Kristin K. Nicodemus

To examine whether postmenopausal women with diabetes experienced a higher incidence of hip fracture than women without diabetes.A prospective cohort of 32,089 postmenopausal women residing in Iowa were surveyed by mail in 1986 and followed for 11 years. Diabetes status and other potential risk factors were assessed by questionnaires at baseline; incidence of hip fracture was ascertained by follow-up questionnaires.A total of 490 hip fractures were reported over 306,900 person-years of follow-up. After adjustment for age, smoking status, estrogen use, BMI, and waist-to-hip ratio, women with type 1 diabetes (n = 47) were 12.25 times (95% CI 5.05-29.73) more likely to report an incident hip fracture than women without diabetes. Women with type 2 diabetes had a 1.70-fold higher risk (1.21-2.38) of incident hip fracture than women without diabetes. Longer duration of type 2 diabetes was associated with higher incidence, as was use of insulin or oral diabetes medications in women with type 2 diabetes. Furthermore, women who were initially free of diabetes but in whom diabetes developed had a relative risk of hip fracture of 1.60 (1.14-2.25) compared with women who never had diabetes.Postmenopausal women who have diabetes or in whom diabetes develops are at higher risk for hip fracture than nondiabetic postmenopausal women. Strategies to prevent osteoporosis and/or falling may be especially warranted in women with diabetes.

Mitochondrial (mt) impairment, particularly within complex I of the electron transport system, has been implicated in the pathogenesis of Parkinson disease (PD). More than half of mitochondrially encoded polypeptides form part of the reduced nicotinamide adenine dinucleotide dehydrogenase (NADH) complex I enzyme. To test the hypothesis that mtDNA variation contributes to PD expression, we genotyped 10 single-nucleotide polymorphisms (SNPs) that define the European mtDNA haplogroups in 609 white patients with PD and 340 unaffected white control subjects. Overall, individuals classified as haplogroup J (odds ratio [OR] 0.55; 95% confidence interval [CI] 0.34-0.91; P=.02) or K (OR 0.52; 95% CI 0.30-0.90; P=.02) demonstrated a significant decrease in risk of PD versus individuals carrying the most common haplogroup, H. Furthermore, a specific SNP that defines these two haplogroups, 10398G, is strongly associated with this protective effect (OR 0.53; 95% CI 0.39-0.73; P=.0001). SNP 10398G causes a nonconservative amino acid change from threonine to alanine within the NADH dehydrogenase 3 (ND3) of complex I. After stratification by sex, this decrease in risk appeared stronger in women than in men (OR 0.43; 95% CI 0.27-0.71; P=.0009). In addition, SNP 9055A of ATP6 demonstrated a protective effect for women (OR 0.45; 95% CI 0.22-0.93; P=.03). Our results suggest that ND3 is an important factor in PD susceptibility among white individuals and could help explain the role of complex I in PD expression.

Bipolar, schizophrenia, and schizoaffective disorders are common, highly heritable psychiatric disorders, for which familial coaggregation, as well as epidemiological and genetic evidence, suggests overlapping etiologies. No definitive susceptibility genes have yet been identified for any of these disorders. Genetic heterogeneity, combined with phenotypic imprecision and poor marker coverage, has contributed to the difficulty in defining risk variants. We focused on families of Ashkenazi Jewish descent, to reduce genetic heterogeneity, and, as a precursor to genomewide association studies, we undertook a single-nucleotide polymorphism (SNP) genotyping screen of 64 candidate genes (440 SNPs) chosen on the basis of previous linkage or of association and/or biological relevance. We genotyped an average of 6.9 SNPs per gene, with an average density of 1 SNP per 11.9 kb in 323 bipolar I disorder and 274 schizophrenia or schizoaffective Ashkenazi case-parent trios. Using single-SNP and haplotype-based transmission/disequilibrium tests, we ranked genes on the basis of strength of association (<i>P</i><.01). Six genes (<i>DAO, GRM3, GRM4, GRIN2B, IL2RB,</i> and <i>TUBA8</i>) met this criterion for bipolar I disorder; only <i>DAO</i> has been previously associated with bipolar disorder. Six genes (<i>RGS4, SCA1, GRM4, DPYSL2, NOS1,</i> and <i>GRID1</i>) met this criterion for schizophrenia or schizoaffective disorder; five replicate previous associations, and one, <i>GRID1,</i> shows a novel association with schizophrenia. In addition, six genes (<i>DPYSL2, DTNBP1, G30/G72, GRID1, GRM4,</i> and <i>NOS1</i>) showed overlapping suggestive evidence of association in both disorders. These results may help to prioritize candidate genes for future study from among the many suspected/proposed for schizophrenia and bipolar disorders. They provide further support for shared genetic susceptibility between these two disorders that involve glutamate-signaling pathways.

Random forests (RF) have been increasingly used in applications such as genome-wide association and microarray studies where predictor correlation is frequently observed. Recent works on permutation-based variable importance measures (VIMs) used in RF have come to apparently contradictory conclusions. We present an extended simulation study to synthesize results.In the case when both predictor correlation was present and predictors were associated with the outcome (HA), the unconditional RF VIM attributed a higher share of importance to correlated predictors, while under the null hypothesis that no predictors are associated with the outcome (H0) the unconditional RF VIM was unbiased. Conditional VIMs showed a decrease in VIM values for correlated predictors versus the unconditional VIMs under HA and was unbiased under H0. Scaled VIMs were clearly biased under HA and H0.Unconditional unscaled VIMs are a computationally tractable choice for large datasets and are unbiased under the null hypothesis. Whether the observed increased VIMs for correlated predictors may be considered a "bias" - because they do not directly reflect the coefficients in the generating model - or if it is a beneficial attribute of these VIMs is dependent on the application. For example, in genetic association studies, where correlation between markers may help to localize the functionally relevant variant, the increased importance of correlated predictors may be an advantage. On the other hand, we show examples where this increased importance may result in spurious signals.

Anxiety disorders are common, complex psychiatric disorders with twin heritabilities of 30-60%. We conducted a genome-wide association study of Lifetime Anxiety Disorder (ncase = 25 453, ncontrol = 58 113) and an additional analysis of Current Anxiety Symptoms (ncase = 19 012, ncontrol = 58 113). The liability scale common variant heritability estimate for Lifetime Anxiety Disorder was 26%, and for Current Anxiety Symptoms was 31%. Five novel genome-wide significant loci were identified including an intergenic region on chromosome 9 that has previously been associated with neuroticism, and a locus overlapping the BDNF receptor gene, NTRK2. Anxiety showed significant positive genetic correlations with depression and insomnia as well as coronary artery disease, mirroring findings from epidemiological studies. We conclude that common genetic variation accounts for a substantive proportion of the genetic architecture underlying anxiety.

Organized neuronal firing is crucial for cortical processing and is disrupted in schizophrenia. Using rapid amplification of 5' complementary DNA ends in human brain, we identified a primate-specific isoform (3.1) of the ether-a-go-go-related K(+) channel KCNH2 that modulates neuronal firing. KCNH2-3.1 messenger RNA levels are comparable to full-length KCNH2 (1A) levels in brain but three orders of magnitude lower in heart. In hippocampus from individuals with schizophrenia, KCNH2-3.1 expression is 2.5-fold greater than KCNH2-1A expression. A meta-analysis of five clinical data sets (367 families, 1,158 unrelated cases and 1,704 controls) shows association of single nucleotide polymorphisms in KCNH2 with schizophrenia. Risk-associated alleles predict lower intelligence quotient scores and speed of cognitive processing, altered memory-linked functional magnetic resonance imaging signals and increased KCNH2-3.1 mRNA levels in postmortem hippocampus. KCNH2-3.1 lacks a domain that is crucial for slow channel deactivation. Overexpression of KCNH2-3.1 in primary cortical neurons induces a rapidly deactivating K(+) current and a high-frequency, nonadapting firing pattern. These results identify a previously undescribed KCNH2 channel isoform involved in cortical physiology, cognition and psychosis, providing a potential new therapeutic drug target.

To identify genes influencing age at onset (AAO) in two common neurodegenerative diseases, a genomic screen was performed for AAO in families with Alzheimer disease (AD; <i>n</i>=449) and Parkinson disease (PD; <i>n</i>=174). Heritabilities between 40%–60% were found in both the AD and PD data sets. For PD, significant evidence for linkage to AAO was found on chromosome 1p (LOD = 3.41). For AD, the AAO effect of APOE (LOD = 3.28) was confirmed. In addition, evidence for AAO linkage on chromosomes 6 and 10 was identified independently in both the AD and PD data sets. Subsequent unified analyses of these regions identified a single peak on chromosome 10q between D10S1239 and D10S1237, with a maximum LOD score of 2.62. These data suggest that a common gene affects AAO in these two common complex neurodegenerative diseases.

We examined the association of mtDNA variation with Alzheimer disease (AD) risk in Caucasians (989 cases and 328 controls) testing the effect of individual haplogroups and single nucleotide polymorphisms (SNPs). Logistic regression analyses were used to assess risk of haplogroups and SNPs with AD in both main effects and interaction models. Males classified as haplogroup U showed an increase in risk (OR = 2.30; 95% CI, 1.03-5.11; P = 0.04) of AD relative to the most common haplogroup H, while females demonstrated a significant decrease in risk with haplogroup U (OR = 0.44 ; 95% CI, 0.24-0.80; P = 0.007). Our results were independent of APOE genotype, demonstrating that the effect of mt variation is not confounded by APOE4 carrier status. We suggest that variations within haplogroup U may be involved in AD expression in combination with environmental exposures or nuclear proteins other than APOE.

In mammals, the neuropeptide vasopressin is a key molecule for complex emotional and social behaviours. Two microsatellite polymorphisms, RS1 and RS3, near the promoter of AVPR1A, encoding the receptor subtype most heavily implicated in behaviour regulation, have been linked to autism and behavioural traits. However, the impact of these variants on human brain function is unknown. Here we show that human amygdala function is strongly associated with genetic variation in AVPR1A. Using an imaging genetics approach in a sample of 121 volunteers studied with an emotional face-matching paradigm, we found that differential activation of amygdala is observed in carriers of risk alleles for RS3 and RS1. Alleles in RS1 previously reported to be significantly over- and undertransmitted to autistic probands showed opposing effects on amygdala activation. Furthermore, we show functional difference in human brain between short and long repeat lengths that mirror findings recently obtained in a corresponding variant in voles. Our results indicate a neural mechanism mediating genetic risk for autism through an impact on amygdala signalling and provide a rationale for exploring therapeutic strategies aimed at abnormal amygdala function in this disorder.

The etiology of schizophrenia is thought to include both epistasis and gene-environment interactions. We sought to test whether a set of schizophrenia candidate genes regulated by hypoxia or involved in vascular function in the brain (AKT1, BDNF, CAPON, CHRNA7, COMT, DTNBP1, GAD1, GRM3, NOTCH4, NRG1, PRODH, RGS4, TNF-alpha) interacted with serious obstetric complications to influence risk for schizophrenia. A family-based study of transmission disequilibrium was conducted in 116 trios. Twenty-nine probands had at least one serious obstetric complication (OC) using the McNeil-Sjostrom Scale, and many of the OCs reported were associated with the potential for fetal hypoxia. Analyses were conducted using conditional logistic regression and a likelihood ratio test (LRT) between nested models was performed to assess significance. Of the 13 genes examined, four (AKT1 (three SNPs), BDNF (two SNPs), DTNBP1 (one SNP) and GRM3 (one SNP)) showed significant evidence for gene-by-environment interaction (LRT P-values ranged from 0.011 to 0.037). Although our sample size was modest and the power to detect interactions was limited, we report significant evidence for genes involved in neurovascular function or regulated by hypoxia interacting with the presence of serious obstetric complications to increase risk for schizophrenia.

A recent study examined the stability of rankings from random forests using two variable importance measures (mean decrease accuracy (MDA) and mean decrease Gini (MDG)) and concluded that rankings based on the MDG were more robust than MDA. However, studies examining data-specific characteristics on ranking stability have been few. Rankings based on the MDG measure showed sensitivity to within-predictor correlation and differences in category frequencies, even when the number of categories was held constant, and thus may produce spurious results. The MDA measure was robust to these data characteristics. Further, under strong within-predictor correlation, MDG rankings were less stable than those using MDA.

The advent of high-throughput genomics has produced studies with large numbers of predictors (e.g. genome-wide association, microarray studies). Machine learning algorithms (MLAs) are a computationally efficient way to identify phenotype-associated variables in high-dimensional data. There are important results from mathematical theory and numerous practical results documenting their value. One attractive feature of MLAs is that many operate in a fully multivariate environment, allowing for small-importance variables to be included when they act cooperatively. However, certain properties of MLAs under conditions common in genomic-related data have not been well-studied--in particular, correlations among predictors pose a problem.Using extensive simulation, we showed considering correlation within predictors is crucial in making valid inferences using variable importance measures (VIMs) from three MLAs: random forest (RF), conditional inference forest (CIF) and Monte Carlo logic regression (MCLR). Using a case-control illustration, we showed that the RF VIMs--even permutation-based--were less able to detect association than other algorithms at effect sizes encountered in complex disease studies. This reduction occurred when 'causal' predictors were correlated with other predictors, and was sharpest when RF tree building used the Gini index. Indeed, RF Gini VIMs are biased under correlation, dependent on predictor correlation strength/number and over-trained to random fluctuations in data when tree terminal node size was small. Permutation-based VIM distributions were less variable for correlated predictors and are unbiased, thus may be preferred when predictors are correlated. MLAs are a powerful tool for high-dimensional data analysis, but well-considered use of algorithms is necessary to draw valid conclusions.Supplementary data are available at Bioinformatics online.

There is a great deal of interest in a fine-scale population structure in the UK, both as a signature of historical immigration events and because of the effect population structure may have on disease association studies. Although population structure appears to have a minor impact on the current generation of genome-wide association studies, it is likely to have a significant part in the next generation of studies designed to search for rare variants. A powerful way of detecting such structure is to control and document carefully the provenance of the samples involved. In this study, we describe the collection of a cohort of rural UK samples (The People of the British Isles), aimed at providing a well-characterised UK-control population that can be used as a resource by the research community, as well as providing a fine-scale genetic information on the British population. So far, some 4000 samples have been collected, the majority of which fit the criteria of coming from a rural area and having all four grandparents from approximately the same area. Analysis of the first 3865 samples that have been geocoded indicates that 75% have a mean distance between grandparental places of birth of 37.3 km, and that about 70% of grandparental places of birth can be classed as rural. Preliminary genotyping of 1057 samples demonstrates the value of these samples for investigating a fine-scale population structure within the UK, and shows how this can be enhanced by the use of surnames.

AKT1-dependent molecular pathways control diverse aspects of cellular development and adaptation, including interactions with neuronal dopaminergic signaling. If AKT1 has an impact on dopaminergic signaling, then genetic variation in AKT1 would be associated with brain phenotypes related to cortical dopaminergic function. Here, we provide evidence that a coding variation in AKT1 that affects protein expression in human B lymphoblasts influenced several brain measures related to dopaminergic function. Cognitive performance linked to frontostriatal circuitry, prefrontal physiology during executive function, and frontostriatal gray-matter volume on MRI were altered in subjects with the AKT1 variation. Moreover, on neuroimaging measures with a main effect of the AKT1 genotype, there was significant epistasis with a functional polymorphism (Val158Met) in catechol-O-methyltransferase [COMT], a gene that indexes cortical synaptic dopamine. This genetic interaction was consistent with the putative role of AKT1 in dopaminergic signaling. Supportive of an earlier tentative association of AKT1 with schizophrenia, we also found that this AKT1 variant was associated with risk for schizophrenia. These data implicate AKT1 in modulating human prefrontal-striatal structure and function and suggest that the mechanism of this effect may be coupled to dopaminergic signaling and relevant to the expression of psychosis.

