Converting Minimac Dosage Files To Plink Data, Preserving Probabilities
I've got a minimac.dose file which contains probabilities for the imputed SNPs, looks like this:
11464_ATCACG DOSE 2.000 1.985 1.987 1.882 1.902 ...
11465_CGATGT DOSE 2.000 1.981 1.982 1.883 1.907 ...
I'd like to convert it into a raw plink file which also contains probabilities.
If you know any way to convert minimac dosage files into raw probability files please do share.
---WHAT I'VE TRIED---
I've been able to convert it into raw plink format with the commands:
gcta64 --dosage-mach ${MINIMAC_OUTPUT}.dose ${MINIMAC_OUTPUT}.info --make-bed --out ${GCTA_OUTPUT}
plink --noweb --bfile ${GCTA_OUTPUT} --recodeA --out ${RAW_IMPUTED_DATA}
Unfortunately this produces files like this (the dosages are now discrete values, not probabilities - not what I want):
11471_ACTTGA 11471_ACTTGA 0 0 -9 -9 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
11472_GATCAG 11472_GATCAG 0 0 -9 -9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0
Bonus points if anyone can tell me whether it is gcta or plink that rounds off the values.
• 824 views
•
link
0 answers
No answers yet.
Log in to answer this question.
Could you tell us the why, what are you going to do once converted? Regarding "bonus point", just open
${GCTA_OUTPUT}and${RAW_IMPUTED_DATA}files and have a look. Btw, plink can read dosage files.Thanks for the reply. I want to find the correlation between imputed and real SNPs. Problem is, IIRC, that the .dose files contains twice as many columns as the reference VCF (converted to raw data). Hence, I do not know how to compare them.
This Computing R Squared Statistics For Imputation Results - How? to your previous Computing R Squared Statistics For Imputation Results - How? doesn't work?
Is a minimac dose file the same as a plink dosage file?