search

AlexandrovLab / SigProfilerClusters

Tool for analyzing the inter-mutational distances between SNV-SNV and INDEL-INDEL mutations. Tool separates mutations into clustered and non-clustered groups on a sample-dependent basis.
BSD 2-Clause "Simplified" License
11 stars 1 forks source link

Problem with MNV records in Mutect2 VCFs #10

Closed eallen-bbi closed 2 years ago

eallen-bbi commented 2 years ago

When the input VCF has rows with MNVs, then SigProfilerClusters crashes during subclassification of clustered mutations. Example VCF row excerpt and error message are copied below:

VCF file: chr9 534258 . GT CC . PASS

err log file: Traceback (most recent call last): File "/home/eallen/miniconda3/lib/python3.8/site-packages/SigProfilerClusters/classifyFunctions.py", line 237, in pullVaf vaf = vafs[sample][newKey] KeyError: '9:534258:G:C'

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "", line 1, in File "/home/eallen/miniconda3/lib/python3.8/site-packages/SigProfilerClusters/SigProfilerClusters.py", line 685, in analysis classifyFunctions.pullVaf (project, input_path, sanger, TCGA, standardVC, correction) File "/home/eallen/miniconda3/lib/python3.8/site-packages/SigProfilerClusters/classifyFunctions.py", line 240, in pullVaf vaf = vafs[sample][newKey] KeyError: 'chr9:534258:G:C'

ebergstr commented 2 years ago

Hi,

Thank you for pointing out this limitation and apologies for the delay. We have now incorporated supported for single line MNVs. Please update your version of the tool to 1.0.11 to reflect these changes.

Best, Erik

eallen-bbi commented 2 years ago

Great, thanks for the quick action. Will try out in next 2-3 days and reply back on results.

eallen-bbi commented 2 years ago

OK, good news - the fix seems to have worked. Ran on several samples without error. Thanks again.