NRG1 is a schizophrenia candidate gene and plays an important role in brain development and neural function. Schizophrenia is a complex disorder, with etiology likely due to epistasis.To examine epistasis between NRG1 and selected N-methyl-d-aspartate-glutamate pathway partners implicated in its effects, including ERBB4, AKT1, DLG4, NOS1, and NOS1AP.Schizophrenia case-control sample analyzed using machine learning algorithms and logistic regression with follow-up using neuroimaging on an independent sample of healthy controls.A referred sample of schizophrenic patients (n = 296) meeting DSM-IV criteria for schizophrenia spectrum disorder and a volunteer sample of controls for case-control comparison (n = 365) and a separate volunteer sample of controls for neuroimaging (n = 172).Epistatic association between single-nucleotide polymorphisms (SNPs) and case-control status; epistatic association between SNPs and the blood oxygen level-dependent physiological response during working memory measured by functional magnetic resonance imaging.We observed interaction between NRG1 5' and 3' SNPs rs4560751 and rs3802160 (likelihood ratio test P = .00020) and schizophrenia, which was validated using functional magnetic resonance imaging of working memory in healthy controls; carriers of risk-associated genotypes showed inefficient processing in the dorsolateral prefrontal cortex (P = .015, familywise error corrected). We observed epistasis between NRG1 (rs10503929; Thr286/289/294Met) and its receptor ERBB4 (rs1026882; likelihood ratio test P = .035); a 3-way interaction with these 2 SNPs and AKT1 (rs2494734) was also observed (odds ratio, 27.13; 95% confidence interval, 3.30-223.03; likelihood ratio test P = .042). These same 2- and 3-way interactions were further biologically validated via functional magnetic resonance imaging: healthy individuals carrying risk genotypes for NRG1 and ERBB4, or these 2 together with AKT1, were disproportionately less efficient in dorsolateral prefrontal cortex processing. Lower-level interactions were not observed between NRG1 /ERBB4 and AKT1 in association or neuroimaging, consistent with biological evidence that NRG1 × ERBB4 interaction modulates downstream AKT1 signaling.Our data suggest complex epistatic effects implicating an NRG1 molecular pathway in cognitive brain function and the pathogenesis of schizophrenia.

<h2>Abstract</h2><h3>Objectives</h3> Researchers need to be confident about the reliability of epidemiologic studies that quantify medication use through self-report. Some evidence suggests that psychiatric medications are systemically under-reported. Modern record linkage enables validation of self-report with national prescribing data as gold standard. Here, we investigated the validity of medication self-report for multiple medication types. <h3>Study Design and Setting</h3> Participants in the Generation Scotland population-based cohort (<i>N</i> = 10,244) recruited 2009–2011 self-reported regular usage of several commonly prescribed medication classes. This was matched against Scottish NHS prescriptions data using 3- and 6-month fixed time windows. Potential predictors of discordant self-report, including general intelligence and psychological distress, were studied via multivariable logistic regression. <h3>Results</h3> Antidepressants self-report showed very good agreement (κ = 0.85, [95% confidence interval (CI) 0.84–0.87]), comparable to antihypertensives (κ = 0.90 [CI 0.89–0.91]). Self-report of mood stabilizers showed moderate-poor agreement (κ = 0.42 [CI 0.33–0.50]). Relevant past medical history was the strongest predictor of self-report sensitivity, whereas general intelligence was not predictive. <h3>Conclusion</h3> In this large population-based study, we found self-report validity varied among medication classes, with no simple relationship between psychiatric medication and under-reporting. History of indicated illness predicted more accurate self-report, for both psychiatric and nonpsychiatric medications. Although other patient-level factors influenced self-report for some medications, none predicted greater accuracy across all medications studied.

Kidney cancer incidence rates in the United States have been increasing and are not fully accounted for by better diagnostic techniques. Risk factors in women are incompletely described. A total of 34,637 Iowan women initially free of cancer completed a mailed questionnaire in 1986. Kidney cancer incidence was identified over 15 years of follow-up (n = 124) through linkage to a statewide cancer registry. Adjusted for age and other risk factors, kidney cancer was associated positively with maximum adult weight (p for trend = 0.02) and current waist-to-hip ratio (p for trend = 0.002). Compared to nondrinkers, consumers of alcohol of 3.0 or more grams per day had a relative risk (RR) of 0.52 (95% CI = 0.29-0.92). Total vitamin C intake was associated positively with risk of kidney cancer (p for trend = 0.04), whereas total vitamin E intake was associated negatively with risk (p for trend = 0.002). The few women who used copper supplements had a 6.52-fold increase in risk of kidney cancer (95% CI = 1.95-21.8). Compared to never users, women who were former users of estrogen had an increased risk of kidney cancer (RR = 1.62; 95% CI = 1.11-2.36), but current users of estrogen were not at a higher risk. Women who were nulliparous or had more than 2 live births were also at increased risk of kidney cancer compared with women who had 1 or 2 live births. In conclusion, in these postmenopausal women, overweight, particularly central adiposity, was an important risk factor for kidney cancer. Potential risk factors that warrant further exploration were low intake of alcohol and vitamin E, higher intake of vitamin C, nulli- or multiparity and use of copper supplements.

It is hypothesized that diets deficient in folate, methionine, and vitamins B-6 and B-12 cause DNA hypomethylation and, as a result, increase risk of colorectal cancers. Furthermore, it is proposed that alcohol, a methyl group antagonist, increases risk of colorectal cancers among those with low intake of folate. Data from the Iowa Women's Health Study, a population-based cohort of incident cancer, were used to examine the relationship of folate, methionine, and vitamins B-6 and B-12 to occurrence of cancers of the colon (n = 598) and rectum (n = 123) over 13 yr of follow-up. There were no independent associations of folate, methionine, or vitamins B-6 and B-12 derived from a food frequency questionnaire with incidence of colon cancer. Adjusted relative risks (RRs) of rectal cancer were similar across categories of folate, vitamin B-12, and methionine intake, but RRs increased progressively with increasing intake of vitamin B-6 [P (for trend) = 0.03]. RRs suggested that incidence of cancer of the proximal colon was lower among those with 1) high folate and high vitamin B-12 intake [RR = 0.59, 95% confidence interval (CI) = 0.39-0.89] and 2) high folate and high vitamin B-6 intake (RR = 0.65, 95% CI = 0.50-0.84) than among those with the lowest intake of these nutrients. Incidence of cancer of the proximal colon was also somewhat lower among those with high folate and low alcohol intake (RR = 0.44, 95% CI = 0.22-0.89). Findings provide limited support for an association between dietary factors involved in DNA methylation and risk of cancers of the colon and rectum.

Abstract Objective: Controversy remains regarding the association between type 2 diabetes mellitus (DM) and colorectal cancer (CRC) risk. To clarify and extend the existing data, we prospectively evaluated the association between self-reported type 2 DM (onset at &amp;gt;30 years of age) and incident CRC, overall and by anatomic subsite, among postmenopausal women in the Iowa Women's Health Study (n = 35,230). Methods: After 14 years of follow-up, a total of 870 incident CRC cases were identified through annual linkage to the Iowa Cancer Registry. DM was analyzed as reported at baseline and as a time-dependent variable using information obtained during follow-up. CRC risks were estimated using Cox proportional hazards regression models. Results: After adjusting for age, body mass index and other potential confounding variables, the relative risk (RR) for women with DM versus women without DM was modestly increased at 1.4 [95% confidence interval (95% CI), 1.1-1.8]. By anatomic subsite, the RR for proximal colon cancer was statistically significantly increased (RR, 1.9; 95% CI, 1.3-2.6), whereas the RRs for distal colon (RR, 1.1; 95% CI, 0.6-1.8) and rectal cancer (RR, 0.8; 95% CI, 0.4-1.6) were not statistically different from unity. Analyses that included DM ascertained at baseline and follow-up yielded similar results. Conclusion: In this large, prospective study of postmenopausal women, the association between DM and incident CRC was found to be subsite specific. If confirmed by others, this finding implies that CRC prevention strategies among type 2 DM patients should include examination of the proximal colon.

Although scientific knowledge regarding the influence of nutritional factors on health and disease serves as the basis for specific recommendations included in the Dietary Guidelines for Americans, limited empirical epidemiologic data are available to verify that adherence to the cluster of nutrition-related behaviors included in the Dietary Guidelines will reduce the incidence of disease.We examined the association of compliance with the Dietary Guidelines and incident cancers.Data from a population-based cohort of postmenopausal women (n = 34 708) were examined. A dietary guidelines index was derived as a summary measure of compliance with the Dietary Guidelines, and the association of this index and cancer incidence was examined for all cancers combined and for site-specific cancers with > 100 events.For all cancers combined, the relative risks associated with the upper 4 quintiles of the dietary guidelines index in reference to the bottom quintile were 0.95 (95% CI: 0.87, 1.05) for quintile 2, 0.88 (95% CI: 0.80, 0.97) for quintile 3, 0.88 (95% CI: 0.80, 0.96) for quintile 4, and 0.85 (95% CI: 0.77, 0.93) for quintile 5 (P for trend < 0.01). Similar patterns in relative risks were found for cancers of the colon, bronchus and lung, breast, and uterus. In contrast, ovarian cancer incidence was positively associated with the dietary guidelines index.Our findings suggest that adherence to the cluster of nutrition-related behaviors included in the Dietary Guidelines for Americans may be associated with a lower risk of cancer.

We examined the association between whole-grain intake and incident upper aerodigestive tract cancer in a cohort of 34,651 postmenopausal, initially cancer-free women. We also studied established risk factors for upper aerodigestive cancers, including fruit and vegetable intake, smoking and alcohol intake. A mailed questionnaire at baseline in 1986 included a food-frequency questionnaire and assessment of other cancer risk factors. During the 14-year follow-up period, 169 women developed cancer of the upper aerodigestive tract. For all upper aerodigestive cancers together, significant inverse associations were observed for the highest compared to the lowest tertile of whole grains [relative risk (RR) = 0.53, 95% confidence interval (CI) 0.34-0.81] and yellow/orange vegetables (RR = 0.58, 95% CI 0.39-0.87). In addition, those in the highest compared to lowest tertile of fiber intake from whole grain were less likely to develop upper aerodigestive tract cancer (RR = 0.56, 95% CI 0.37-0.84); fiber intake from refined grain was not significantly associated with upper aerodigestive tract cancer. Findings were generally similar for oropharyngeal (n = 53), laryngeal (n = 21), nasopharyngeal/salivary (n = 18), esophageal (n = 21) and gastric (n = 56) cancers, though numbers of cases were too small for statistical testing within individual cancers. These findings confirm previous observations that high intake of fruits and vegetables and that intake of whole grains and the fiber derived from them may reduce risk of upper aerodigestive tract cancers.

Imaging genetics provides an enormous amount of functional-structural data on gene effects in living brain, but the sheer quantity of potential phenotypes raises concerns about false discovery. Here, we provide the first empirical results on false positive rates in imaging genetics. We analyzed 720 frequent coding SNPs without significant association with schizophrenia and a subset of 492 of these without association with cognitive function. Effects on brain structure (using voxel-based morphometry, VBM) and brain function, using two archival imaging tasks, the n-back working memory task and an emotional face matching task, were studied in whole brain and regions of interest and corrected for multiple comparisons using standard neuroimaging procedures. Since these variants are unlikely to impact relevant brain function, positives obtained provide an upper empirical estimate of the false positive association rate. In a separate analysis, we randomly permuted genotype labels across subjects, removing any true genotype-phenotype association in the data, to derive a lower empirical estimate. At a set correction level of 0.05, in each region of interest and data set used, the rate of positive findings was well below 5% (0.2-4.1%). There was no relationship between the region of interest and the false positive rate. Permutation results were in the same range as empirically derived rates. The observed low rates of positives provide empirical evidence that the type I error rate is well controlled by current commonly used correction procedures in imaging genetics, at least in the context of the imaging paradigms we have used. In fact, our observations indicate that these statistical thresholds are conservative.

PRODH, encoding proline oxidase (POX), has been associated with schizophrenia through linkage, association, and the 22q11 deletion syndrome (Velo-Cardio-Facial syndrome). Here, we show in a family-based sample that functional polymorphisms in PRODH are associated with schizophrenia, with protective and risk alleles having opposite effects on POX activity. Using a multimodal imaging genetics approach, we demonstrate that haplotypes constructed from these risk and protective functional polymorphisms have dissociable correlations with structure, function, and connectivity of striatum and prefrontal cortex, impacting critical circuitry implicated in the pathophysiology of schizophrenia. Specifically, the schizophrenia risk haplotype was associated with decreased striatal volume and increased striatal-frontal functional connectivity, while the protective haplotype was associated with decreased striatal-frontal functional connectivity. Our findings suggest a role for functional genetic variation in POX on neostriatal-frontal circuits mediating risk and protection for schizophrenia.

Category fluency is a widely used task that relies on multiple neurocognitive processes and is a sensitive assay of cortical dysfunction, including in schizophrenia. The test requires naming of as many words belonging to a certain category (e.g., animals) as possible within a short period of time. The core metrics are the overall number of words produced and the number of errors, namely non-members generated for a target category. We combine a computational linguistic approach with a candidate gene approach to examine the genetic architecture of this traditional fluency measure. In addition to the standard metric of overall word count, we applied a computational approach to semantics, Latent Semantic Analysis (LSA), to analyse the clustering pattern of the categories generated, as it likely reflects the search in memory for meanings. Also, since fluency performance probably also recruits verbal learning and recall processes, we included two standard measures of this cognitive process: the Wechsler Memory Scale and California Verbal Learning Test (CVLT). To explore the genetic architecture of traditional and LSA-derived fluency measures we employed a candidate gene approach focused on SNPs with known function that were available from a recent genome-wide association study (GWAS) of schizophrenia. The selected candidate genes were associated with language and speech, verbal learning and recall processes, and processing speed. A total of 39 coding SNPs were included for analysis in 665 subjects. Given the modest sample size, the results should be regarded as exploratory and preliminary. Nevertheless, the data clearly illustrate how extracting the meaning from participants' responses, by analysing the actual content of words, generates useful and neurocognitively viable metrics. We discuss three replicated SNPs in the genes ZNF804A, DISC1 and KIAA0319, as well as the potential for computational analyses of linguistic and textual data in other genomics tasks.

Data science uses computer science and statistics to extract new knowledge from high-dimensional datasets (ie, those with many different variables and data types). Mental health research, diagnosis, and treatment could benefit from data science that uses cohort studies, genomics, and routine health-care and administrative data. The UK is well placed to trial these approaches through robust NHS-linked data science projects, such as the UK Biobank, Generation Scotland, and the Clinical Record Interactive Search (CRIS) programme. Data science has great potential as a low-cost, high-return catalyst for improved mental health recognition, understanding, support, and outcomes. Lessons learnt from such studies could have global implications.

Antidepressants demonstrate modest response rates in the treatment of major depressive disorder (MDD). Despite previous genome-wide association studies (GWAS) of antidepressant treatment response, the underlying genetic factors are unknown. Using prescription data in a population and family-based cohort (Generation Scotland: Scottish Family Health Study; GS:SFHS), we sought to define a measure of (a) antidepressant treatment resistance and (b) stages of antidepressant resistance by inferring antidepressant switching as non-response to treatment. GWAS were conducted separately for antidepressant treatment resistance in GS:SFHS and the Genome-based Therapeutic Drugs for Depression (GENDEP) study and then meta-analysed (meta-analysis n = 4213, cases = 358). For stages of antidepressant resistance, a GWAS on GS:SFHS only was performed (n = 3452). Additionally, we conducted gene-set enrichment, polygenic risk scoring (PRS) and genetic correlation analysis. We did not identify any significant loci, genes or gene sets associated with antidepressant treatment resistance or stages of resistance. Significant positive genetic correlations of antidepressant treatment resistance and stages of resistance with neuroticism, psychological distress, schizotypy and mood disorder traits were identified. These findings suggest that larger sample sizes are needed to identify the genetic architecture of antidepressant treatment response, and that population-based observational studies may provide a tractable approach to achieving the necessary statistical power.

A recent development in the genetic architecture of schizophrenia suggested that an omnigenic model may underlie the risk for this disorder. The aim of our study was to use polygenic profile scoring to quantitatively assess whether a number of experimentally derived sets would contribute to the disorder above and beyond the omnigenic effect. Using the PGC2 secondary analysis schizophrenia case-control cohort (N = 29,125 cases and 34,836 controls), a robust polygenic signal was observed from gene sets based on TCF4, FMR1, upregulation from MIR137 and downregulation from CHD8. Additional analyses revealed a constant floor effect in the amount of variance explained, consistent with the omnigenic model. Thus, we report that putative core gene sets showed a significant effect above and beyond the floor effect that might be linked with the underlying omnigenic background. In addition, we demonstrate a method to quantify the contribution of specific gene sets within the omnigenic context.

