|
From: Sarah S. <ss...@us...> - 2018-02-27 21:17:44
|
Hello all, I am using a vcf file that I did not produce, it was provided as supplementary material from a manuscript. It is old, produced most likely with versions of all of this software from around 2012. I am trying to merge the VCFs as the authors provided them split by genotype, and I keep getting this error: 'Could not determine the ploidy (nals=1, nvals=3). If using vcftools version 1.12, and this error: Use of uninitialized value within @gtypes in hash element at /Users/sarah/Downloads/vcftools_0.1.10/perl//Vcf.pm line 1753, <__ANONIO__> line 287. If using 1.10. And indeed, vcf validator has the following to say: FORMAT tag [PL] expected different number of values (expected 1, found 3) The most recent version of vcftools errors as follows: Wrong number of values in 23/PL at YHet:102954 .. nAlleles=1, nValues=3. Expected 1 values for diploid genotypes or 1 for haploid genotypes. I"ll admit I was using legacy versions of VCFtools to try and avoid any difficulties relating to changes in file format, but clearly its not working. Did VCF files used to have a different number of PL fields? Is there something deeply wrong with this VCF file? Is this some kind of bug with VCFtools? I realize I might be asking a question about VCF files in general rather than VCFtools in particular, but any insight would really be appreciated. Thanks, Sarah |