Discrepancy between genotype and reference genome build - issue with strand flips

Question

[dam4l]

I have genotypes that are aligned with Hg38. I am hoping to impute them with HRC (Hg37) as the reference panel via the Michigan Imputation Server. However, there are significant strand flips and I am getting this error message:

Error: More than 100 obvious strand flips have been detected. Please check strand. Imputation cannot be started!

I have avoided this issue in the past by aligning my genotype file with the reference panel using scripts from here: https://www.well.ox.ac.uk/~wrayner/tools/.

How can I fix these strand flips when my genotype file genome build differs from the reference panel genome build? All input/suggestions appreciated. Thanks in advance.

Answer 1

[Nicolas Rosewick]

check here : https://imputationserver.readthedocs.io/en/latest/prepare-your-data/

And more precisely here : https://www.well.ox.ac.uk/~wrayner/tools/

[dam4l]

Yeah, this is what I have used previously when my genotype file was aligned with Hg37. It does not seem appropriate given that the genome builds differ between by genotype file and HRC this time around. Please let me know if I am missing something.

[Nicolas Rosewick]

Indeed HRC is only in hg19 ( in my knowledge ). Maybe you could try to liftover your data, then perform pre-imputation data preparation with McCarthy tool, then impute with HRC.

Other idea : impute with TOPMed. Pre-imputaion data preparation can be done on hg38 : https://imputation.biodatacatalyst.nhlbi.nih.gov/