Further evidence for association between ErbB4 and schizophrenia and influence on cognitive intermediate phenotypes in healthy controls

Abstract Summary: snp.plotter is a newly developed R package which produces high-quality plots of results from genetic association studies. The main features of the package include options to display a linkage disequilibrium (LD) plot below the P-value plot using either the r2 or D′ LD metric, to set the X-axis to equal spacing or to use the physical map of markers, and to specify plot labels, colors, symbols and LD heatmap color scheme. snp.plotter can plot single SNP and/or haplotype data and simultaneously plot multiple sets of results. R is a free software environment for statistical computing and graphics available for most platforms. The proposed package provides a simple way to convey both association and LD information in a single appealing graphic for genetic association studies. Availability: Downloadable R package and example datasets are available at http://cbdb.nimh.nih.gov/~kristin/snp.plotter.html and http://www.r-project.org Contact: nicodemusk@mail.nih.gov

The G protein-coupled receptor (GPCR) family is highly diversified and involved in many forms of information processing. SREB2 (GPR85) is the most conserved GPCR throughout vertebrate evolution and is expressed abundantly in brain structures exhibiting high levels of plasticity, e.g., the hippocampal dentate gyrus. Here, we show that SREB2 is involved in determining brain size, modulating diverse behaviors, and potentially in vulnerability to schizophrenia. Mild overexpression of SREB2 caused significant brain weight reduction and ventricular enlargement in transgenic (Tg) mice as well as behavioral abnormalities mirroring psychiatric disorders, e.g., decreased social interaction, abnormal sensorimotor gating, and impaired memory. SREB2 KO mice showed a reciprocal phenotype, a significant increase in brain weight accompanying a trend toward enhanced memory without apparent other behavioral abnormalities. In both Tg and KO mice, no gross malformation of brain structures was observed. Because of phenotypic overlap between SREB2 Tg mice and schizophrenia, we sought a possible link between the two. Minor alleles of two SREB2 SNPs, located in intron 2 and in the 3' UTR, were overtransmitted to schizophrenia patients in a family-based sample and showed an allele load association with reduced hippocampal gray matter volume in patients. Our data implicate SREB2 as a potential risk factor for psychiatric disorders and its pathway as a target for psychiatric therapy.

Major depressive disorder (MDD) is a heritable and highly debilitating condition. It is commonly associated with subcortical volumetric abnormalities, the most replicated of these being reduced hippocampal volume. Using the most recent published data from Enhancing Neuroimaging Genetics through Meta-analysis (ENIGMA) consortium's genome-wide association study of regional brain volume, we sought to test whether there is shared genetic architecture between seven subcortical brain volumes and intracranial volume (ICV) and MDD. We explored this using linkage disequilibrium score regression, polygenic risk scoring (PRS) techniques, Mendelian randomisation (MR) analysis and BUHMBOX. Utilising summary statistics from ENIGMA and Psychiatric Genomics Consortium, we demonstrated that hippocampal volume was positively genetically correlated with MDD (rG=0.46, P=0.02), although this did not survive multiple comparison testing. None of the other six brain regions studied were genetically correlated and amygdala volume heritability was too low for analysis. Using PRS analysis, no regional volumetric PRS demonstrated a significant association with MDD or recurrent MDD. MR analysis in hippocampal volume and MDD identified no causal association, however, BUHMBOX analysis identified genetic subgrouping in GS:SFHS MDD cases only (P=0.00281). In this study, we provide some evidence that hippocampal volume and MDD may share genetic architecture in a subgroup of individuals, albeit the genetic correlation did not survive multiple testing correction and genetic subgroup heterogeneity was not replicated. In contrast, we found no evidence to support a shared genetic architecture between MDD and other regional subcortical volumes or ICV.

Although permutation testing has been the gold standard for assessing significance levels in studies using multiple markers, it is time-consuming. A Bonferroni correction to the nominal p-value that uses the underlying pair-wise linkage disequilibrium (LD) structure among the markers to determine the number of effectively independent tests has recently been proposed. We propose using the number of independent LD blocks plus the number of independent single-nucleotide polymorphisms for correction. Using the Collaborative Study on the Genetics of Alcoholism LD data for chromosome 21, we simulated 1,000 replicates of parent-child trio data under the null hypothesis with two levels of LD: moderate and high. Assuming haplotype blocks were independent, we calculated the number of independent statistical tests using 3 haplotype blocking algorithms. We then compared the type I error rates using a principal components-based method, the three blocking methods, a traditional Bonferroni correction, and the unadjusted p-values obtained from FBAT. Under high LD conditions, the PC method and one of the blocking methods were slightly conservative, whereas the 2 other blocking methods exceeded the target type I error rate. Under conditions of moderate LD, we show that the blocking algorithm corrections are closest to the desired type I error, although still slightly conservative, with the principal components-based method being almost as conservative as the traditional Bonferroni correction.

Risk for complex disease is thought to be controlled by multiple genetic risk factors, each with small individual effects. Meta-analyses of several independent studies may be helpful to increase the ability to detect association when effect sizes are modest. Although many software options are available for meta-analysis of genetic case-control data, no currently available software implements the method described by Kazeem and Farrall (2005), which combines data from independent family-based and case-control studies.I introduce the package catmap for the R statistical computing environment that implements fixed- and random-effects pooled estimates for case-control and transmission disequilibrium methods, allowing for the use of genetic association data across study types. In addition, catmap may be used to create forest and funnel plots and to perform sensitivity analysis and cumulative meta-analysis. catmap is available from the Comprehensive R Archive Network http://www.r-project.org.catmap allows researchers to synthesize data to assess evidence for association in studies of genetic polymorphisms, facilitating the use of pooled data analyses which may increase power to detect moderate genetic associations.

Abnormalities in social interaction are a common feature of several psychiatric disorders, aligning with the recent move towards using Research Domain Criteria (RDoC) to describe disorders in terms of observable behaviours rather than using specific diagnoses. Neuroeconomic games are an effective measure of social decision-making that can be adapted for use in neuroimaging, allowing investigation of the biological basis for behaviour. This review summarises findings of neuroeconomic gameplay studies in Axis 1 psychiatric disorders and advocates the use of these games as measures of the RDoC Affiliation and Attachment, Reward Responsiveness, Reward Learning and Reward Valuation constructs. Although research on neuroeconomic gameplay is in its infancy, consistencies have been observed across disorders, particularly in terms of impaired integration of social and cognitive information, avoidance of negative social interactions and reduced reward sensitivity, as well as a reduction in activity in brain regions associated with processing and responding to social information.

Abstract We examined whether there is an association between whole grain intake and incident endometrial cancer and whether the association varied by use of hormone replacement therapy. The study included 23,014 Iowa women, aged 55-69 years in 1986. A mailed food frequency questionnaire was used to estimate grain intake, hormone replacement therapy use, and other cancer risk factors. Cancer incidence from 1986 to 1998 was also collected. In analyses stratified by hormone replacement therapy use, an inverse association between whole grain intake and endometrial cancer was observed among never-users of hormone replacement therapy (p for trend = 0.05). Never-users in the highest quintile of whole grain intake were 0.63 times as likely to develop endometrial cancer as those in the lowest quintile of whole grain intake (95% confidence interval = 0.39-1.01). Among hormone replacement therapy users, no association between whole grain intake and endometrial cancer was evident. There was no statistically significant association between whole grain intake and incident endometrial cancer when users of hormone replacement therapy and nonusers were analyzed together. There also was no association between refined grain intake and endometrial cancer. Whole grain intake may protect against endometrial cancer among never-users of hormone replacement therapy.

No AccessJournal of UrologyReview article1 May 2006A Systematic Review of Vitamin D Receptor Gene Polymorphisms and Prostate Cancer Risk Sonja I. Berndt, Jennifer L. Dodson, Wen-Yi Huang, and Kristin K. Nicodemus Sonja I. BerndtSonja I. Berndt Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health Baltimore and Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland More articles by this author , Jennifer L. DodsonJennifer L. Dodson James Buchanan Brady Urological Institute, Johns Hopkins Hospital More articles by this author , Wen-Yi HuangWen-Yi Huang Baltimore and Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland More articles by this author , and Kristin K. NicodemusKristin K. Nicodemus Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health More articles by this author View All Author Informationhttps://doi.org/10.1016/S0022-5347(05)00958-4AboutFull TextPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract Purpose: Polymorphisms in the vitamin D receptor gene have been hypothesized to alter the risk of prostate cancer. However, studies investigating the associations between specific vitamin D receptor polymorphisms and prostate cancer risk have yielded inconsistent results. Materials and Methods: We performed a meta-analysis of 26 studies evaluating the association between vitamin D receptor TaqI, poly(A), BsmI, ApaI, and/or FokI polymorphisms, and prostate cancer risk. Results: The studies were heterogeneous in terms of study design, selection of cases and controls, and racial composition. Random effects models were used to estimate the pooled OR and 95% CI of each vitamin D receptor polymorphism under codominant, additive, dominant and recessive genetic models. Overall we did not find evidence to support an association between any of the vitamin D receptor polymorphisms and the risk of prostate cancer. For TaqI, which is the most studied vitamin D receptor polymorphism with 18 studies (total of 2,727 cases and 3,685 controls), the pooled OR was 1.00 (95% CI 0.85 to 1.18) for the Tt vs TT genotypes, 0.94 (95% CI 0.78 to 1.13) for the tt vs TT genotypes and 0.89 (95% CI 0.71 to 1.10) for the recessive model (tt vs Tt plus TT). ORs for the poly(A) microsatellite, BsmI, ApaI and FokI polymorphisms were similar. Conclusions: The results of this meta-analysis suggest that the vitamin D receptor TaqI, poly(A), BsmI, ApaI and FokI polymorphisms are not related to prostate cancer risk. References 1 : Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res1990; 10: 1307. Google Scholar 2 : Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res1994; 54: 805. Medline, Google Scholar 3 : Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1alpha,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res1995; 1: 997. Medline, Google Scholar 4 : Geographic patterns of prostate cancer mortality. Evidence for a protective effect of ultraviolet radiation. Cancer1992; 70: 2861. Crossref, Medline, Google Scholar 5 : Vitamin D receptor expression is required for growth modulation by 1 alpha,25-dihydroxyvitamin D3 in the human prostatic carcinoma cell line ALVA-31. J Steroid Biochem Mol Biol1996; 58: 277. Google Scholar 6 : The human prostatic carcinoma cell line LNCaP expresses biologically active, specific receptors for 1 alpha,25-dihydroxyvitamin D3. Cancer Res1992; 52: 515. Google Scholar 7 : Risk of early-onset prostate cancer in relation to germ line polymorphisms of the vitamin D receptor. Cancer Epidemiol Biomarkers Prev2004; 13: 1325. Google Scholar 8 : Prediction of bone density from vitamin D receptor alleles. Nature1994; 367: 284. Google Scholar 9 : Comparative high-resolution analysis of linkage disequilibrium and tag single nucleotide polymorphisms between populations in the vitamin D receptor gene. Hum Mol Genet2004; 13: 1633. Google Scholar 10 : A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res1997; 12: 915. Google Scholar 11 : Association of prostate cancer with vitamin D receptor gene polymorphism. Cancer Res1996; 56: 4108. Medline, Google Scholar 12 : Association of prostate cancer risk with genetic polymorphisms in vitamin D receptor and androgen receptor. J Natl Cancer Inst1997; 89: 166. Crossref, Medline, Google Scholar 13 : Vitamin D receptor polymorphisms and prostate cancer. Mol Carcinog2000; 27: 18. Google Scholar 14 : Polymorphisms in the vitamin D receptor gene, ultraviolet radiation, and susceptibility to prostate cancer. Environ Mol Mutagen2004; 43: 121. Google Scholar 15 : Vitamin D receptor gene polymorphisms and prostate cancer risk. Prostate2004; 59: 409. Google Scholar 16 : Vitamin D receptor polymorphisms as markers in prostate cancer. Hum Genet1999; 105: 281. Google Scholar 17 : Vitamin D receptor gene polymorphism in Japanese patients with prostate cancer. Endocr J1999; 46: 467. Google Scholar 18 : Vitamin D receptor gene polymorphism and prostate cancer risk. Prostate2002; 51: 30. Google Scholar 19 : Association of vitamin D receptor gene polymorphism with prostate cancer and benign prostatic hyperplasia in a Japanese population. Cancer Res2000; 60: 305. Google Scholar 20 : Clinical and pathological significance of vitamin D receptor gene polymorphism for prostate cancer which is associated with a higher mortality in Japanese. Endocr J2001; 48: 543. Google Scholar 21 : Association between vitamin D receptor polymorphisms and prostate cancer risk in a Taiwanese population. Cancer Lett2004; 207: 69. Google Scholar 22 : Association of prostate cancer with vitamin D receptor haplotypes in African-Americans. Cancer Res1998; 58: 1620. Google Scholar 23 : Vitamin D receptor polymorphisms and lethal prostate cancer. J Urol1998; 160: 1405. Link, Google Scholar 24 : Vitamin D receptor polymorphisms, circulating vitamin D metabolites, and risk of prostate cancer in United States physicians. Cancer Epidemiol Biomarkers Prev1998; 7: 385. Google Scholar 25 : Vitamin D receptor polymorphisms and prostate cancer risk in Brazilian men. Int J Biol Markers2004; 19: 245. Google Scholar 26 : The role of vitamin D receptor gene polymorphisms in the susceptibility to prostate cancer of a southern European population. J Hum Genet2002; 47: 413. Google Scholar 27 : Vitamin D receptor gene polymorphism in familial prostate cancer in a Japanese population. Int J Urol2003; 10: 261. Google Scholar 28 : CYP3A4 and VDR gene polymorphisms and the risk of prostate cancer in men with benign prostate hyperplasia. Br J Cancer2003; 88: 928. Google Scholar 29 : Vitamin D receptor, HER-2 polymorphisms and risk of prostate cancer in men with benign prostate hyperplasia. Saudi Med J2004; 25: 447. Google Scholar 30 : Significance of vitamin D receptor gene polymorphism for prostate cancer risk in Japanese. Anticancer Res1999; 19: 4511. Google Scholar 31 : Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based case-control study in China. Cancer Res2001; 61: 4333. Google Scholar 32 : Vitamin D receptor gene Bsm I polymorphism and the susceptibility to prostate cancer in northern Chinese Han population. Zhonghua Nan Ke Xue2003; 9: 413. Google Scholar 33 : Vitamin D receptor (VDR) gene polymorphisms (Bsm1, Fok1, Cdx2) in total prostate cancer (CaP) risk among United States men. AACR Front Cancer Prev Res2004; . p. 141, abstract B120. Google Scholar 34 : Comprehensive assessment of candidate genes and serological markers for the detection of prostate cancer. Cancer Epidemiol Biomarkers Prev2003; 12: 1429. Google Scholar 35 : Association of single nucleotide polymorphism of vitamin D receptor gene start codon and the susceptibility to prostate cancer in the Han nationality in Hubei area. Zhonghua Nan Ke Xue2004; 10: 411. Google Scholar 36 : Vitamin D receptor gene polymorphisms and risk of prostate cancer: a meta-analysis. Cancer Epidemiol Biomarkers Prev2003; 12: 1395. Google Scholar 37 : Polymorphisms in the vitamin D receptor gene and the androgen receptor gene and the risk of benign prostatic hyperplasia. Eur Urol2000; 37: 234. Google Scholar 38 : Meta-analysis in clinical trials. Control Clin Trials1986; 7: 177. Google Scholar 39 : Outcome in prostate cancer associations with skin type and polymorphism in pigmentation-related genes. Carcinogenesis2001; 22: 1343. Google Scholar 40 : Vitamin D receptor start codon polymorphism (FokI) and prostate cancer progression. Cancer Epidemiol Biomarkers Prev2003; 12: 23. Google Scholar 41 : Analysis of polymorphic patterns in candidate genes in Israeli patients with prostate cancer. Isr Med Assoc J2003; 5: 741. Google Scholar 42 : Vitamin D receptor gene polymorphisms and disease free survival after radical prostatectomy. Prostate2004; 61: 267. Google Scholar 43 : Significance of vitamin D receptor gene polymorphism for risk and disease severity of prostate cancer and benign prostatic hyperplasia in Japanese. Urol Int2002; 68: 226. Google Scholar 44 : Association of vitamin D receptor gene polymorphism with prostate cancer. Nippon Rinsho, suppl.2002; 11: 465. Google Scholar 45 : Association of vitamin D receptor gene polymorphism with protection against prostate cancer and benign prostate hyperplasia. Prostate Cancer Prostatic Dis1999; 2: S24. Google Scholar 46 : Prostate cancer risk: associations with ultraviolet radiation, tyrosinase and melanocortin-1 receptor genotypes. Br J Cancer2001; 85: 1504. Google Scholar 47 : Vitamin D receptor polymorphisms and prostate cancer in African-Americans. Proc Annu Meet Am Assoc Cancer Res1997; . p. 38, abstract. Google Scholar 48 : Links between genetic and environmental factors and prostate cancer risk. Prostate1999; 39: 262. Google Scholar 49 : Molecular characterisation and racial distribution of androgen and vitamin D receptor polymorphisms in patients with prostate cancer, benign prostatic hyperplasia and age-matched healthy controls. Prostate Cancer Prostatic Dis1999; 2: S30. Google Scholar 50 : Association of vitamin D receptor polymorphisms with prostate cancer in non-Hispanic white men. Proc Annu Meet Am Assoc Cancer Res1996; . p. 37, abstract. Google Scholar 51 : The association of lumbar disc disease with vitamin-D receptor gene polymorphism. J Bone Joint Surg Am2002; 84-A: 2022. Google Scholar 52 : ApaI, BsmI, EcoRV and TaqI polymorphisms at the human vitamin D receptor gene locus in Caucasians, blacks and Asians. Hum Mol Genet1993; 2: 487. Google Scholar 53 : Genetic variations in the vitamin D receptor, androgen receptor and enzymes that regulate androgen metabolism. J Urol2004; 171: S45. Link, Google Scholar 54 : Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol2001; 177: 145. Google Scholar 55 : Vitamin D receptor: no evidence for allele-specific mRNA stability in cells which are heterozygous for the Taq I restriction enzyme polymorphism. Biochem Biophys Res Commun1997; 238: 77. Google Scholar 56 : Vitamin D receptor 3′-untranslated region polymorphisms: lack of effect on mRNA stability. Biochim Biophys Acta1999; 1453: 311. Google Scholar 57 : Quantification of vitamin D receptor mRNA by competitive polymerase chain reaction in PBMC: lack of correspondence with common allelic variants. J Bone Miner Res1997; 12: 726. Google Scholar 58 : Consequences of vitamin D receptor gene polymorphisms for growth inhibition of cultured human peripheral blood mononuclear cells by 1, 25-dihydroxyvitamin D3. Clin Endocrinol (Oxf)2000; 52: 211. Google Scholar 59 : Sequence variation within the 5′ regulatory regions of the vitamin D binding protein and receptor genes and prostate cancer risk. Prostate2005; 64: 272. Google Scholar 60 : Vitamin D receptor content and transcriptional activity do not fully predict antiproliferative effects of vitamin D in human prostate cancer cell lines. Mol Cell Endocrinol1997; 126: 83. Google Scholar © 2006 by American Urological AssociationFiguresReferencesRelatedDetailsCited byTaneja S (2019) Re: Vitamin D Supplements and Prevention of Cancer and Cardiovascular DiseaseJournal of Urology, VOL. 202, NO. 2, (211-212), Online publication date: 1-Aug-2019. Volume 175Issue 5May 2006Page: 1613-1623 Advertisement Copyright & Permissions© 2006 by American Urological AssociationKeywordsprostatic neoplasmspolymorphismprostatecalcitriolgeneticreceptorsrisk factorsAcknowledgmentsSue Ann Ingles, Robert Nam and Igor Snitcovsky provided race specific genotype frequencies from their studies, and Steve Goodman, Richard Hayes and Robert Getzenberg provided insight and critical reviewed the manuscript.Metrics Author Information Sonja I. Berndt Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health Baltimore and Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland More articles by this author Jennifer L. Dodson James Buchanan Brady Urological Institute, Johns Hopkins Hospital More articles by this author Wen-Yi Huang Baltimore and Division of Cancer Epidemiology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, Maryland More articles by this author Kristin K. Nicodemus Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health More articles by this author Expand All Advertisement PDF downloadLoading ...

