2020-04-08

027_ Collaboration of PsyCourse with the International Suicide Genetics Consortium

Research Question and Aims

A very large proportion of people who attempt suicide have a psychiatric diagnosis. Population-based and family studies have shown a substantial heritability of suicide behavior, with estimates from 17 to 55% (Voracek et al., 2007; Sokolowski et al., 2014). A genetic correlation between suicide attempt and several psychiatric/personality traits has recently been reported (Ruderfer et al. 2019). Moreover, a recent GWAS on suicide attempt has started to shed light on the genetic underpinnings of suicidal behavior in psychiatric populations (Mullins et al. 2019).
Here we propose to contribute to the upcoming GWAS on suicidal attempt from the International Suicide Genetics Consortium. We will leverage the fine-grained information available in PsyCourse regarding suicidiology. Briefly, GWAS analyses based on PsyCourse suicide variables will be run locally at the LRZ servers and only summary statistics will be shared with the International Suicide Genetics Consortium in order to perform a meta-analysis. Sergi Papiol will also participate in such meta-analyses as analyst.

Analytic Plan

Main hypothesis: common genetic variants contribute to the risk of suicide attempt/completion in psychiatric patients.

Participants: data from all samples of patients and healthy controls in PsyCourse who have genotype/suicide phenotype data available will be included in this study.

Statistical analysis: local analyses will involve a genome-wide approach (GWAS) based on imputed variants that will be carried out using PLINK1.90. Population stratification will be corrected for using ancestry principal components. Likewise, center of recruitment will be used as covariate. Age and sex are not mandatory according to the guidelines of the International Suicide Genetics Consortium but they will also be used to compare the results with or without them.

Several models will be used:
Model 1 (within-case) - suicide attempters with a psychiatric disorder vs. non-attempters with the same psychiatric disorder as controls.
Model 2 (case vs all) - suicide attempters vs. all others (the controls in this model could be population-based and may include non-attempters with psychiatric disorders)
Model 3 (case vs healthy controls) - suicide attempters vs. non-psychiatric controls (these controls should have some level of screening for the absence of psychiatric disorders and should not include known cases of psychiatric disorders)

The resulting summary statistics will be shared with the International Suicide Genetics Consortium.

Resources needed

Phenotype:
Diagnosis/status (v1_scid_dsm_dx_cat)
Suicide attempt (categorical [1,2,3], v1_suic_attmpt)
Number of suicide attempts (ordinal [1,2,3,4,5,6], v1_scid_no_suic_attmpt)
Age at first interview (continuous [years], v1_age)
Sex (dichotomous [M,F], v1_sex)
Recruitment center variable (categorical [see below], v1_center)

Genotype:
Raw genotypes pre-imputation to calculate PCAs
Imputed genotypes