config.yaml

# General configuration
# ---------------------
output_dir: 'results'

# Genotype position definition
genodata: 
  json: "/scratch/mfilosi/test_gwas_regenie/genetic_data.json"
  name: "HRC13K"

# Sample file to use
# ------------------
# See https://www.cog-genomics.org/plink/1.9/filter#indiv
# for details on the sample file
sample_file: "/scratch/mfilosi/test_gwas_regenie/metaboGWAS/sample.samplist"

# Phenotype file used in the GWAS
# -------------------------------
# tab separated.
# First two colums should be FID and IID
pheno_file: "/scratch/mfilosi/metaboGWAS/input_files/phenotype_filt_winsorized_scaled.txt"
run_list: "/scratch/mfilosi/test_gwas_regenie/finemap_pipeline/pheno_to_run.csv" 

# Clumping
# --------
# NB: logp1 and logp2 will only work with plink2
clumping:
  # logp1: 10.769551078621726
  logp1: 7.3
  logp2: 1.3010299956639813
  r2: 0.1
  kb: 10000
  p1: 1.7e-11
  p2: 0.05
  totsize: 1e6


# Summary Statistic
# -----------------
# Define where the summary stat results will be found
# which is the pvalue column and the threshold to use to define
# putative credible sets
# Option log: True not implemented for now
sumstat:
  resdir: "/scratch/mfilosi/test_gwas_regenie/metaboGWAS"
  pvalcol: "LOG10P"
  log: True
  pthr: 1.7e-11
  tophits_dir: "/scratch/mfilosi/test_gwas_regenie/metaboGWAS/tophits"
  annotate: False

# SusieR parameters
# -----------------
susieR:
  # The following parameter will enable the use of correlation matrix based
  # on LD as specified in [https://stephenslab.github.io/susieR/articles/finemapping_summary_statistics.html](https://stephenslab.github.io/susieR/articles/finemapping_summary_statistics.html)
  # If set to False (default), it will use the genotypes coded with additive model
  # together with the phenotype to evaluate the RSS model.
  use_ld: False
  # When using this pipeline on CHRIS samples, the IDs
  # have leading zeros, and will have a total length of 10 characters.
  # Thus within the `scripts/finemapping.R` will do a conversion
  # with for zero padding of the IDs to match the ones in the genotypes.
  # Set this value to `False` for remove 0 padding to 10 character.
  chris_id: True
  min_abs_corr: 0.1
  iter: 1000