Computational protocol: A Combined Pathway and Regional Heritability Analysis Indicates NETRIN1 Pathway Is Associated With Major Depressive Disorder

Similar protocols

Protocol publication

[…] This study included 21,387 subjects (8772 men, 12,615 women; mean age, 47.2 years). Participants were recruited from the registers of collaborating general practices by their Community Health Index (). A Structured Clinical Interview for DSM-IV was used for the diagnosis of MDD mood disorders () (). By the time we performed this study, 9863 individuals were genotyped using the Illumina (San Diego, CA) Human OmniExpressExome-8- v1.0 array (). Details of genotyping are described in detail elsewhere (). Quality control (QC) and imputation method are described in the . In total, 592,690 genotyped and 2,163,848 imputed autosomal SNPs passed QC criteria and were used in subsequent analyses. Because close relatives can bias the pathway analysis and SNP heritability estimation, the function “--grm-cutoff 0.025” in GCTA was used to remove one of each pair of individuals with estimated relatedness larger than 0.025 while maximizing the remaining sample size (); 6455 subjects (1123 MDD case subjects; 5332 control subjects) remained in the analyses described below. [...] For both samples, SNPs were annotated to 1035 pathways (640 from Reactome [], 216 from BioCarta [], and 179 from Kyoto Encyclopedia of Genes and Genomes []) (). For the GS:SFHS genotype data set (nSNP = 592,690; nsample = 6455), the GRASS (gene set ridge regression in association studies) algorithm () was used to identify pathway MDD associations using only the genotyped SNPs (). False discovery rate (FDR)-adjusted p values (nFDR = 1035) were calculated with the function p.adjust in the R package “stats” (, ). For the PGC:MDD GWAS summary results data set (nSNP = 1,074,100), MAGENTA (meta-analysis gene-set enrichment of variant associations () was used to test for the enrichment of genetic associations in each pathway for MDD, because only summary statistics were available for this part of the study (, and MAGENTA was designed to exploit summary data from GWAS results (). MAGENTA reports a nominal p value and an estimated FDR per pathway (nFDR = 1035). […]

Pipeline specifications

Software tools GCTA, GRASS, MAGENTA
Databases Reactome KEGG BioCarta
Application GWAS