Right now we calculate maf for every variant by looking for any of maf, af, ac, mac and num_samples. Should we calculate mac in the same way, or only allow assoc_min_mac if there's a maf column?
We already support a r2 field. Just add a threshold for it, and explain more clearly in the README how to tell pheweb about header names ("aliases").
Should we allow variant_min_mac or variant_min_r2? I don't see a need, but it's odd to have variant_min_maf without those. Perhaps remove variant_min_maf? People initially supported that feature, but I feel like they'd be fine without it.
pjvandehaar
commented
3 years ago
Right now we calculate
maffor every variant by looking for any ofmaf,af,ac,macandnum_samples. Should we calculatemacin the same way, or only allowassoc_min_macif there's a maf column?We already support a
r2field. Just add a threshold for it, and explain more clearly in the README how to tell pheweb about header names ("aliases").Should we allow
variant_min_macorvariant_min_r2? I don't see a need, but it's odd to havevariant_min_mafwithout those. Perhaps removevariant_min_maf? People initially supported that feature, but I feel like they'd be fine without it.