filtering 1kg data to hm3 and recoding snp names

filter_1kg.sh

#!/bin/bash

module add apps/plink-1.90b3v

cd /panfs/panasas01/shared/alspac/deprecated/gib/data_for_suzi

cp /panfs/panasas01/shared/alspac/deprecated/alspac_combined_1kg_20140424/combined/alspac_1kg_p1v3_maf0.01_info0.8.bed als.bed
cp /panfs/panasas01/shared/alspac/deprecated/alspac_combined_1kg_20140424/combined/alspac_1kg_p1v3_maf0.01_info0.8.bim als.bim
cp /panfs/panasas01/shared/alspac/deprecated/alspac_combined_1kg_20140424/combined/alspac_1kg_p1v3_maf0.01_info0.8.fam als.fam

# Recode snp ids into chr_pos format
cp als.bim als.bim.orig
awk '{ print $1, $1"_"$4, $3, $4, $5, $6 }' als.bim.orig > als.bim

# Remove any duplicate SNP IDs

cp als.bim als.bim.orig2
awk '{
if (++dup[$2] > 1) { 
	print $1, $2".duplicate."dup[$2], $3, $4, $5, $6 
} else { 
	print $1,$2,$3,$4,$5,$6 }
}' als.bim.orig2 > als.bim

grep "duplicate" als.bim | awk '{ print $2 }' > duplicates.txt

plink1.90 --bfile als --exclude duplicates.txt --make-bed --out als_clean
plink1.90 --bfile als_clean --extract hm3.snplist --make-bed --out als_clean_hm3