IMPORTANCEWe investigated the variation in neuropsychological function explained by risk alleles at the psychosis susceptibility gene ZNF804A and its interacting partners using single nucleotide polymorphisms (SNPs), polygenic scores, and epistatic analyses.Of particular importance was the relative contribution of the polygenic score vs epistasis in variation explained.OBJECTIVES To (1) assess the association between SNPs in ZNF804A and the ZNF804A polygenic score with measures of cognition in cases with psychosis and (2) assess whether epistasis within the ZNF804A pathway could explain additional variation above and beyond that explained by the polygenic score.DESIGN, SETTING, AND PARTICIPANTS Patients with psychosis (n = 424) were assessed in areas of cognitive ability impaired in schizophrenia including IQ, memory, attention, and social cognition.We used the Psychiatric GWAS Consortium 1 schizophrenia genome-wide association study to calculate a polygenic score based on identified risk variants within this genetic pathway.Cognitive measures significantly associated with the polygenic score were tested for an epistatic component using a training set (n = 170), which was used to develop linear regression models containing the polygenic score and 2-SNP interactions.The best-fitting models were tested for replication in 2 independent test sets of cases: (1) 170 individuals with schizophrenia or schizoaffective disorder and (2) 84 patients with broad psychosis (including bipolar disorder, major depressive disorder, and other psychosis). MAIN OUTCOMES AND MEASURESParticipants completed a neuropsychological assessment battery designed to target the cognitive deficits of schizophrenia including general cognitive function, episodic memory, working memory, attentional control, and social cognition. RESULTSHigher polygenic scores were associated with poorer performance among patients on IQ, memory, and social cognition, explaining 1% to 3% of variation on these scores (range, P = .01to .03).Using a narrow psychosis training set and independent test sets of narrow phenotype psychosis (schizophrenia and schizoaffective disorder), broad psychosis, and control participants (n = 89), the addition of 2 interaction terms containing 2 SNPs each increased the R 2 for spatial working memory strategy in the independent psychosis test sets from 1.2% using the polygenic score only to 4.8% (P = .11and .001,respectively) but did not explain additional variation in control participants.CONCLUSIONS AND RELEVANCE These data support a role for the ZNF804A pathway in IQ, memory, and social cognition in cases.Furthermore, we showed that epistasis increases the variation explained above the contribution of the polygenic score.

Glycogen synthase kinase 3β (GSK3β) has been implicated in mood disorders. We previously reported associations between a GSK3β polymorphism and hippocampal volume in major depressive disorder (MDD). We then reported similar associations for a subset of GSK3β-regulated genes. We now investigate an algorithm-derived comprehensive list of genes encoding proteins that directly interact with GSK3β to identify a genotypic network influencing hippocampal volume in MDD.We used discovery (N=141) and replication (N=77) recurrent MDD samples. Our gene list was generated from the NetworKIN database. Hippocampal measures were derived using an optimized Freesurfer protocol. We identified interacting single nucleotide polymorphisms using the machine learning algorithm Random Forest and verified interactions using likelihood ratio tests between nested linear regression models.The discovery sample showed multiple two-single nucleotide polymorphism interactions with hippocampal volume. The replication sample showed a replicable interaction (likelihood ratio test: P=0.0088, replication sample; P=0.017, discovery sample; Stouffer's combined P=0.0007) between genes associated previously with endoplasmic reticulum stress, calcium regulation and histone modifications.Our results provide genetic evidence supporting associations between hippocampal volume and MDD, which may reflect underlying cellular stress responses. Our study provides evidence of biological mechanisms that should be further explored in the search for disease-modifying therapeutic targets for depression.

NRG1 is a schizophrenia candidate gene which regulates brain development and neural function. The minor allele of rs7014762 in the NRG1 5′ core promoter was associated with schizophrenia (p=0.031) and significant predicted reduced NRG1 Type III isoform expression in postmortem human brain of schizophrenia cases (p=0.001). Our results provide additional evidence for transcriptional dysregulation as a biological mechanism implicating NRG1 in schizophrenia risk. Association between NRG1 and schizophrenia was originally discovered via haplotype analysis in an Icelandic sample (HAPICE) at the 5′ end of the gene1, further replicated in a Scottish population2. In the present report, we examined association between schizophrenia and the NRG1 SNP rs7014762 because it is situated within a core promoter region3 and is physically proximal (87bp) to a functional SNP which has been shown to influence NRG1 type IV isoform expression4. In addition to clinical association analyses, we validated the functional effects of rs7014762 by testing for effects on mRNA expression in postmortem human brain. Cases (N=296) and controls (N=365) were ascertained as part of the Clinical Brain Disorders Branch Sibling Study. Probands met DSM-IV criteria for a broad diagnosis category including schizophrenia, schizoaffective disorder, psychosis NOS, delusional disorder, schizotypal, schizoid, or paranoid personality disorder. Control individuals were screened to exclude individuals with psychiatric diagnoses. All participants gave informed consent and self-identified as Caucasian. Blood was collected and DNA was extracted using standard methods. Genotypes were obtained using the Taqman 5′-exonuclease allelic discrimination assay. Postmortem brain tissue was collected with informed consent from the legal next-of-kin. The sample was previously described in detail along with the NRG1 primer and probe sets3,4. Briefly, hippocampi from 84 normal controls (22 females/62 males, 53 African American/25 American Caucasian/5 Hispanic and 1 Asian individual, mean age 40.5 ±(SD) 15.4 years, post mortem interval (PMI) 30.7 ± 13.9 hrs, pH 6.59 ± 0.32); and 44 schizophrenic patients (15 females/29 males, 24 African Americans/20 Caucasians, mean age 49.7 ± 17.2 years, PMI, 36.3 ± 17.7 hrs, pH 6.48 ± 0.28) were available for study. Diagnoses were determined by independent reviews of clinical records and family interviews by two psychiatrists using DSM-IV criteria. Macro- and microscopic neuropathological examinations and toxicology screening were performed prior to inclusion. No differences were observed on variables that potentially affect gene expression in human postmortem brain (i.e. age, PMI, pH and RIN) by rs7014762 genotype group. NRG1 (types I-IV) mRNA splice isoform expression was measured by real-time quantitative RT-PCR using an ABI Prism 7900 sequence detection system with 384-well format (Applied Biosystems, Foster City, CA, USA). Case-control analyses used unconditional logistic regression. Effects of rs7014762 on NRG1 isoform mRNA expression were examined using ANOVA with genotype and diagnosis as independent factors, controlling for race. Where there was a significant genotype bydiagnosis interaction, individual group post hoc tests were examined. P-values were not adjusted for multiple testing. rs7014762 was in Hardy Weinberg equilibrium in cases and controls (p > 0.05). The minor allele of rs7014762 showed significant association with schizophrenia case status (minor allele (T) carrier OR = 1.49 (1.04, 2.15), p-value = 0.031). mRNA expression analysis revealed a significant diagnosis by genotype interaction on type III isoform expression in the hippocampus (F (5, 106) = 5.98; p-value = 0.003). Post hoc analysis showed the effect of rs7014762 was significant only in patients, whereby individuals heterozygous (LSD; p-value = 0.001) or homozygous (LSD p-value = 0.002) for the minor allele exhibited significantly lower levels of NRG1 type III expression compared to major allele homozygotes (Figure 1). No effects of race or race by genotype interactions were observed. No effects of genotype were observed for any other NRG1 isoform. Figure 1 Normalized NRG1 Type III Expression by rs7014762 Genotype in Schizophrenia Cases and Healthy Controls We report association between schizophrenia and rs7014762 in a case-control sample and show that the same allele of rs7014762 in the NRG1 5′ promoter region significantly predicts lower type III isoform expression levels in patients. NRG1 is expressed throughout the human brain, including the hippocampus and prefrontal cortex5, two areas implicated in schizophrenia, and individuals with schizophrenia show abnormalities in ErbB4–NRG1 signaling in the brain versus healthy controls6. In animal models, NRG1 type III isoform has been associated with axonal myelination7, lateral ventricle enlargement, and reduced function in the prefrontal cortex and hippocampus8. Disturbances in myelination/oligodendroglial density in individuals with schizophrenia have been observed9 and suggest reduced structural connectivity may be part of the neurobiology of schizophrenia. The same allele at this SNP has been shown to be associated with increased risk for bipolar disorder10. In addition, rs7014752 is 87 bp from rs6994992 (HAPICE SNP8NRG1243177) and is in moderate LD with this HAPICE SNP (D′ = 0.96, r2 = 0.21). SNP rs6994992 has previously been reported to be a functional promoter variant associated with schizophrenia and the regulated expression of a novel brain-specific isoform of NRG1, type IV, in humans3,4 and was associated with alterations in activation in frontal/temporal lobes, higher risk of psychotic symptomology and reduction in premorbid IQ in schizophrenia patients11. Together these observations suggest that variation in the HAPICE region may impact risk for schizophrenia via transcriptional regulation of multiple NRG1 isoforms.

The National Institute of Mental Health's Research Domain Criteria (RDoC) Initiative “calls for the development of new ways of classifying psychopathology based on dimensions of observable behavior.” As a result of this ambitious initiative, language has been identified as an independent construct in the RDoC matrix. In this article, we frame language within an evolutionary and neuropsychological context and discuss some of the limitations to the current measurements of language. Findings from genomics and the neuroimaging of performance during language tasks are discussed in relation to serious mental illness and within the context of caveats regarding measuring language. Indeed, the data collection and analysis methods employed to assay language have been both aided and constrained by the available technologies, methodologies, and conceptual definitions. Consequently, different fields of language research show inconsistent definitions of language that have become increasingly broad over time. Individually, they have also shown significant improvements in conceptual resolution, as well as in experimental and analytic techniques. More recently, language research has embraced collaborations across disciplines, notably neuroscience, cognitive science, and computational linguistics and has ultimately re‐defined classical ideas of language. As we move forward, the new models of language with their remarkably multifaceted constructs force a re‐examination of the NIMH RDoC conceptualization of language and thus the neuroscience and genetics underlying this concept. © 2016 The Authors. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics Published by Wiley Periodicals, Inc.

Abstract Background Self-harm is common, debilitating and associated with completed suicide and increased all-cause mortality, but there is uncertainty about its causal risk factors, limiting risk assessment and effective management. Neuroticism is a stable personality trait associated with self-harm and suicidal ideation, and correlated with coping styles, but its value as an independent predictor of these outcomes is disputed. Methods Prior history of hospital-treated self-harm was obtained by record-linkage to administrative health data in Generation Scotland:Scottish Family Health Study ( N = 15,798; self-harm cases = 339) and by a self-report variable in UK Biobank ( N = 35,227; self-harm cases = 772). Neuroticism in both cohorts was measured using the Eysenck Personality Questionnaire-Short Form. Associations of neuroticism with self-harm were tested using multivariable regression following adjustment for age, sex, cognitive ability, educational attainment, socioeconomic deprivation, and relationship status. A subset of GS:SFHS was followed-up with suicidal ideation elicited by self-report ( n = 3342, suicidal ideation cases = 158) and coping styles measured by the Coping Inventory for Stressful Situations. The relationship of neuroticism to suicidal ideation, and the role of coping style, was then investigated using multivariable logistic regression. Results Neuroticism was positively associated with hospital-associated self-harm in GS:SFHS (per EPQ-SF unit odds ratio 1.2 95% credible interval 1.1–1.2, p FDR 0.0003) and UKB (per EPQ-SF unit odds ratio 1.1 95% confidence interval 1.1–1.2, p FDR 9.8 × 10 −17 ). Neuroticism, and the neuroticism-correlated coping style, emotion-oriented coping (EoC), were also associated with suicidal ideation in multivariable models. Conclusions Neuroticism is an independent predictor of hospital-treated self-harm risk. Neuroticism and emotion-orientated coping styles are also predictive of suicidal ideation.

