Once METASOFT and/or RE2C have run, we currently match on the coordinate position of each SNP, but we leave out the reference and alternate allele.
This should be included as a priority. It should be easy to include those columns from the VCF when adding the coordinates.
Stretch goal: we should do a sanity check that each cohort has the same reference an alternate allele for each SNP - if a cohort has them the other way round then potentially the direction of effect (beta) would be flipped. This would hurt the meta-analysis considerably.
Once METASOFT and/or RE2C have run, we currently match on the coordinate position of each SNP, but we leave out the reference and alternate allele.
This should be included as a priority. It should be easy to include those columns from the VCF when adding the coordinates.
Stretch goal: we should do a sanity check that each cohort has the same reference an alternate allele for each SNP - if a cohort has them the other way round then potentially the direction of effect (beta) would be flipped. This would hurt the meta-analysis considerably.