Intersection-between-UK-Biobank-and-Illumina-GSA-arrays.md

@@ -78,3 +78,21 @@ cut -f 6 compare_alleles.txt  | sort | uniq -c
 | UKB | Both | 279300 / 34.0% | 267308 / 32.4% |
 | GSA | Unique | 435910 / 60.9% | 450562 / 62.8% |
 | GSA | Both | 279300 / 39.1% | 267308 / 37.2% |
+
+### Extract the rsids and get the MAFs
+
+**NB** Could try mapping by position and/or alleles again to the MAF files for any GSA-only sites that are imputed in UKB.
+
+```
+grep -v gsa_only compare_alleles.txt | cut -f 5 | sed -e 's/,/\n/g' > rsids.ukb_all.txt
+grep ukb_only compare_alleles.txt | cut -f 5 | sed -e 's/,/\n/g' > rsids.ukb_only.txt
+grep both compare_alleles.txt | cut -f 5 | sed -e 's/,/\n/g' > rsids.both.txt
+wc -l rsids.*
+  269399 rsids.both.txt
+  845485 rsids.ukb_all.txt
+  576086 rsids.ukb_only.txt
+
+cat UKB_MAFs/* | perl scripts/extract_UKB_mafs_by_rsid.pl rsids.ukb_all.txt > mafs.ukb_all.txt
+cat UKB_MAFs/* | perl scripts/extract_UKB_mafs_by_rsid.pl rsids.ukb_only.txt > mafs.ukb_only.txt
+cat UKB_MAFs/* | perl scripts/extract_UKB_mafs_by_rsid.pl rsids.both.txt > mafs.both.txt
+```
\ No newline at end of file