Page 1 of 1

coding of "half deletions"

Posted: Wed Jun 29, 2016 11:34 am
by CRW

I am trying to load canola hybrid SNP data in GenABEL. Since I created the hybrids in silico from the Illumina genotyping files of the parents, some hybrids can contain "half deletions" (e.g. A-).
According to the GenABEL manual

"Accepted allele codes: 1/2, A/B, A/T, A/G, A/C, T/G, T/C, G/C, A/-, T/-, G/-, C/-. Here, "-" stands of a deletion. Missing data can be coded as "–" or "00". Make sure that the coding for missing is "00" if you use one of the codings A/-, T/-, G/-, C/-"

I tried to code my half deletions like this (in this short testfile below):

ID chrom snp_pos G1 G2 G3
M1 1 36516 T- TG TG
M2 1 50056 AT TA TA

which gives me the following error:
"error in in convert.snp.illumina(inf = "Genabel_CoreSet_test2.pop", out = "Genabel_CoreSet_test2.raw", :
illegal genotype (half missing) SNP 'M1' file 'Genabel_CoreSet_test2.pop' line 0 !"

I tried several other codings for the matrix, including T/-, -T, -/T, T0, but none of them is able to handle my "half deletions". However, it works as soon as I replace the first SNP genotype by TT, so the rest of my file seems to be correct.

Does anyone know how to solve this problem?

Many thanks in advance!