[ensembl-dev] Clarifying VEP outputs for inframe insertions

[John M.C. Ma]

Hi,

I'm reviewing my VEP output from VEP 84 for parsing scripts. I'm
confused regarding the inconsistent annotations for inframe
insertions.

I have encountered, in the output from a single VEP run, two ways how
VEP notates protein-level changes. For your convenience, the fields
names used here are the same ones used in CSQ outputs.

1. The first type would have only a single number in the
Protein_position field, and the Animo_acids field would include the
residue immediately before the insertion. For example,
Protein_position = 114, Animo_acids=A/AILH.

2. The second type includes two numbers in the Protein_position field,
and in the Anime_acids field the ref AA is marked as -. For example,
Protein_position=271-272, Animo_acids=-/APTP.

While I have an ides about how to parse these two types, I'm puzzled
about this inconsistency.

For your reference, I was running with local e84 GRCm38 cache with
fasta file with VEP 84 and perl 5.22.1 in an Linux environment.

Thanks for the assistance.

Cheers,

John