Annotate Variant List

The script annotate_variant_list.py can be used to match variants from input files to an instance of GraphKB. Copy the script locally and install the package dependencies.

This example script expects a file with a single field (variant) and no header. Each line should be a separate HGVS-like variant notation. For example

KRAS:p.G12D
KRAS:p.G13E

Must use Python3.6 or higher

pip3 install graphkb pandas

Then the annotator can be run as follows

python annotate_variant_list.py <INPUT FILE> --output graphkb_annotations.tsv

By default this will use the pori-demo version of GraphKB which has a limited amount of data. This demo version is intended for demonstration/testing only and a custom or shared production instance of the GraphKB API. This can be configured via the GraphKB arguments. See the help menu for a full list of arguments.

python annotate_variant_list.py -h

The output file will contain the variant name and the annotations pulled from GraphKB.

The names of the variants matched will be included in the output file as "variant_matches", this will be a semi-colon delimited list of all the variants which were considered to be present/equivalent based on the input variant. For example if the input where KRAS:p.G12D we might expect to see something like this

KRAS mutation;KRAS:p.(G12_G13)mut;KRAS:p.?12mut;KRAS:p.G12;KRAS:p.G12D;KRAS:p.G12mut;chr12:g.25398284C>T

We can see the variant has matched less specific forms of the same variant such as KRAS mutation or KRAS:p.(G12_G13)mut (any KRAS mutation at G12 or G13)