From 845dfd26a185a82e04388df0e50227b078e29656 Mon Sep 17 00:00:00 2001 From: hanna julienne <hanna.julienne@pasteur.fr> Date: Wed, 30 Oct 2019 12:02:09 +0100 Subject: [PATCH] filter ambiguous SNPs --- jass_preprocessing/map_reference.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/jass_preprocessing/map_reference.py b/jass_preprocessing/map_reference.py index 1b0f395..c630149 100644 --- a/jass_preprocessing/map_reference.py +++ b/jass_preprocessing/map_reference.py @@ -23,6 +23,9 @@ def read_reference(gwas_reference_panel, mask_MHC=False, minimum_MAF=None, regio ref = pd.read_csv(gwas_reference_panel, header=None, sep= "\t", names =['chr', "pos", "snp_id", "ref", "alt", "MAF"], index_col="snp_id") + #Filter Strand ambiguous + ref = ref.loc[~(ref.ref+ref.alt).isin(["AT", "TA", 'CG','GC'])] + if mask_MHC: ref = ref.loc[(ref.chr !=6)|(ref.pos < 28477797)|(ref.pos > 33448354)] -- GitLab