search

cortes-ciriano-lab / SComatic

A tool for detecting somatic variants in single cell data
Other
170 stars 27 forks source link

Detecting somatic SNPs with VAF=1 #71

Open Swap90 opened 5 days ago

Swap90 commented 5 days ago

Hi,

I noticed that the pipeline detects some somatic mutations with VAF = 1 in a specific cell type. Such mutations may be an example of cell-type-specific clonal expansion which may not be true somatic mutations (as per my understanding). Could you please advise on such situation and if the pipeline detects somatic mutations specific to cell type regardless of the VAF cutoff?

Following is an example of one of such mutations.

CHROM Start End REF ALT FILTER Cell_types Up_context Down_context N_ALT Dp Nc Bc Cc VAF CCF BCp CCp Cell_types_min_BC Cell_types_min_CC Rest_BC Rest_CC Fisher_p Cell_type_Filter INFO OPC EN Oligo Micro IN Astro

chr5 37610130 37610130 G A PASS Astro GCGTG TGGCA 1 5 5 5 5 1.0 1.0 0.0 0.0 2 2 0;10;1 0;10;1 . PASS DP|NC|CC|BC|BQ|BCf|BCr NA NA 10|10|0:0:0:10:0:0|0:0:0:10:0:0|0:0:0:370:0:0|0:0:0:10:0:0|0:0:0:0:0:0 NA NA 5|5|5:0:0:0:0:0|5:0:0:0:0:0|185:0:0:0:0:0|5:0:0:0:0:0|0:0:0:0:0:0

ArthurDondi commented 9 hours ago

Hi, SComatic doesn't directly have a VAF filter. It does filter mutations based on a beta-binomial test, but this test just checks if the number of mutated reads observed at the positions could be due to chance. A VAF of 1 is not a rare occurence in cancer, for example when there is an allele deletion and the remaining allele is mutated, so I would not recommend filtering them out.

Swap90 commented 32 minutes ago

Thank you for the explanation.