Question on clinvar annotation of sgRNA guides

[MGordon09]

Hi all, thanks for developing this very useful tool.

I have a question on how the clinvar variant significance annotation table is matched to the generated guides. I've attached a subset of my output files below for a single guide (clinvar_annotations_test.txt). The guide is antisense and targeted to gene on -1 strand.

Looking at the clinvar file, this guide recovers 7 clinvar records for the two possible edits in edit window(A_5 (n=2) and A_7 (n=5)), with 3 unique Clinvar SNP clinical significance values reported for mutations in this window. In the sgrna_designs output, this guide is collapsed to two Clinvar SNP clinical significance values; 'Likely benign' for A_5 and 'None' for A_7 (assuming order is maintained acrossAmino acid edits, Mutation category, Clinical significance columns in sgrna_designs output).

My understanding is both these edits should be assigned 'None' in Clinvar SNP clinical significance sgrna_designs output, as none of the clinvar SNP edits match the edits we are introducing when comparing the sgRNA amino acid change and SNP amino acid change columns in the clinvar_annotations.

I would appreciate a little more insight into the merging for the Clinvar and sgRNA guide tables is performed, and how signficance terms are ranked and selected.

Thanks for your time.

python /wynton/group/krogan/mgordon/projects/072224_RBabu_CRISPReGuideDesign/script/base-editor-design-tool/base_editing_guide_designs.py \
    --input-file /xxx/guideInput.txt \
    --variant-file /xxx/docs/variant_summary.txt \
    --input-type tid \
    --pam NG \
    --edit-window 4-8 \
    --edit A-G \
    --sg-len 20 \
    --filter-gc True \
    --output-name 072424.Abe8e.guides

sgrna_designs_test.txt

[MGordon09]

Attaching relevant files here:

clinvar_annotations_test.txt sgrna_designs_test.txt