The HLA region on chromosome 6 is gene-rich and under selective pressure because of the high proportion of immunity-related genes. Linkage disequilibrium (LD) patterns and allele frequencies in this region are highly differentiated across broad geographical populations, making it a region of interest for population genetics and immunity-related disease studies. We examined LD in this important region of the genome in six European populations using 166 putatively neutral SNPs and the classical HLA-A, -B and -C gene alleles. We found that the pattern of association between classic HLA gene alleles and SNPs implied that most of the SNPs predated the origin of classic HLA gene alleles. The SNPs most strongly associated with HLA gene alleles were in some cases highly predictive of the HLA allele carrier status (misclassification rates ranged from <1 to 27%) in independent populations using five or fewer SNPs, a much smaller number than tagSNP panels previously proposed and often with similar accuracy, showing that our approach may be a viable solution to designing new HLA prediction panels. To describe the LD within this region, we developed a new haplotype clustering method/software based on r(2), which may be more appropriate for use within regions of strong LD. Haplotype blocks created using this proposed method, as well as classic HLA gene alleles and SNPs, were predictive of a northern versus southern European population membership (misclassification error rates ranged from 0 to 23%, depending on which independent population was used for prediction), indicating that this region may be a rich source of ancestry informative markers.

Genome-wide association studies using thousands to hundreds of thousands of single nucleotide polymorphism (SNP) markers and region-wide association studies using a dense panel of SNPs are already in use to identify disease susceptibility genes and to predict disease risk in individuals. Because these tasks become increasingly important, three different data sets were provided for the Genetic Analysis Workshop 15, thus allowing examination of various novel and existing data mining methods for both classification and identification of disease susceptibility genes, gene by gene or gene by environment interaction. The approach most often applied in this presentation group was random forests because of its simplicity, elegance, and robustness. It was used for prediction and for screening for interesting SNPs in a first step. The logistic tree with unbiased selection approach appeared to be an interesting alternative to efficiently select interesting SNPs. Machine learning, specifically ensemble methods, might be useful as pre-screening tools for large-scale association studies because they can be less prone to overfitting, can be less computer processor time intensive, can easily include pair-wise and higher-order interactions compared with standard statistical approaches and can also have a high capability for classification. However, improved implementations that are able to deal with hundreds of thousands of SNPs at a time are required. Genet. Epidemiol. 31(Suppl. 1):S51–S60, 2007. © 2007 Wiley-Liss, Inc.

Background Genetic studies of single gene variants have been criticized as providing a simplistic characterization of the genetic basis of illness risk that ignores the effects of other variants within the same biological pathways. Of candidate biological pathways for schizophrenia (SZ), the cell adhesion molecule (CAM) pathway has repeatedly been linked to both psychosis and neurocognitive dysfunction. Here we tested, using risk allele scores derived from the Schizophrenia Psychiatric Genome-Wide Association Study Consortium (PGC-SCZ), whether alleles within the CAM pathway were correlated with poorer neuropsychological function in patients. Method In total, 424 patients with psychosis were assessed in areas of cognitive ability typically found to be impaired in SZ: intelligence quotient, memory, working memory and attentional control. CAM pathway genes were identified using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Alleles within these genes identified as significantly associated with SZ risk in the PGC-SCZ were then used to calculate a CAM pathway-based polygenic risk allele score for each patient and these scores were tested for association with cognitive ability. Results Increased CAM pathway polygenic risk scores were significantly associated with poorer performance on measures of memory and attention, explaining 1–3% of variation on these measures. Notably, the most strongly associated single nucleotide polymorphism (SNP) in the CAM pathway (rs9272105 within HLA-DQA1 ) explained a similar amount of variance in attentional control, but not memory, as the polygenic risk analysis. Conclusions These data support a role for the CAM pathway in cognitive function, both at the level of individual SNPs and the wider pathway. In so doing these data highlight the value of pathway-based polygenic risk score studies as well as single gene studies for understanding SZ-associated deficits in cognition.

The authors examined prospectively between 1986 and 1997 the relation of irregular menstrual cycles and irregular menstrual bleeding duration earlier in life with risk of hip fracture in 33,434 postmenopausal Iowa women. Over the 318,522 person-years of follow-up, 523 hip fractures were reported. Adjusted for age, smoking, body mass index, waist/hip ratio, and estrogen use, the relative risk of hip fracture in women who reported always having irregular menstrual cycles, compared with women who never had irregular cycles, was 1.36 (95% confidence interval (CI): 1.03, 1.78). Women who reported having irregular menstrual bleeding duration had a 1.40-fold (95% CI: 1.10, 1.78) increased risk of hip fracture compared with women with regular bleeding duration. In addition, women who reported having both irregular menstrual cycles and irregular menstrual bleeding had a 1.82-fold (95% CI: 1.55, 2.15) higher risk of hip fracture than did women who reported neither irregularity. Women who reported only one menstrual disturbance did not have a risk of hip fracture that was significantly different from women who reported no menstrual disturbances. The authors conclude that women with menstrual irregularities are at increased risk of hip fracture, probably because they are estrogen or progesterone deficient.

The latent structure of schizotypy and psychosis-spectrum symptoms remains poorly understood. Furthermore, molecular genetic substrates are poorly defined, largely due to the substantial resources required to collect rich phenotypic data across diverse populations. Sample sizes of phenotypic studies are often insufficient for advanced structural equation modeling approaches. In the last 50 years, efforts in both psychiatry and psychological science have moved toward (1) a dimensional model of psychopathology (eg, the current Hierarchical Taxonomy of Psychopathology [HiTOP] initiative), (2) an integration of methods and measures across traits and units of analysis (eg, the RDoC initiative), and (3) powerful, impactful study designs maximizing sample size to detect subtle genomic variation relating to complex traits (the Psychiatric Genomics Consortium [PGC]). These movements are important to the future study of the psychosis spectrum, and to resolving heterogeneity with respect to instrument and population. The International Consortium of Schizotypy Research is composed of over 40 laboratories in 12 countries, and to date, members have compiled a body of schizotypy- and psychosis-related phenotype data from more than 30000 individuals. It has become apparent that compiling data into a protected, relational database and crowdsourcing analytic and data science expertise will result in significant enhancement of current research on the structure and biological substrates of the psychosis spectrum. The authors present a data-sharing infrastructure similar to that of the PGC, and a resource-sharing infrastructure similar to that of HiTOP. This report details the rationale and benefits of the phenotypic data collective and presents an open invitation for participation.

Abstract Anxiety disorders are common, complex psychiatric disorders with twin heritabilities of 30-60%. We conducted a genome-wide association study of Lifetime Anxiety Disorder (n = 83 565) and an additional Current Anxiety Symptoms (n= 77 125) analysis. The liability scale common variant heritability estimate for Lifetime Anxiety Disorder was 26%, and for Current Anxiety Symptoms was 31%. Five novel genome-wide significant loci were identified including an intergenic region on chromosome 9 that has previously been associated with neuroticism, and a locus overlapping the BDNF receptor gene, NTRK2 . Anxiety showed significant genetic correlations with depression and insomnia as well as coronary artery disease, mirroring findings from epidemiological studies. We conclude that common genetic variation accounts for a substantive proportion of the genetic architecture underlying anxiety.

Antidepressants are the most commonly prescribed psychiatric medication but concern has been raised about significant increases in their usage in high income countries. We aimed to quantify antidepressant prevalence, incidence, adherence and predictors of use in the adult population.The study record-linked administrative prescribing and morbidity data to the Generation Scotland cohort ( N = 11,052), between 2009 and 2016. Prevalence and incidence of any antidepressant use was determined. Antidepressant adherence was measured using Proportion of Days Covered and Medication Possession Ratio. Time-to-event analysis for incident antidepressant use within 5 years of Generation Scotland: Scottish Family Health Study (GS:SFHS) recruitment was performed to reveal patient-level predictors of use.Almost one-third (28.0%, 95%CI 26.9-29.1) of the adults in our sample were prescribed at least one antidepressant in the 5-year period 2012-2016. There was a 36.2% increase in annual prevalence between 2010 and 2016. Incidence was 2.4(2.1-2.7)% per year. The majority of antidepressant episodes (57.6%) were greater than 9 months duration and adherence was generally high (69.0% with Proportion of Days Covered >80%). Predictors of new antidepressant use included history of affective disorder, being female, physical comorbidities, higher neuroticism scores, and lower cognitive function scores.Antidepressant prevalence is greater than previously reported but incidence remains relatively stable. We found the majority of antidepressant episodes to be of relatively long duration with good estimated adherence. Our study supports the hypothesis that increased long-term use among existing (and returning) users, along with wider ranges of indications for antidepressants, has significantly increased the prevalence of these medications.

Abstract Motivation The genomic architecture of human complex diseases is thought to be attributable to single markers, polygenic components and epistatic components. No study has examined the ability of tree-based methods to detect epistasis in the presence of a polygenic signal. We sought to apply decision tree-based methods, C5.0 and logic regression, to detect epistasis under several simulated conditions, varying strength of interaction and linkage disequilibrium (LD) structure. We then applied the same methods to the phenotype of educational attainment in a large population cohort. Results LD pruning improved the power and reduced the type I error. C5.0 had a conservative type I error rate whereas logic regression had a type I error rate that exceeded 5%. Despite the more conservative type I error, C5.0 was observed to have higher power than logic regression across several conditions. In the presence of a polygenic signal, power was generally reduced. Applying both methods on educational attainment in a large population cohort yielded numerous interacting SNPs; notably a SNP in RCAN3 which is associated with reading and spelling and a SNP in NPAS3, a neurodevelopmental gene. Availability and implementation All methods used are implemented and freely available in R. Supplementary information Supplementary data are available at Bioinformatics online.

