Closed nettam closed 2 weeks ago
In line 51 we have: df_regions_keep = df_regions.apply(lambda r: np.sum((df_sumstats['CHR']==r['CHR']) & (df_sumstats['BP'].between(r['START'], r['END']))) > 1, axis=1)
Shouldn't this be: df_regions_keep = df_regions.apply(lambda r: np.sum((df_sumstats['CHR']==r['CHR']) & (df_sumstats['BP'].between(r['START'], r['END']))) >= 1, axis=1) ?
couldn't this single-snps be causal? I see that they will be dropped in finemap.py as well
Found the reason
In line 51 we have: df_regions_keep = df_regions.apply(lambda r: np.sum((df_sumstats['CHR']==r['CHR']) & (df_sumstats['BP'].between(r['START'], r['END']))) > 1, axis=1)
Shouldn't this be: df_regions_keep = df_regions.apply(lambda r: np.sum((df_sumstats['CHR']==r['CHR']) & (df_sumstats['BP'].between(r['START'], r['END']))) >= 1, axis=1) ?