Issue Feedback on HGVS Nomenclature in Annotation Results

[PickingBook]

Issue Description

The following three variants were annotated using our pipeline:

chr1-145318950-C-G → NM_001039703.6:c.2998G>G (p.Glu1000=)

chr1-145316672-G-T → NM_001039703.6:c.2746T>T (p.Phe916=)

chr11-124085462-G-A → NM_001395906.1:c.1A>A (p.Met1=)

Input VCF format:

1 145318950 . C G . . .

All three descriptions show redundant DNA change syntax (>same base), which is invalid according to HGVS rules. The correct representation should be c.X= instead of c.X>X, indicating a DNA variant that does not change the encoded amino acid.

Our annotation pipeline produced these unusual notations, possibly due to normalization or transcript alignment issues. However, the same variants annotated with the web-based tool Ensembl VEP gave identical results, confirming that our local server is functioning correctly. Please help troubleshoot the root cause.

System

• VEP version: 115.2
• VEP Cache version: 115
• Perl version: 5.26

[ainefairbrother]

Hi @PickingBook,

This is occurring due to a GRCh37/RefSeq mismatch. On CLI VEP, you can apply the --use_given_ref option to use the input reference base, however, this option is not available on web VEP.

Hope this helps,

@ainefairbrother