Researchers conducting family-based association studies have a wide variety of transmission/disequilibrium (TD)–based methods to choose from, but few guidelines exist in the selection of a particular method to apply to available data. Using a simulation study design, we compared the power and type I error of eight popular TD-based methods under different family structures, frequencies of missing parental data, genetic models, and population stratifications. No method was uniformly most powerful under all conditions, but type I error was appropriate for nearly every test statistic under all conditions. Power varied widely across methods, with a 46.5% difference in power observed between the most powerful and the least powerful method when 50% of families consisted of an affected sib pair and one parent genotyped under an additive genetic model and a 35.2% difference when 50% of families consisted of a single affection-discordant sibling pair without parental genotypes available under an additive genetic model. Methods were generally robust to population stratification, although some slightly less so than others. The choice of a TD-based test statistic should be dependent on the predominant family structure ascertained, the frequency of missing parental genotypes, and the assumed genetic model. Researchers conducting family-based association studies have a wide variety of transmission/disequilibrium (TD)–based methods to choose from, but few guidelines exist in the selection of a particular method to apply to available data. Using a simulation study design, we compared the power and type I error of eight popular TD-based methods under different family structures, frequencies of missing parental data, genetic models, and population stratifications. No method was uniformly most powerful under all conditions, but type I error was appropriate for nearly every test statistic under all conditions. Power varied widely across methods, with a 46.5% difference in power observed between the most powerful and the least powerful method when 50% of families consisted of an affected sib pair and one parent genotyped under an additive genetic model and a 35.2% difference when 50% of families consisted of a single affection-discordant sibling pair without parental genotypes available under an additive genetic model. Methods were generally robust to population stratification, although some slightly less so than others. The choice of a TD-based test statistic should be dependent on the predominant family structure ascertained, the frequency of missing parental genotypes, and the assumed genetic model. The testing of preferential transmission of alleles from parents to affected offspring is a common method of assessing association between genetic markers and disease status. In essence, transmission/disequilibrium (TD)–based methods compare the distribution of alleles transmitted to an affected offspring with the distribution of alleles not transmitted. Therefore, TD-based methods often claim to be robust to confounding from population admixture or stratification because the “case” and “control” alleles come from the same set of parents; therefore, they are expected to have identical genetic backgrounds. The role of population subdivision in the confounding of genetic epidemiologic case-control studies has been controversial.1Wacholder S Rothman N Caporaso N Population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias.J Natl Cancer Inst. 2000; 92: 1151-1158Crossref PubMed Scopus (357) Google Scholar, 2Millikan RC Re: population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias.J Natl Cancer Inst. 2001; 93: 156-157Crossref PubMed Scopus (21) Google Scholar, 3Wacholder S Rothman N Caporaso N Response: re: population stratification in epidemiologic studies of common genetic variants and cancer: quantification of bias.J Natl Cancer Inst. 2001; 93: 157-158Crossref Google Scholar, 4Thomas DC Witte JS Point: population stratification: a problem for case-control studies of candidate-gene associations?.Cancer Epidemiol Biomarkers Prev. 2002; 11: 505-512PubMed Google Scholar, 5Wacholder S Rothman N Caporaso N Counterpoint: bias from population stratification is not a major threat to the validity of conclusions from epidemiological studies of common polymorphisms and cancer.Cancer Epidemiol Biomarkers Prev. 2002; 11: 513-520PubMed Google Scholar, 6Khlat M Cazes MH Génin E Guiguet M Robustness of case-control studies of genetic factors to population stratification: magnitude of bias and type I error.Cancer Epidemiol Biomarkers Prev. 2004; 13: 1660-1664PubMed Google Scholar, 7Heiman GA Hodge SE Gorroochurn P Zhang J Greenberg DA Effect of population stratification on case-control association studies. I. Elevation in false positive rates and comparison to confounding risk ratios (a simulation study).Hum Hered. 2004; 58: 30-39Crossref PubMed Scopus (38) Google Scholar, 8Heiman GA Gorroochurn P Hodge SE Greenberg DA Roubustness of case-control studies to population stratification.Cancer Epidemiol Biomarkers Prev. 2005; 14: 1579-1582Crossref PubMed Google Scholar Recently, several researchers have shown evidence of population stratification and that it can lead to spurious association, even in populations once thought to be homogeneous, such as Europeans.9Helgason A Yngvadóttir B Hrafnkelsson B Gulcher J Stefánsson K An Icelandic example of the impact of population structure on association studies.Nat Genet. 2005; 37: 90-95PubMed Google Scholar, 10Campbell CD Ogburn EL Lunetta KL Lyon HN Freedman ML Groop LC Altshuler D Ardlie KG Hirschhorn JN Demonstrating stratification in a European American population.Nat Genet. 2005; 37: 868-872Crossref PubMed Scopus (344) Google Scholar, 11Clayton DG Walker NM Smyth DJ Pask R Cooper JD Maier LM Smink LJ Lam AC Ovington NR Stevens HE et al.Population structure, differential bias and genomic control in a large-scale, case-control association study.Nat Genet. 2005; 37: 1243-1246Crossref PubMed Scopus (442) Google Scholar However, choosing an appropriate TD-based method to employ in a particular study may be difficult because no practical guidelines have been provided about which method may be more powerful for a set of families with a particular structure. Although it is feasible to ascertain full parents-child trios when the disease being studied can be diagnosed during childhood (e.g., autism), studies of late-onset diseases (e.g., diabetes) often suffer from incomplete familial ascertainment. In this report, we focus on the evaluation of various versions of software that have implemented TD-based tests, hoping to provide practical guidelines for users in terms of the choice of study design and the use of statistical methods. Our goal is to clearly illustrate decreases in power when an unwise or invalid test statistic is chosen and to dispel confusion about selection of a TD-based methodology. We sought to assess power and type I error by using a simulation study design to test eight commonly used TD-based methods while varying the genetic model (additive, dominant, or recessive), the family structures, and the frequencies of missing data. We also evaluated TD-based methods under population stratification. The statistical methods considered included the association-in-the-presence-of-linkage statistic (APL)12Martin ER Bass MP Hauser ER Kaplan NL Accounting for linkage in family-based tests of association with missing parental genotypes.Am J Hum Genet. 2003; 73: 1016-1026Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar; the family-based association test (FBAT)13Horvath S Xu X Laird NM The family based association test method: strategies for studying general genotype-phenotype associations.Eur J Hum Genet. 2001; 9: 301-306Crossref PubMed Scopus (654) Google Scholar, 14Lake SL Blacker D Laird NM Family-based tests in the presence of association.Am J Hum Genet. 2000; 67: 1515-1525Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar; the pedigree disequilibrium test (PDT)15Martin ER Monks SA Warren LL Kaplan NL A test for linkage and association in general pedigrees: the pedigree disequilibrium test.Am J Hum Genet. 2000; 67: 146-154Abstract Full Text Full Text PDF PubMed Scopus (518) Google Scholar; the sibship disequilibrium test (SDT)16Horvath S Laird NM A discordant-sibship test for disequilibrium and linkage: no need for parental data.Am J Hum Genet. 1998; 63: 1886-1897Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar; Spielman’s TD test (TDT),17Spielman RS McGinnis RE Ewens WJ Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM).Am J Hum Genet. 1993; 52: 506-516PubMed Google Scholar as implemented in Haploview18Barrett JC Fry B Maller J Daly MJ Haploview: analysis and visualization of LD and haplotype maps.Bioinformatics. 2005; 21: 263-265Crossref PubMed Scopus (11431) Google Scholar (hereafter referred to as the “Haploview TDT”); TDTPhase19Dudbridge F Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol. 2003; 25: 115-121Crossref PubMed Scopus (1044) Google Scholar; TRANSMIT,20Clayton D A generalization of the transmission/disequilibrium test for uncertain-haplotype transmission.Am J Hum Genet. 1999; 65: 1170-1177Abstract Full Text Full Text PDF PubMed Scopus (565) Google Scholar with use of both analytically derived and permutation-based P values (permutationimplementation information is available at Transmit [version 2.5.4] Web site); and the Weinberg log-linear method.21Weinberg CR Wilcox AJ Lie RT A log-linear approach to case-parent-triad data: assessing effects of disease genes that act either directly or through maternal effects and that may be subject to parental imprinting.Am J Hum Genet. 1998; 62: 969-978Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar, 22Weinberg CR Allowing for missing parents in genetic studies of case-parent triads.Am J Hum Genet. 1999; 64: 1186-1193Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar The Weinberg log-linear method is implemented in a statistical analysis system (SAS)–based program and can be found at Clarice R. Weinberg's Web site. For ease of use of this method, we created an R script (available from K.K.N. at [email protected] ) that takes a standard-formatted pedigree file (.ped file) as input and creates a count-based output file formatted for use with the SAS program. PDTPhase19Dudbridge F Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol. 2003; 25: 115-121Crossref PubMed Scopus (1044) Google Scholar is an implementation of the PDT,15Martin ER Monks SA Warren LL Kaplan NL A test for linkage and association in general pedigrees: the pedigree disequilibrium test.Am J Hum Genet. 2000; 67: 146-154Abstract Full Text Full Text PDF PubMed Scopus (518) Google Scholar which, when analysis involves a single SNP, performs identically to the PDT (data not shown) and will not be discussed further. Methods were selected on the basis of frequency of use in published applied-data analyses. To get a baseline estimate of power and type I error for each method, we simulated 1,000 fully genotyped trios. In addition, to determine which methods fared best for late-onset diseases or in studies that have predominantly one parent ascertained, we simulated 1,000 incomplete trios with one parental genotype missing. The assessment of families with two offspring included the following simulation conditions: 50% of families (n=500) were fully genotyped parents–affected child trios, and the remaining 50% of families (n=500) were composed of an affection-discordant sibling pair without parental genotypes, an affection-discordant sibling pair with one parent genotyped, an affected sibling pair without parental genotypes, or an affected sibling pair with one parent genotyped, for a total of 18 simulation conditions. In addition, we simulated 1,000 trios and 1,000 incomplete trios under population stratification. Simulation of 1,000 replicates per each association-present condition was conducted via SIMLA version 2.2.23Bass MP Martin ER Hauser ER Pedigree generation for analysis of genetic linkage and association.Pac Symp Biocomput. 2004; 9: 93-103Google Scholar The minor-allele frequency (MAF) for observed associated SNPs was set at 0.50, except in the simulations under population stratification (discussed below). Type I error was evaluated using a separate set of 1,000 data sets with the same family structure, simulated under the null hypothesis of no linkage and no association by use of Merlin version 1.0.124Abecasis GR Cherney SS Cookson WO Cardon LR Merlin—rapid analysis of dense genetic maps using sparse gene flow trees.Nat Genet. 2002; 30: 97-101Crossref PubMed Scopus (2697) Google Scholar; for the replicates simulated under population stratification, the MAFs under the null hypothesis were set to be the same as in the associated conditions, thus retaining the difference in MAFs. Power was calculated—with the α level held constant within TD-based methods—as the proportion of data sets showing significant evidence for association divided by the total number of replicates. To mimic a realistic candidate-gene family-based study with locus heterogeneity, in all associated conditions, we simulated 75% of the families to have association between the observed associated SNP and disease status and 25% of the families to not have association between disease status and the observed SNP. The nonobserved disease-allele frequency was set to 20% for simulations not under population stratification. For the population-stratification simulations, the nonobserved disease-allele frequency was set to 20% for the first population and to 10% for the second population, and observed allele frequencies varied between the populations: population 1 had an observed associated-allele frequency of 50%, and population 2 had an observed associated-allele frequency of 25%. Disease-allele penetrances for the additive conditions were 0, 0.25, and 0.50, for the three possible genotypes; penetrance increased as the number of disease-associated alleles increased. For dominant disease models, the disease-allele carrier penetrance was 0.50, and, for recessive disease models, the disease-allele homozygote penetrance was 0.50. Simulations considering population stratification used the additive model. Although underlying genetic models used to generate data were varied, all analyses were done blind to genetic model, to closely approximate applied analyses in which the genetic model is unknown. The linkage disequilibrium between the unobserved disease allele and the observed associated marker allele was simulated to be incomplete (D′=0.50), to mimic applied association studies. All methods showed type I error close to 0.05 in fully genotyped trios (table 1), although several methods (FBAT, Haploview TDT, and TRANSMIT) appeared slightly conservative (empirical type I error<0.04). PDT/PDTPhase was the most powerful method under all genetic models, although with a slight increase in type I error (table 1). TRANSMIT was the least powerful method for fully genotyped trios. All other methods performed equally well under the additive and dominant models. Under the recessive model, all methods performed strongly. Among the three methods that perform association testing of incomplete trios (APL, TDTPhase, and TRANSMIT), we observed difference in power by genetic model (table 1). Under the additive model, APL and TDTPhase showed equivalent power, performing significantly better than TRANSMIT. However, under the dominant model, APL was clearly more powerful than TDTPhase and TRANSMIT, and, under the recessive model, TDTPhase showed much higher power to detect association than did TRANSMIT or APL, although the empirical type I error for TDTPhase was slightly anticonservative (0.068).Table 1Power and Type I Error of TD-Based Methods: Full Trios and Incomplete TriosAPLFBATPDT/PDTPhaseHaploview TDTTDTPhaseTransmit: AnalyticalTransmit: PermutationLog-Linear TDTConditionsType I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Full trios: Additive model4.964.6 (60.1–69.1)3.963.6 (59.1–68.1)5.579.2 (76.0–82.4)3.963.5 (59.2–67.8)4.563.6 (59.1–68.1)3.849.4 (44.5–54.3)3.748.2 (43.3–53.1)4.553.5 (48.6–58.4) Dominant model4.951.8 (46.9–56.7)3.950.8 (45.9–55.7)5.568.1 (63.8–72.4)3.950.5 (45.6–55.4)4.554.4 (49.5–59.3)3.839.9 (35.2–44.6)3.740.0 (35.3–44.7)4.545.7 (40.8–50.6) Recessive model4.999.2 (99.0–99.4)3.998.9 (98.8–99.0)5.51.03.998.9 (98.9–99.0)4.598.9 (98.9–99.0)3.895.3 (94.4–96.2)3.794.8 (94.3–95.3)4.598.0 (97.6–98.4)Incomplete trios: Additive model4.837.4 (32.8–42.0)………………6.833.2 (28.9–37.5)5.019.6 (16.5–22.7)4.520.3 (17.1–23.5)…… Dominant model4.830.7 (26.5–34.9)………………6.822.5 (19.1–25.9)5.013.7 (11.4–16.0)4.513.5 (11.2–15.8)…… Recessive model4.866.4 (62.0–70.8)………………6.890.2 (88.5–91.9)5.069.7 (65.6–73.8)4.569.8 (65.7–73.9)……Note.—Data are percentages. Open table in a new tab Note.— Data are percentages. Within the condition of families with 50% affected sib pairs with or without a single parent genotyped, TDTPhase consistently showed high power to detect association (table 2). In the condition of 50% affected sib pairs without parental genotypes, both TDTPhase and TRANSMIT with use of the analytically derived P values performed similarly well; after a single pair of parents' genotypes were added, APL, TDTPhase, and TRANSMIT with use of analytically derived P values showed the highest power to detect association. As expected, the log-linear model was more powerful than the Haploview TDT under conditions with 50% affected sib pairs and one parental genotype available, because the log-linear method is able to use the incomplete trio families. Indeed, under a recessive model, the log-linear method is as powerful as APL, TDTPhase, and TRANSMIT. The Haploview TDT, PDT, and FBAT with either variance estimate showed the lowest power to detect association in all affected sib pair conditions. The greatest differences in power for affected sib pair families was 44% under the condition of one parental genotype available with the use of an additive model (in the comparison between TDTPhase and PDT). All methods performed very well under recessive genetic models.Table 2Power and Type I Error of TD-Based Methods: 50% Fully Genotyped Trios and 50% Affected Sib PairsAPLFBATFBAT Empirical VariancePDT/PDTPhaseHaploview TDTTDTPhaseTransmit: AnalyticalTransmit: PermutationLog-Linear TDTConditionsType I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Affected sib pairs: Additive model4.652.2 (47.3–57.1)5.137.7 (33.1–42.3)5.439.1 (34.4–43.8)5.328.3 (24.3–32.3)5.137.7 (33.1–42.3)4.660.9 (56.2–65.6)5.257.2 (52.4–62.0)5.150.9 (46.0–55.8)6.634.9 (30.4–39.4) Dominant model4.634.5 (30.0–38.9)5.131.6 (27.4–35.8)5.433.1 (28.8–37.4)5.323.4 (19.9–26.9)5.131.9 (27.6–36.2)4.646.6 (41.7–51.5)5.245.3 (40.4–50.2)5.139.4 (34.7–44.1)6.628.1 (26.1–30.1) Recessive model4.696.6 (96.0–97.2)5.192.2 (90.8–93.6)5.492.2 (90.8–93.6)5.387.0 (84.8–89.2)5.192.2 (90.8–93.6)4.698.1 (97.7–98.5)5.296.5 (95.8–97.2)5.188.7 (86.7–90.7)6.689.1 (87.2–91.0)Affected sib pairs and one parent: Additive model5.873.8 (70.0–77.6)5.041.5 (36.7–46.3)4.941.0 (36.3–45.7)4.829.8 (25.7–33.9)5.041.5 (36.7–46.3)5.476.3 (72.8–79.8)5.272.8 (68.9–76.7)5.168.4 (64.2–72.6)5.657.3 (52.5–62.1) Dominant model5.852.0 (47.1–56.9)5.031.5 (27.3–35.7)4.931.5 (27.3–35.7)4.822.6 (19.2–26.0)5.031.9 (27.6–36.2)5.460.3 (55.6–65.0)5.257.6 (52.8–62.4)5.152.3 (47.4–57.2)5.642.8 (38.0–47.6) Recessive model5.899.1 (98.9–99.3)5.092.2 (90.8–93.6)4.992.2 (90.8–93.6)4.886.5 (84.2–88.8)5.092.2 (90.8–93.6)5.499.0 (98.8–99.2)5.298.9 (98.9–99.0)5.193.4 (92.2–94.6)5.697.2 (96.7–97.7)Note.—Data are percentages. Open table in a new tab Note.— Data are percentages. Under conditions of 50% affection-discordant families ascertained, APL and TRANSMIT with use of either analytically derived or permutation-based P values showed the highest power, compared with other test statistics (table 3). FBAT and PDT performed reasonably well under all genetic models when 50% of families consisted of discordant sib pairs without parental genotypes; however, both methods showed less-than-optimal power in the conditions of discordant sibship plus one parental genotype under an additive or dominant model. Interestingly, the opposite trend was observed with the TDTPhase method; this method performed strongly among families with one parental genotype and discordant sib pairs but performed less well when both parental genotypes were missing. The methods giving the lowest power to detect association were the Haploview TDT, the SDT, and the log-linear TDT under conditions of 50% discordant sibships with missing genotype data for both parents. The greatest difference in power was observed for the discordant sibship with no parental data with use of a dominant model: 34.2% (in the comparison between FBAT and the log-linear TDT). The main explanation for why the Haploview TDT and the log-linear TDT performed less well than the other methods is that both of these methods were able to use only 50% of the families in each replicate because the implementation of both methods cannot use families with both parental genotypes missing. The recessive genetic model condition gave the highest power, and the dominant model gave the lowest.Table 3Power and Type I Error of TD-Based Methods: 50% Fully Genotyped Trios and 50% Discordant Sib PairsAPLFBATPDT/PDTPhaseSDTHaploview TDTTDTPhaseTransmit: AnalyticalTransmit: PermutationLog-Linear TDTConditionsType I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Discordant sib pairs: Additive model4.360.7 (56.0–65.4)4.658.3 (53.5–63.1)4.658.6 (53.8–63.4)4.726.5 (22.7–30.3)3.932.3 (28.0–36.6)3.545.9 (41.0–50.8)4.259.4 (64.6–64.1)4.159.0 (54.3–63.7)4.026.5 (22.7–30.3) Dominant model4.354.2 (49.3–59.1)4.657.9 (63.1–62.7)4.657.1 (52.3–61.9)4.728.5 (24.5–32.5)3.930.4 (26.3–34.5)3.538.6 (34.0–43.2)4.255.3 (50.5–60.1)4.155.7 (50.9–60.5)4.022.7 (19.3–26.1) Recessive model4.399.4 (99.3–99.5)4.699.5 (99.4–99.6)4.699.5 (99.4–99.6)4.775.7 (72.1–79.3)3.990.5 (88.8–92.2)3.596.8 (96.2–97.4)4.299.5 (99.4–99.6)4.199.5 (99.4–99.6)4.085.1 (82.1–87.6)Discordant sib pairs and one parent: Additive model5.170.1 (66.0–74.2)4.458.9 (54.2–63.6)4.358.0 (53.2–62.8)4.627.6 (23.7–31.5)4.536.7 (32.1–41.3)5.469.6 (65.5–73.7)5.168.4 (64.2–72.6)4.866.9 (62.6–71.2)4.550.8 (45.9–55.7) Dominant model5.160.1 (55.4–64.8)4.452.1 (47.2–57.0)4.352.0 (47.1–56.9)4.628.6 (24.6–32.6)4.530.4 (26.3–34.5)5.460.4 (55.7–65.1)5.162.4 (57.8–67.0)4.861.3 (56.7–65.9)4.541.5 (36.7–46.3) Recessive model5.195.3 (94.4–96.2)4.494.0 (92.9–95.1)4.393.8 (95.7–94.9)4.676.1 (72.5–79.7)4.591.2 (89.6–92.8)5.492.4 (91.0–93.8)5.193.2 (92.0–94.4)4.892.6 (91.3–93.9)4.582.0 (79.1–84.9)Note.—Data are percentages. Open table in a new tab Note.— Data are percentages. Virtually all methods showed reduced power under the presence of population stratification (table 4). This reduction in power was most likely caused by the reduction in MAFs and the resulting smaller number of informative families, although the loss of power was very moderate for fully genotyped trios (variations ranged from 1.8% higher power for TRANSMIT with permutation-based P values to 6.3% lower power for PDT/PDTPhase) and was only ∼10% for incomplete trios (reductions ranged from 7.1% lower power for TRANSMIT with analytical P values to 13.7% lower power for APL) versus simulations conducted without population stratification. As expected, for fully genotyped trios, the type I error was not increased for joint association and linkage methods such as PDT, the log-linear TDT, and the Haploview TDT. Similarly, we did not observe an increase in type I error in associationbased methods such as TRANSMIT and TDTPhase. PDT/PDTPhase had the highest power (72.9%) to detect association under population stratification, followed by the Haploview TDT (62.6% power) and TDTPhase (61.7% power). However, with incomplete trios, we observed a slightly anticonservative type I error for the association-based method TDTPhase (0.062) but not for the other association-based method, TRANSMIT (both analytical and permutation-based type I error<0.04). For incomplete trios, the joint linkage and association method APL (23.7% power) and the association method TDTPhase (25.2% power) showed equivalent power, although APL did not show an inflated type I error rate. Both APL and TDTPhase showed significantly increased power versus TRANSMIT (12.5% and 11.3% power with use of analytic and permutation-based P values, respectively) for incomplete trios.Table 4Power and Type I Error of TD-Based Methods under Population Stratification and the Additive Model: Fully Genotyped Trios and Incomplete TriosAPLFBATPDT/PDTPhaseHaploview TDTTDTPhaseTransmit: AnalyticalTransmit: PermutationLog-Linear TDTConditionType I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Type I ErrorPower (95% CI)Trios3.959.2 (54.5–63.9)5.261.7 (57.1–66.3)4.572.9 (69.0–76.8)5.562.6 (60.3–64.9)5.061.7 (57.1–66.3)5.249.3 (46.8–51.8)5.450.0 (47.5–52.5)4.448.3 (43.4–53.2)Incomplete trios3.623.7 (21.9–25.5)………………6.225.2 (23.3–27.1)3.812.5 (11.2–13.4)3.011.3 (10.3–12.3)……Note.—Data are percentages. Open table in a new tab Note.— Data are percentages. Here, we report differences in power across family type, missing-parental-data conditions, genetic models, and population stratification for several commonly used TD-based methods implemented in freely available software. In planning and executing candidate-gene or genomewide association data analyses with the use of family-based association tests, it is important to consider the type of sibship ascertained (concordant or discordant for disease status), the proportion of missing parental genotypes, and the assumed genetic model as key factors in the decision making about appropriate TD-type statistics applied to achieve the highest likelihood of detecting association when association exists. Simulation study designs are useful in assessing the behavior of test statistics under a restricted set of scenarios, but generalization to all possible scenarios must be cautiously applied. However, we believe our recommendations are a helpful addition to the applied researcher’s selection of appropriate test statistics, and we are encouraged by the ability of all methods to detect association under locus heterogeneity, incomplete LD, and incomplete penetrance, which are likely to be found in applied candidate-gene association studies. In our view, the reason for such vast differences in power across TD-based test statistics is twofold. First, there are differences in the number of families that are informative for different methods; second, how each method handles missing parental genotypes varies (table 5). In the simulations with affected sibling pairs and missing parental data, both FBAT and PDT cannot use 50% of the families in the analysis, so the resulting loss in power is not surprising. We also note that the Haploview TDT considers only families with fully genotyped parents, so it used only trios for all conditions in the calculation of the test statistic. A graphical-user-interface–based software for the Windows operating system that implements the full method proposed by Spielman et al.17Spielman RS McGinnis RE Ewens WJ Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM).Am J Hum Genet. 1993; 52: 506-516PubMed Google Scholar and that can include families with missing parental genotypes is available from the Spielman Lab: TDT & S-TDT Web site. This software also implements the S-TDT28 and the combined TDT/S-TDT28 statistic and will perform better than the method implemented in Haploview under conditions of missing parental data and discordant sibships, because it can use a larger number of sibships and can score transmissions to unaffected offspring. In addition, the log-linear method has been implemented only for trio data, although it can assess association in families with some missing parental data.22Weinberg CR Allowing for missing parents in genetic studies of case-parent triads.Am J Hum Genet. 1999; 64: 1186-1193Abstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar, 23Bass MP Martin ER Hauser ER Pedigree generation for analysis of genetic linkage and association.Pac Symp Biocomput. 2004; 9: 93-103Google Scholar The SDT uses only discordant sibships, so the results in table 3 are based on 50% of families. We include these “unfair” comparisons as an illustration for researchers who have to decide which test is most appropriate and to show the magnitude of decreases in power that occur when an unwise or invalid choice of test statistic is made. TRANSMIT infers missing parental genotypes by using sibling genotype information, which is appropriate in situations where there are not multiple affected siblings and there is no linkage or when discordant sibships are used.12Martin ER Bass MP Hauser ER Kaplan NL Accounting for linkage in family-based tests of association with missing parental genotypes.Am J Hum Genet. 2003; 73: 1016-1026Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar However, it has been shown that the score test used in TRANSMIT has an inflated type I error rate when there are missing parental genotypes and multiple affected siblings and when the MAF is <0.50 because of increased allele sharing between siblings as a result of linkage.12Martin ER Bass MP Hauser ER Kaplan NL Accounting for linkage in family-based tests of association with missing parental genotypes.Am J Hum Genet. 2003; 73: 1016-1026Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar The APL statistic handles correlation of transmitted alleles imposed by linkage between affected siblings and thus retains the appropriate type I error rate12Martin ER Bass MP Hauser ER Kaplan NL Accounting for linkage in family-based tests of association with missing parental genotypes.Am J Hum Genet. 2003; 73: 1016-1026Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar; therefore, the use of APL may be preferable to TRANSMIT when multiple affected siblings are included in analysis, missing parental genotypes are frequent, and the observed MAF is not equal to 0.50. TDTPhase implements the full likelihood described by Clayton20Clayton D A generalization of the transmission/disequilibrium test for uncertain-haplotype transmission.Am J Hum Genet. 1999; 65: 1170-1177Abstract Full Text Full Text PDF PubMed Scopus (565) Google Scholar and conducts a likelihood-ratio test between nested models. However, the parental part of the likelihood depends on a population-based model for parental genotype frequencies, which can provide incorrect inference if the population model for parental genotype frequencies is misspecified,19Dudbridge F Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol. 2003; 25: 115-121Crossref PubMed Scopus (1044) Google Scholar especially for a high frequency of missing parental genotypes. The subsequent statistical test is an unconditional logistic regression comparing “case” genotype frequencies with “pseudocontrol” genotype frequencies and is not robust to population stratification unless a permutation procedure that can be time consuming is used,19Dudbridge F Pedigree disequilibrium tests for multilocus haplotypes.Genet Epidemiol. 2003; 25: 115-121Crossref PubMed Scopus (1044) Google Scholar although we observed only a modest increase in type I error (6.2%) in the incomplete-trios condition simulated under population stratification. In addition, the test statistic calculated in TDTPhase is also likely to suffer from the same bias as the score statistic from TRANSMIT, resulting in an inflated type I error rate when multiple affected offspring are ascertained, because it does not take into account correlation between transmissions to affected offspring when linkage is present. Our recommendation is to use APL when a large proportion of families have multiple affected siblings and missing parental genotype data are frequent, to assure appropriate type I error, but to consider APL, TRANSMIT, and TDTPhase as the test statistics of choice when discordant sibships are the predominant ascertained family structure. Under population stratification in fully genotyped trios, we suggest the use of PDT/PDTPhase, TDTPhase, or the Haploview TDT, but, within incomplete trios, we suggest the use of APL to retain adequate power with appropriate type I error rate.Table 5Permissible Family Structures, Markers Accepted, and Input Data File Types of TD-Based Methods EvaluatedMethodPermissible Family Structure(s)Parental GenotypesType(s) of Markers AcceptedData File Type(s)aFull LINKAGE pedigree file contains the following columns before genotype data: family number, individual number, father’s identification (ID) number, mother's ID number, first parental offspring's ID number, next paternal sibling's ID number, next maternal sibling's ID number, sex, proband status, and affection status. The LINKAGE data file is a separate file that lists descriptions of the data contained in the pedigree file. An abbreviated pedigree file contains only the following columns before genotype data: family number, individual number, father’s ID number, mother’s ID number, sex, and affection status.APLParent-child trios, affected sibships, discordant sibshipsComplete and incompleteSNPsFull LINKAGE pedigree file and data fileFBATParent-child trios, discordant sibshipsComplete and incompleteSNPs, multiallelic markersAbbreviated pedigree fileFBAT: empirical variance estimateParent-child trios, affected sibships, discordant sibshipsComplete and incompleteSNPs, multiallelic markersAbbreviated pedigree filePDT/PDTPhaseParent-child trios, affected sibships, discordant sibshipsComplete and incompleteSNPs, multiallelic markersPDT: full LINKAGE pedigree file and data file; PDTPhase: abbreviated pedigree fileSDTDiscordant sibshipsNoneSNPs, multiallelic markersAbbreviated pedigree fileHaploview TDTParent-child triosCompleteSNPsAbbreviated pedigree fileTDTPhaseParent-child trios, discordant sibshipsComplete and incompleteSNPs, multiallelic markersAbbreviated pedigree fileTRANSMIT: analyticalParent-child trios, discordant sibshipsComplete and incompleteSNPs, multiallelic markersAbbreviated pedigree fileTRANSMIT: permutationParent-child trios, discordant sibshipsComplete and incompleteSNPs, multiallelic markersAbbreviated pedigree fileLog-linear TDTParent-child triosComplete and incompleteSNPsCount-based input filebAn R script is available from K.K.N. (at [email protected]), to recode abbreviated pedigree files into count-based input files.a Full LINKAGE pedigree file contains the following columns before genotype data: family number, individual number, father’s identification (ID) number, mother's ID number, first parental offspring's ID number, next paternal sibling's ID number, next maternal sibling's ID number, sex, proband status, and affection status. The LINKAGE data file is a separate file that lists descriptions of the data contained in the pedigree file. An abbreviated pedigree file contains only the following columns before genotype data: family number, individual number, father’s ID number, mother’s ID number, sex, and affection status.b An R script is available from K.K.N. (at [email protected] ), to recode abbreviated pedigree files into count-based input files. Open table in a new tab We note that, for the purposes of this study, we employed the joint null hypothesis of no linkage and no association as our null hypothesis of interest, even though some methods are robust to linkage in the detection of association. As discussed by Laird and Lange,25Laird NM Lange C Family-based designs in the age of large-scale gene-association studies.Nat Rev Genet. 2006; 7: 385-394Crossref PubMed Scopus (316) Google Scholar the joint null hypothesis may be the null hypothesis of interest when considering a candidate gene that is not under a linkage peak or for genomewide association studies. Therefore, the results reported in the present study cannot be generalized to studies designed to follow up on regions identified as potentially harboring a disease gene from a linkage study. One condition not considered in the present study is that of informatively missing parental genotypes. When a parental genotype is correlated with its probability of being missing, the distribution of observed genotypes differs from that of the missing genotypes, and thus the missing parental genotypes are informatively missing.26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 27Chen YH New approach to association testing in case-parental designs under informative missingness.Genet Epidemiol. 2004; 27: 131-140Crossref PubMed Scopus (25) Google Scholar Informatively missing parental genotypes may be induced through population stratification26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar or association between a disease that creates a higher probability of missingness—for example, an aggressive form of cancer26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar or a debilitating psychiatric disorder, such as schizophrenia—and a genetic marker. Many TD-based tests that allow for missing parental data assume that the distributions of observed and unobserved parental genotypes are the same. Simulation studies have shown that this assumption, if violated, can result in inflated type I error.26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar Two methods have been proposed for use under informative missingness to retain appropriate type I error rates.26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar, 27Chen YH New approach to association testing in case-parental designs under informative missingness.Genet Epidemiol. 2004; 27: 131-140Crossref PubMed Scopus (25) Google Scholar Because informative missingness may be a common feature of family-based genetic association studies, a planned simulation study will consider methods assessed in the present study plus the methods proposed by Allen et al.26Allen AS Rathouz PJ Satten GA Informative missingness in genetic association studies: case-parent designs.Am J Hum Genet. 2003; 72: 671-680Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar and Chen,27Chen YH New approach to association testing in case-parental designs under informative missingness.Genet Epidemiol. 2004; 27: 131-140Crossref PubMed Scopus (25) Google Scholar to determine how several TD-based methods fare under varying scenarios of informatively missing parental data. Replication of candidate-gene TD-based association results has been inconsistent for many diseases and may, in part, be because of differences in power of different TD methods used in applied analyses. Given the disparate power of different TD test statistics, we remain optimistic that some failures to replicate can be reconciled by use of the most appropriate TD-based methodology available, using our simulation study results as a guide to selection of a more appropriate TD-based test statistic, given the family structure, type of sibship ascertained, and genetic model, if known. We are grateful for the helpful and insightful comments of two anonymous reviewers and for the thoughtful discussion with Drs. Joan Bailey-Wilson and Robert Elston. This study used the high-performance computational capabilities of the Biowulf Linux cluster at the National Institutes of Health, Bethesda (Biowulf Cluster at NIH Web site). Y.Y.S. is supported in part by grant RO3CA123620-01.

