Hi everyone,
I have followed the gatk somatic short variants discovery workflow but I found that there are some variants that are adjacent to each other as below. I am wondering if there is a reason why mutect does not combine them? Should I be checking all the mutations that are very close by in the bam file one by one?
e.g. M35 (9683756 CC>TT isn't it the same mutation as 9683757 C>T?), M48 (49041108 C>A 49041109 C>A so I should change it into CC>AA?) .
and how would you deal with e.g. M46 (39727366 C>CAA and 39727367 TCC>T there is an insertion and deletion right next to each other? Or is it an indication that I might have done something wrong? These mutaitons represents less than 1% of the total number of mutations but I just want to make sure I am doing the right thing.
Thank you for your help!
id chr pos ref alt
M35 1 9683756 CC TT
M35 1 9683757 C T
M35 10 43114056 G GTT
M35 10 43114057 ACAC A
M45 14 22978099 G A
M45 14 22978100 A AGGTGAGGAAGGAAGGAG
M45 20 32434632 C T
M45 20 32434633 AT A
M45 4 186589050 G T
M45 4 186589051 AG A
M46 14 78835175 G A
M46 14 78835176 G A
M46 17 7674220 C G
M46 17 7674221 G C
M46 17 39727366 C CAA
M46 17 39727367 TCC T
M46 2 38986124 T TG
M46 2 38986125 T TAGGCTCTCTAGCTGGTAGAATAATCTGAGCTACAA
M47 4 186618715 T C
M47 4 186618716 C T
M47 9 132944745 GG AA
M47 9 132944746 G A
M48 12 49041108 C A
M48 12 49041109 C A
M51 11 24201469 A AGC
M51 11 24201470 A ATTCTAGC
M53 4 186706729 C CCA
M53 4 186706730 ACGG A
M54 17 12127139 C T
M54 17 12127140 GT G
do you have the associated allele frequency for each site ?
Hi Nicolas yes I have added the column which contains the count and af (bold). Please let me know what you think. Thanks again.