12_ldsc_munge.bsub

#!/bin/bash
#BSUB -q long
#BSUB -J ldsc_munge[1-15]
#BSUB -n 4 
#BSUB -R "span[ptile=4]"
#BSUB -R "rusage[mem=8GB]"
#BSUB -o /your/out/dir/ldsc_munge_%J_%I.out
#BSUB -e /your/err/dir/ldsc_munge_%J_%I.err

#This script creates ldsc_munge:sumstats output

#Load modules
module load anaconda3/2021.05

#set env
source /.bashrc
conda activate /./.conda/envs/ldsc
#phenotypes
pheno1=Cell_B
pheno2=Cell_P
pheno3=CLL
pheno4=DLBCL
pheno5=Drug_G1
pheno6=FL
pheno7=HL
pheno8=LM
pheno9=LPL_WM
pheno10=MGUS
pheno11=MM_MGUS
pheno12=MM
pheno13=MZL
pheno14=Soma_G1
pheno15=Soma_G2


#paths for sumstat files
path0=/your/dir/sumstats
path1=${path0}${pheno1}
path2=${path0}${pheno2}
path3=${path0}${pheno3}
path4=${path0}${pheno4}
path5=${path0}${pheno5}
path6=${path0}${pheno6}
path7=${path0}${pheno7}
path8=${path0}${pheno8}
path9=${path0}${pheno9}
path10=${path0}${pheno10}
path11=${path0}${pheno11}
path12=${path0}${pheno12}
path13=${path0}${pheno13}
path14=${path0}${pheno14}
path15=${path0}${pheno15}

# array job var
combined1=path${LSB_JOBINDEX}
combined2=pheno${LSB_JOBINDEX}

cd ${!combined1}




# example input: Cell_B_ldsc.ldsc.tsv.gz
# col.names: CHR     POS     A1      A2      Frq     Beta    Z       P       OR      INFO    N       SNP

############### munge_sumstats ############################

python /./ldsc/munge_sumstats.py \
--sumstats ${!combined2}_ldsc.ldsc.tsv.gz \
--chunksize 500000 \
--merge-alleles /./ldsc/w_hm3.noMHC.snplist \
--maf-min 0.05 \
--frq Frq \
--snp SNP \
--N-col N \
--a1 A1 \
--a2 A2 \
--p P \
--signed-sumstats Z,0 \
--ignore "Beta,OR" \
--out ${!combined1}/${!combined2}_maf0p005_noMHC_munge.sumstats