Lower performances in cognitive ability in individuals with Major Depressive Disorder (MDD) have been observed on multiple occasions. Understanding cognitive performance in MDD could provide a wider insight in the aetiology of MDD as a whole. Using a large, well characterised cohort (N = 7012), we tested for: differences in cognitive performance by MDD status and a gene (single SNP or polygenic score) by MDD interaction effect on cognitive performance. Linear regression was used to assess the association between cognitive performance and MDD status in a case-control, single-episode-recurrent MDD and control-recurrent MDD study design. Test scores on verbal declarative memory, executive functioning, vocabulary, and processing speed were examined. Cognitive performance measures showing a significant difference between groups were subsequently analysed for genetic associations. Those with recurrent MDD have lower processing speed versus controls and single-episode MDD (β = -2.44, p = 3.6 × 10-04; β = -2.86, p = 1.8 × 10-03, respectively). There were significantly higher vocabulary scores in MDD cases versus controls (β = 0.79, p = 2.0 × 10-06), and for recurrent MDD versus controls (β = 0.95, p = 5.8 × 10-05). Observed differences could not be linked to significant single-locus associations. Polygenic scores created from a processing speed meta-analysis GWAS explained 1% of variation in processing speed performance in the single-episode versus recurrent MDD study (p = 1.7 × 10-03) and 0.5% of variation in the control versus recurrent MDD study (p = 1.6 × 10-10). Individuals with recurrent MDD showed lower processing speed and executive function while showing higher vocabulary performance. Within MDD, persons with recurrent episodes show lower processing speed and executive function scores relative to individuals experiencing a single episode.

Compared to their term-born peers, children born very preterm are at risk for poorer cognitive, academic and behavioral outcomes, however this finding may have been confounded by lower parental education level in the very preterm children. Studies that compare very preterm and term-born children with comparable (high) parental education level are needed to assess the true effect of very preterm birth on outcomes.To compare cognitive, academic and behavioral functioning in very preterm and term-born children of highly educated parents. To examine whether outcomes differ for children of whom one or both parents are highly educated.Cross-sectional study with a term-born comparison group.113 very preterm children and 38 term-born children aged 8–12 years old, with highly educated parents.Cognitive functioning (Intelligence Quotient), academic functioning (arithmetic facts and reading) and parent and teacher rated behavioral functioning (attention, executive function, hyperactivity, and emotional, conduct and peer problems). Parental education was considered high when children had two highly educated parents or one highly- and one middle educated parent.Very preterm children had significantly poorer cognitive (difference of 10 IQ points) and behavioral functioning than their term-born peers, but did not differ on academic functioning. Children with one highly educated parent performed poorer than children with two highly educated parents on most outcome measures.Performance of very preterm children should be compared to term-born peers with parents having comparable educational levels for accurate assessment of outcomes. The number of highly educated parents also impacts outcomes.

Previous studies have reported conflicting results regarding the association of the Q7R polymorphism in the Saitohin gene with late-onset Alzheimer disease (AD). Given that AD is a tauopathy but no mutations or polymorphisms in Tau have been consistently associated with AD, and that Saitohin is nested in intron 9 of Tau and shares a similar expression pattern, we tested this association in 690 multiplex AD families and in a case-control sample (903 patients and 320 controls). We found no evidence of significant association of this polymorphism with risk of AD using family-based and case-control tests of association.

Risk of complex disorders is thought to be multifactorial, involving interactions between risk factors. However, many genetic studies assess association between disease status and markers one single-nucleotide polymorphism (SNP) at a time, due to the high-dimensional nature of the search space of all possible interactions. Three ensemble methods have been recently proposed for use in high-dimensional data (Monte Carlo logic regression, random forests, and generalized boosted regression). An intuitive way to detect an association between genetic markers and disease status is to use variable importance measures, even though the stability of these measures in the context of a whole-genome association study is unknown. For the simulated data of Problem 3 in the Genetic Analysis Workshop 15 (GAW15), we examined the variability of both rankings and magnitude of variable importance measures using 10 variables simulated to participate in gene x gene and gene x environment interactions. We conducted 500 analyses per method on one randomly selected replicate, tallying the rankings and importance measures for each of the 10 variables of interest. When the simulated effect size was strong, all three methods showed stable rankings and estimates of variable importance. However, under conditions more commonly expected to be encountered in complex diseases, random forests and generalized boosted regression showed more stable estimates of variable importance and variable rankings. Individuals endeavoring to apply statistical learning methods to detect interaction in complex disease studies should perform repeated analyses in order to assure variable importance measures and rankings do not vary greatly, even for statistical learning algorithms that are thought to be stable.

