TelentiLab / PrimateAI-Biothings-Parser

This is the data parser for PrimateAI data.
0 stars 0 forks source link

Same position may map to different scores #1

Open CristianoYL opened 5 years ago

CristianoYL commented 5 years ago

Noticed an issue within the score file where same positions may map to different scores with different refAA (reference amino acid) and altAA.

For example:

#chr    pos ref alt refAA   altAA   strand_1pos_0neg    trinucleotide_context   UCSC_gene   ExAC_coverage   primateDL_score
chr1    205492739   C   T   P   L   1   CCG uc010pri.2  31.14   0.447502940893
chr1    205492739   C   T   R   C   1   CCG uc001hcr.3  31.14   0.555931210518
CristianoYL commented 5 years ago

Explanation from Kyle:

It’s because in some cases the variant is flipped in the reference with respect to the derived allele. In cases where the derived allele was not the alt allele, we scored the changes in both directions. Almost all of these are neutral human/chimpanzee changes.