I am trying to use selscan to calculate EHH scores. I have 1000 genomes vcf files which I have used to produce map files with vcftools, such that they look like this:
22 rs8142737 50291889 50291889
22 rs570182536 50291936 50291936
22 rs8135816 50291976 50291976
22 rs8140681 50292081 50292081
22 rs9627785 50292178 50292178
22 rs9616779 50292545 50292545
22 rs9616780 50292763 50292763
22 rs139397353 50292931 50292931
22 rs9616364 50292983 50292983
22 rs12159367 50293281 50293281
22 rs7290342 50294176 50294176
22 rs141187212 50294325 50294325
22 rs6520063 50294378 50294378
22 rs6520064 50294469 50294469
My selscan command is:
selscan --ehh 50292931 --vcf chr22.vcf.gz --map plink22.map --maf 0.0001 --out test.txt
This produces the following error:
ERROR: Variant physical position must be strictly increasing.
-- -9999 comes after -- -9999
Now I have tried to identify the problem row by trying to grep for "-9999" but get nothing. I have also tried to sort on the physical position column but get the same error. There are no blank rows at the start or end of the file.
To ensure there was no issue with my map file, I tried using different chromosomes but keep getting this error. Can anyone shed some light on this?
By the way, I have also tried using hapbin with the same files using the following command:
ehhbin --locus 50292931 --hap out.impute.hap --map <(awk '{$3=$4;print}' plink22.map)
But I always get an error:
no locus with the id: 50292931
I have checked and the locus is definitely within the .map and the .hap files (which were created from the vcf files). Therefore I think the problem must be within my map files but I cannot fathom what the issue is.