In order to investigate the selection bias of subjects for inpatient human cocaine studies, characteristics of 859 potential subjects were examined. Excluded subjects compared with accepted group were more likely to be single and male, currently use drugs other than cocaine, have a history of intravenous cocaine use, and have medical or mental health problems or physical complaints. Subjects who were accepted but did not participate, compared with participants, were likely to spend more money on cocaine. These results suggest that potential subjects who were accepted to our research studies may not accurately represent all potential subjects for several important subject characteristics.

Applications of machine learning (ML) in biomedical science are growing rapidly, spurred by interdisciplinary collaborations, aggregation of large datasets, accessibility of analytic routines, and availability of powerful computers. With this increased usage comes a responsibility for education, borne equally by data scientists plying their wares in medical research and biomedical scientists harnessing such methods to glean knowledge from data. This article provides a critical review of ML, covering common ML methods and historical trends of their use in psychiatry, and identifying areas of opportunity for future applications of ML in biological psychiatry. We also establish the ML in Psychiatry (MLPsych) Consortium, enumerate its objectives, and provide a set of standards (Guidelines for REporting ML Investigations in Neuropsychiatry [GREMLIN]) for designing and reporting studies that use ML. This review serves as a cautiously optimistic primer on ML for those on the precipice as they prepare to dive into the field, either as dedicated methodological practitioners or, at the very least, well-informed consumers.

Abstract Major depressive disorder (MDD) is a heritable and highly debilitating condition. It is commonly associated with subcortical volumetric abnormalities, the most replicated of these being reduced hippocampal volume. Using the most recent published data from ENIGMA consortium’s genome-wide association study (GWAS) of regional brain volume, we sought to test whether there is shared genetic architecture between 8 subcortical brain volumes and MDD. Using LD score regression utilising summary statistics from ENIGMA and the Psychiatric Genomics Consortium, we demonstrated that hippocampal volume was positively genetically correlated with MDD (r G =0.46, P =0.02), although this did not survive multiple comparison testing. None of other six brain regions studied were genetically correlated and amygdala volume heritability was too low for analysis. We also generated polygenic risk scores (PRS) to assess potential pleiotropy between regional brain volumes and MDD in three cohorts (Generation Scotland; Scottish Family Health Study (n=19,762), UK Biobank (n=24,048) and the English Longitudinal Study of Ageing (n=5,766). We used logistic regression to examine volumetric PRS and MDD and performed a meta-analysis across the three cohorts. No regional volumetric PRS demonstrated significant association with MDD or recurrent MDD. In this study we provide some evidence that hippocampal volume and MDD may share genetic architecture, albeit this did not survive multiple testing correction and was in the opposite direction to most reported phenotypic correlations. We therefore found no evidence to support a shared genetic architecture for MDD and regional subcortical volumes.

Abstract Depression is a common psychiatric disorder with substantial recurrence risk. Accurate prediction from easily collected data would aid in diagnosis, treatment and prevention. We used machine learning in the Generation Scotland cohort to predict lifetime risk of depression and, among cases, recurrent depression. Rank aggregation was used to combine results across ten different algorithms and identify highly predictive variables. The model containing all but the cardiometabolic predictors had the highest predictive ability on independent data. Rank aggregation produced a reduced set of predictors without decreasing predictive performance (lifetime: 20 out of 154 predictors and Receiver Operating Characteristic area under the curve (AUC)=0·84, recurrent: 10 out of 180 predictors and AUC=0·76). Here we develop a pipeline which leads to a small set of highly predictive variables. This information can be easily collected with a smartphone ‘application’ to help diagnosis and treatment, while longitudinal tracking may help patients in self-management. Significance Depression is the most common psychiatric disorder and a leading cause of disability worldwide. Patients are often diagnosed and treated by non-specialist clinicians who have limited time available to assess them. We present a novel methodology which allowed us to identify a small set of highly predictive variables for a diagnosis of depression, or recurrent depression in patients. This information can easily be collected using a tablet or smartphone application in the clinic to aid diagnosis.

Accumulating evidence suggests obstetric complications (OCs) impact future health and they have been implicated in the aetiology of psychiatric disorders. OCs are hypothesised to interact with underlying genetics to increase risk. We undertook a genome-wide search of interactions with OCs for a variety of outcomes in the Generation Scotland cohort.

Major Depressive Disorder (MDD) is a major cause of suffering and economic burden. Studies have shown that MDD patients perform significantly less well on some cognitive domains, which may be influenced by genetics.

Sufficient connexin-mediated intercellular coupling is critical to maintain gap junctional communication for proper cardiac function. Alterations in connexin phosphorylation state, particularly dephosphorylation of connexin 43 (Cx43), may impact cell coupling and conduction in disease states. Cx43 dephosphorylation may be carried out by protein phosphatase activity. Here, we present an overview of the key phosphatases known to interact with Cx43 or modulators of Cx43, as well as some possible therapeutic targets to regulate phosphatase activity in the heart.

Lower performances in cognitive ability in individuals with Major Depressive Disorder (MDD) have been observed on multiple occasions. Understanding cognitive performance in MDD could provide a wider insight in the aetiology of MDD as a whole. Using a large, well characterised cohort (N = 7012), we tested for: differences in cognitive performance by MDD status and a gene (single SNP or polygenic score) by MDD interaction effect on cognitive performance. Linear regression was used to assess the association between cognitive performance and MDD status in a case-control, single-episode-recurrent MDD and control-recurrent MDD study design. Test scores on verbal declarative memory, executive functioning, vocabulary, and processing speed were examined. Cognitive performance measures showing a significant difference between groups were subsequently analysed for genetic associations. Those with recurrent MDD have lower processing speed versus controls and single-episode MDD (β = -2.44, p = 3.6 × 10-04; β = -2.86, p = 1.8 × 10-03, respectively). There were significantly higher vocabulary scores in MDD cases versus controls (β = 0.79, p = 2.0 × 10-06), and for recurrent MDD versus controls (β = 0.95, p = 5.8 × 10-05). Observed differences could not be linked to significant single-locus associations. Polygenic scores created from a processing speed meta-analysis GWAS explained 1% of variation in processing speed performance in the single-episode versus recurrent MDD study (p = 1.7 × 10-03) and 0.5% of variation in the control versus recurrent MDD study (p = 1.6 × 10-10). Individuals with recurrent MDD showed lower processing speed and executive function while showing higher vocabulary performance. Within MDD, persons with recurrent episodes show lower processing speed and executive function scores relative to individuals experiencing a single episode.

Correction to: Molecular Psychiatry (2006) 11, 1062–1065; doi:10.1038/sj.mp.4001878 Following publication of the above article, the authors noted that two copies of Table 1 (an incomplete version on page 1063 and a complete version on page 1064) were incorrectly printed. The full table appears on page 1064.

We build an agent-based system for supporting correlation analysis between human behavioural and non-behavioural patterns. A novel social norm specification language is leveraged to create an interaction model based communication engine for choreographing distributed systems, offering a communication environment for multiple interacting players. Categorising sets of players based on their interaction behaviours allows labelling the other patterns, which the system uses to further its understanding relationship between the two traits. While existing analysis methods are manually applied, non-user-editable and typically opaque, the system offers an end-to-end computing framework and protocols which are modifiable for specific users. Evaluation for this system relies on tests for categories of people who are mentally depressed, where traditional questionnaire-based methods are superseded by methods that use more objective behavioural tests. This approach to evaluation through behavioural experimentation is intended not only to classify sub-types of depression cases which would facilitate elucidation of aetiology but evaluates system performance in a real-world scenario.

Obstetric Complications (OC) have been implicated in the aetiology of psychiatric disorders, and are hypothesised to interact with underlying genetics to increase risk. We undertook a genome-wide search of interactions with OCs for psychopathology and personality traits related to psychiatric disorders in the Generation Scotland cohort. OCs were obtained using medical record linkage. We used general linear models to test for interactions between OCs (birth-weight, labour induction, Caesarean section, use of forceps, gestational age and neonatal care admission) and over 500,000 SNPs on measures of psychopathology (General Health Questionnaire (GHQ), Schizotypal Personality Questionnaire (SPQ), Mood Disorder Questionnaire (MDQ) and personality (Eysenck Extraversion and Introversion) as outcomes. Numbers for each OC analysis were based on the availability of medical records. All p-values reported are after Bonferroni correction. Genome-wide significant SNP-OC interactions were observed for the GHQ with neonatal care admission (N=1539; rs17141144; p=6.70×10-7; LOC107986773; intronic), use of forceps (N=1669; rs17065704; p=3.24×10-6; PEX7; intronic) and birthweight (N=2420; rs9608151; p=0.02; intergenic). GWAS-significant interactions were also found for the SPQ with neonatal care admission (N=932; rs12512245; p=0.02; intergenic) and birthweight (N=1536; rs7803908, p=0.005; TNS3; intronic) and for Eysenck Extraversion with birthweight (N=2629; rs7137811; p=0.007; intergenic) and gestational age (N=1541; rs17708877; p=0.01; DGKB; intronic). To our knowledge, this is the first genome-wide study to reveal GWAS-significant gene-by-environment interactions with OCs impact a wide variety of psychological traits associated with psychiatric disorders. As the OC data were obtained via medical record linkage, they are not likely to suffer from recall bias. Genes implicated include PEX7, involved in neuronal migration, and DGKB, highly expressed in the hippocampus and linked to cognition and schizophrenia.

Major Depressive Disorder (MDD) is one of the most common psychiatric disorders, with a prevalence of ~15%. Diagnosis is often inaccurate, based on self-report and is largely under-diagnosed. Data-driven approaches to predict lifetime risk for MDD and single versus recurrent MDD, especially using a small number of highly-accurate predictors that could be easily collected in clinic, would be a step forward in realising the promises of personalised medicine. We applied machine learning algorithms (MLAs) to the Generation Scotland cohort study (N > 21,000), a deeply-phenotyped cohort. The cohort was first divided into a training (63%) and held-out test (37%) set for both lifetime and single versus recurrent MDD analyses. In the training set, 10-fold cross-validation was used to estimate optimal hyperparameters for the following algorithms: Random Forest, Conditional Inference Forest, Gradient Descent Boosting (GBM), Support Vector Machines (with linear, polynomial and radial basis function kernels), Neural Networks, C5.0, Elastic Net and Forward Stepwise Regression. Using the optimal hyperparameters, we ran each MLA on the full training set, then ran the independent test data through the best model derived from the training data to predict case-control or single-recurrent status. Receiving Operator Characteristic Area Under the Curve (AUC) values were used to assess performance of the algorithms on the test data. To obtain a reduced set of predictors with optimised predictive value, the Markov Chain 4 (MC4) algorithm was used. MC4 is a rank aggregation algorithm, originally designed for meta-Internet search rankings. The aggregated rankings of predictors across all MLAs were then used to determine the smallest set with equal or better predictive value on the test data versus using all of the predictors. AUC values for the prediction of lifetime MDD were in the range of 0.81–0.84 for all MLAs; similarly, the AUC range for single versus recurrent MDD were between 0.69 and 0.76. All AUCs were significantly better than expected by chance (AUC = 0.5). Using the MC4 ranked variables and the best-performing model (GBM), we found that equal performance was obtained using only 20 variables for lifetime MDD (AUC = 0.84) compared to 155 variables in the full set, and 10 for single versus recurrent MDD (AUC = 0.76) compared to 180 variables in the full set. The MC4 ranked subsets performed equally well to the full subset, although it is likely other subsets exist that could perform similarly. The subsets identified for lifetime MDD included neuroticism, general psychological distress, age, income, family history of depression, living alone, sex, home ownership, smoking, education, pain status, mania and schizotypy. Age at MDD onset, neuroticism, psychological distress, measures of cognition, age, smoking and home ownership were the most relevant predictors for whether someone would have a single episode of MDD or recurrent episodes. These highly-predictive questionnaires and demographic information can be easily collected in clinic to assist in accurate diagnosis and preventative treatment for recurrent MDD.

Studies in psychiatric genetics have focused almost solely on single-loci and polygenic associations and have not included possible interactions. Nicodemus et al. (2014, JAMA Psychiatry) introduced a novel statistical model that incorporates single marker, polygenic and epistatic components to assess the association between SNPs in the ZNF804A pathway and cognitive performance in psychosis. Two-SNP interaction modelling was conducted to test for epistasis. This resulted in a regression model that contained the polygenic score and two two-SNP interaction terms, where the epistatic component explained more variation in cognition than the polygenic score. This model is still relatively simplistic in modelling the genetic architecture of complex traits; in particular, the epistatic component, as the model only allowed for pairwise interactions between SNPs. We sought to implement a more flexible specification of the epistatic component by the use of non-parametric rule sets via the algorithm C5.0. As a proof-of-principle we simulated multiple phenotypes using SNPs located on chromosome 19 explaining A) 30% polygenic variation B) 30% epistatic variation using 2-SNP models not included in the polygenic set, not in LD and under weak (β=0.09), intermediate (β=0.15) and strong (β=0.24) interaction strength and C) a combination containing both a polygenic and epistatic component. Five hundred replicates were simulated per condition. We applied C5.0 in an effort to detect these epistatic interactions. C5.0 is a non-parametric algorithm to generate decision tree-based rulesets. Because of its decision tree basis C5.0 only allows for the Boolean operator OR but does allow for non-binary predictors. Each possible combination from the top node to bottom node in the tree is a so-called ruleset and can be used as a predictor in a regression model. We found C5.0 is highly capable of detecting both epistatic SNPs in 100% of the 500 simulations of the strong interaction, 99.2% of the intermediate interaction and 19.2% of the weak interaction models. For the intermediate and weak interactions C5.0 detected only one of the two epistatic SNPs in 0.8% and 41.2% of the replicates. C5.0 was immune to detecting polygenic additivity at the single SNP level. In the polygenic setting of 30% variation explained by an additive effect, C5.0 detected in 11.4% simulations at least one ruleset with no SNP being included more than 13 times These results show that C5.0 is able to detect epistasis even in the presence of a phenotype containing strong-single loci and polygenic components. Further work includes creating an R package that embeds these different components through a penalised regression framework to combine all three types of genetic variation (single loci, polygenic components, epistatic components) to better reflect the underlying biology, and to apply this method to cognitive performance in Major Depressive Disorder.

Little is known about the prognostic role of fasting glucose after mechanical thrombectomy (MT).We investigated whether fasting glucose on the next day after MT was associated with long-term outcome in acute ischemic stroke patients according to diabetes.We retrospectively analyzed 181 consecutive patients with acute anterior circulation ischemic stroke who underwent MT in 2 comprehensive stroke centers in Poland. Glucose levels were evaluated on admission and on the next day after MT. Fasting hyperglycemia (FHG) was defined as the glucose level above 5.5 mmol/L. Unfavorable outcome was defined as modified Rankin scale (mRS) of 3-6 at day 90 from stroke onset.Patients with FHG had higher mRS at 3-month follow-up compared with those without FHG (3.71 ± 2.56 versus 1.87 ± 2.22, P < .001). In the subgroup analyses, FHG was associated with poor neurological outcome in the group without diabetes (3.74 ± 2.52 versus 1.81 ± 3.74, P < .001) but not with diabetes (3.64 ± 2.67 versus 2.30 ± 3.74, P= .11). Patients without diabetes who had FHG were older, had higher glucose on admission, higher prevalence of atrial fibrillation, cardioembolic stroke etiology and bleeding brain complications compared with the group with normal fasting glucose. After adjustment for potential confounders, fasting glucose (odds ratio [OR] 1.46; 95% CI 1.19-1.79, P < .001), age (OR 1.06; 95% CI 1.02-1.10, P = .001), successful reperfusion (OR 0.09; 95% CI 0.04-0.22, P < .001) and baseline NIHSS score (OR 1.18; 95% CI 1.08-1.29, P < .001) were predictors of mRS 3-6 at 3-month follow-up in the whole group. In the subgroup without diabetes, fasting glucose (OR 1.57; 95% CI 1.17-2.11, P = .002), age (OR 1.05; 95% CI 1.01-1.08, P = .008), successful reperfusion (OR 0.11; 95% CI 0.04-0.30, P < .001) and baseline NIHSS score (OR 1.14; 95% CI 1.04-1.26, P = .011) were independent predictors of unfavorable 3-month outcome.Fasting glucose on the next day after MT in patients with acute ischemic stroke is an independent risk factor for worse 3-month outcome.