Add get_loftee_end_trunc_filter_expr and update_loftee_end_trunc_filter to patch the LOFTEE END_TRUNC filter
#740
Conversation
|
|
||
| :param csq_expr: StructExpression containing the LOFTEE annotation 'lof_info', with | ||
| 'GERP_DIST' and '50_BP_RULE' info. | ||
| :param gerp_dist_cutoff: GERP distance cutoff for end truncation. Default is 0.0. |
There was a problem hiding this comment.
Thanks for making this notebook! So the gerp_dist_cutoff is changing from -58 to 0 and some "HC" became "LC", is there a justification for this cutoff that we could document?
I summarized our discussion with Konrad last year to answer the questions in the emails and on the forum.
There was a problem hiding this comment.
I only checked variants in BRCA2 gene, the last 2 LOF now have "END_TRUNC" filter and became "LC".
There was a problem hiding this comment.
Yup, I'm hoping Julia can help me add to the documentation to justify the cutoff, and to also confirm that the patch is being done correctly.
| :return: Consequence Struct with updated LOFTEE annotations. | ||
| """ | ||
| end_trunc_expr = get_loftee_end_trunc_filter_expr(csq_expr, gerp_dist_cutoff) | ||
| filter_expr = hl.or_else( |
There was a problem hiding this comment.
Do you use hl.set here and hl.delimit below because there might be multiple lof_filter per transcript? I saw in your test, after explode, there seemed to be only one.
There was a problem hiding this comment.
The lof_filter is a list of filters represented as a string delimited by ,, for example: "5UTR_SPLICE,ANC_ALLELE,GC_TO_GT_DONOR"
If end_trunc_expr is True, I add END_TRUNC to the filters, but it might already be in the filters and I only want it represented once.
Tested in this notebook
test_loftee_end_trunc.html.zip