topherhuebel/ukbgwas

Scripts for body composition GWAS in UK Biobank

ukbgwas

Scripts for body composition GWAS in UK Biobank

1. Folder structure
2. Phenotype

• Extract columns from UKBB files (awk)
• Exclusion [phenotype_an_free.R]
• Descriptives
• Covariates
• Regression [regression_linear.R]
• Transfer descriptives into doc
• Make scatterplots [plotting_descriptives_healthy.R]
• Make phenotypic correlation matrix
• Calculate independent tests
• Make correlation heatmap [full_correlation_alphabetical_healthy.R]

3. GWAS

• per chromosome [gwas_chr.sh]

4. Post-GWAS filtering [post_gwas_filtering/]

• Remove SNP duplicates [concatenate_filtering.sh]
• Retain HRC SNPs
• Cocatenate to one sumstats file

4. Post-GWAS filtering 2 [post_gwas_filtering]

• MAF 0.01 [maf_0.01.sh]

5. Meta-analyse

• Metal [bfpc_h_metal_se.script & metal_jobscript_bfpc.sh]

6. Make sex-specific and metal files [sex_specific_sumstats_files/]
   • [sex_sumstats.sh]
   • [metal_sumstats.sh]
   • [add_pos_to_meta_sumstats.sh]
7. Extract gws SNPs [post_gwas_filtering/]

• Genome-wide significant SNPs [sign_snps.sh]
• gzip
• Put in spreadsheet

5. Plots [plots]

• qqplot
• sex-specific qqplot
• Manhattan plot
• Miami plot

6. Clumping

• FUMA

7. Annotate

• FUMA

8. LDSC [ldsc]

• Munge
• Calculate h2
• Put inflation statistics into table
• run rgs

9. Part heritability [ldsc_part_h2]

• submit script