Modify default_compute_info with the option to use the AS_ annotations in gvcf_info for allele specific aggregations
#560
Conversation
… into jg/change_compute_info_to_use_AS_ann # Conflicts: # gnomad/utils/annotations.py
gnomad/utils/sparse_mt.py
Outdated
| AS_INFO_SUM_AGG_FIELDS = ["AS_QUALapprox", "AS_RAW_MQ"] | ||
| AS_INFO_INT32_SUM_AGG_FIELDS = ["AS_VarDP"] | ||
| AS_INFO_MEDIAN_AGG_FIELDS = ["AS_RAW_ReadPosRankSum", "AS_RAW_MQRankSum"] | ||
| AS_INFO_ARRAY_SUM_AGG_FIELDS = ["AS_SB_TABLE"] |
There was a problem hiding this comment.
Just an idea, should we nest these in a dict called AS_INFO_FIELDS or something -- maybe a better name than that? I'm thinking of the utility of this module and using tab when writing code to see available variables and functions within this module. Theres just a lot of high level objects.
There was a problem hiding this comment.
Should I change the non AS ones to that also?
| # decimal place followed by a comma and the corresponding count for | ||
| # that value, so we want to sum the rank sum value (first element). | ||
| # Rename annotation in the form 'AS_RAW_*_RankSum' to 'AS_*_RankSum'. | ||
| if k.startswith("AS_RAW_") and k.endswith("RankSum"): |
There was a problem hiding this comment.
Thoughts on making a drop_raw parameter that accepts a list of fields? I know this bit of code is really specific to these fields so no strong preference here and I dont even know if any other fields would ever apply given the above comment.
There was a problem hiding this comment.
I think I want to keep it as is since there is a lot in this code that is very specific and that is the way Laura defines those annotations here: https://docs.google.com/document/d/1ZdRERavtO1k4v0ppVzJ7gu6PHzSydznhYDmvZUb8qPo/edit#heading=h.gyrw4ywmz8b3
| if "AS_SB_TABLE" in info or "AS_SB" in info: | ||
| # Rename AS_SB to AS_SB_TABLE if present and add SB Ax2 aggregation logic. | ||
| if "AS_SB" in agg_expr: |
There was a problem hiding this comment.
Not sure this will happen but do we always want to overwrite AS_SB_TABLE with AS_SB when both are present? Should we have logic that throws a warning if this happens? Or is that the intended use?
There was a problem hiding this comment.
hmm, good point, I don't think it will happen, but I added a warning
Co-authored-by: Mike Wilson <mwilson@broadinstitute.org>
….com/broadinstitute/gnomad_methods into jg/change_compute_info_to_use_AS_ann
No description provided.