inst/markdown/methods/germline-gwas.md

In CRI-iAtlas/iatlas-app: What the Package Does (One Line, Title Case)

Title: Germline GWAS

Descriptions: We selected each of the immune phenotypes that demonstrated nominally significant genome-wide heritability (N=33) for GWAS. GWAS was conducted on all of the genotyped SNPs that passed QC and all of the imputed SNPs that had imputation R2 > 0.5 and minor allele frequency > 0.005 in the 9,603 unrelated individuals (PLINK 1.9 identity by descent, IBD, pihat < 0.25). Minor allele frequencies were recalculated post-imputation for only the subset of 9,603 individuals (PLINK 1.9). Of the 39,127,678 SNPs available after imputation, 10,955,441 passed both imputation quality and frequency thresholds and thus were included in the association analysis. GWAS was performed using PLINK 1.9. Immune phenotypes that were approximately normally distributed or normally distributed after stratification by covariates were tested for association with SNPs using linear regression with age, tumor type, sex and PC1-7 as covariates. Immune traits that were dichotomized for heritability analyses were analyzed using logistic regression models, with the same covariates. For each GWAS we also calculated the genome-wide inflation coefficient (lambda). We used the traditional cutoff of p < 5x10-8 as a cutoff for genome-wide significance and p < 1x10-6 to denote suggestive loci. Since we only selected the subset of phenotypes that was heritable and since many of the phenotypes were highly correlated, we did not correct the GWAS for the number of phenotypes analyzed. SNPs were annotated based on spanned genomic ranges (R v3.5.0, Bioconductor package GenomicRanges_1.32.6) with rsIDs (R v3.5.0, R package snplist_0.18.1, Bioconductor package SNPlocs.Hsapiens.dbSNP144.GRCh37_0.99.20) and with genes within +/- 50KB, +/- 500KB, and +/- 1MB of the SNP (R v3.5.0, Bioconductor package biomaRt_2.36.1 using grch37.ensembl.org as host). Variant annotations for all genome-wide and suggestive SNPs were determined using the web interface of the Ensembl Variant Effect Predictor (VEP, https://grch37.ensembl.org/info/docs/tools/vep/index.html. All annotations were based on Homo sapiens (human) genome assembly GRCh37 (hg19) from Genome Reference Consortium. All association statistics for the GWAS are available at Figshare https://figshare.com/articles/dataset/Sayaman_et_al_TCGA_Germline-Immune_GWAS_Summary_Statistics/13077920.

Reference Listing

Contributors: Rosalyn Sayaman, Elad Ziv

CRI-iAtlas/iatlas-app documentation built on Feb. 7, 2025, 9:02 p.m.