Generate de novo calls #654
Conversation
| @@ -1015,7 +1017,33 @@ def dense_trio_mt( | |||
| version: MatrixTableResource( | |||
| f"{get_sample_qc_root(version, test, data_type='exomes')}" | |||
| f"/relatedness/trios/gnomad.{data_type}.v{version}.trios" | |||
| f"{'.releasable' if releasable else ''}.dense.mt" | |||
| f"{'.releasable' if releasable else ''}.dense" | |||
| f"{'.split' if split else ''}.mt" | |||
There was a problem hiding this comment.
this update means that you can no longer easily access the dense but unsplit MT, right?
There was a problem hiding this comment.
No, they are both there, Julia had both versions saved already.
There was a problem hiding this comment.
yes, I mean that this won't return .mt at the end of the path if you only want the dense, unsplit MT, right?
There was a problem hiding this comment.
It will, like this:
from gnomad_qc.v4.resources.sample_qc import dense_trio_mt
print(dense_trio_mt().path)
print(dense_trio_mt(split=True).path)
outputing:
gs://gnomad/v4.0/sample_qc/exomes/relatedness/trios/gnomad.exomes.v4.0.trios.releasable.dense.mt
gs://gnomad/v4.0/sample_qc/exomes/relatedness/trios/gnomad.exomes.v4.0.trios.releasable.dense.split.mt
| @@ -104,52 +106,64 @@ def filter_de_novos(ht: hl.Table) -> hl.Table: | |||
| return ht | |||
|
|
|||
|
|
|||
| def get_releasable_trios_dense_mt( | |||
There was a problem hiding this comment.
I think we should keep this function since it documents how the dense MT was created
There was a problem hiding this comment.
It's not how the dense MT was created, it's here now, and I think we should remove them.
There was a problem hiding this comment.
ah, OK, fine with me to remove if the code is saved elsewhere
| logger.info("Filtering to chr20 for testing...") | ||
| mt = hl.filter_intervals(mt, [hl.parse_locus_interval("chr20")]) | ||
|
|
||
| mt = mt.annotate_rows( |
There was a problem hiding this comment.
are you adding these annotations because some of these variants weren't present in our release table (and therefore won't have information in our previously calculated allele_info struct)?
There was a problem hiding this comment.
Julia added this before splitting, since the split table is already there, I kept a record here.
| filter_samples_ht=meta_ht, | ||
| filter_variant_ht=var_ht, | ||
| checkpoint_variant_data=True, | ||
| # Approximate the AD and PL fields when missing. |
There was a problem hiding this comment.
ah I asked about this in the other PR -- can you add this code to the gnomAD methods function, or at least add a note that this is a way to approximate AD and PL?
There was a problem hiding this comment.
also, can you add in the comment here that these are missing for a reason, and only for homref genotypes (we dropped them to save on storage space/costs)?
There was a problem hiding this comment.
I think it's better to be here because it approximate before convert to a trio matrix.
| from gnomad.utils.slack import slack_notifications | ||
| from gnomad.utils.vep import process_consequences | ||
| from hail.methods.family_methods import trio_matrix |
There was a problem hiding this comment.
can you call this method rather than import it here? https://hail.is/docs/0.2/methods/genetics.html#hail.methods.trio_matrix
There was a problem hiding this comment.
Thank you! I didn't know this.
| tm = trio_matrix(mt, ped, complete_trios=True) | ||
| tm = tm.checkpoint(new_temp_file("trio_matrix", "mt")) | ||
|
|
||
| ht = tm.entries() |
There was a problem hiding this comment.
do you need all of the entries? can you drop some to save space?
There was a problem hiding this comment.
Julia already selected entries for the dense MT here, do we want to delete "SB" etc?
Co-authored-by: Katherine Chao <[email protected]>
This cleaned up some redundant dense mt functions and arguments since we had PR #651. Julia created the dense MT in job 0a3e98e75ba14b2f8341012515d11f8b before the PR was merged.
This is waiting on PR #760 in gnomad_methods.
Test run on chr20: fbca7f3460b84ceab02933a644d22cc2