Hello,
When using VarScan in somatic mode, I filter out the positions with less than 10 read counts.
I ran VarScan for 7 samples this way:
When checking my results, I noticed that some potential somatic mutations had less than 10 read counts in either normal ( normal_reads1+normal_reads2<10 ) or tumor sample ( tumor_reads1+tumor_reads2<10 ). It was not supposed to be so since I asked for at least 10 reads in both samples...
Then, I reanalyse my data: for 7 samples, from the output files (.indel & .snp), I had to remove several thousands of positions having less than 10 read counts.
Do someone notice that too?
I wonder why these positions are still here... Do you think there might be conflicts between several parameters? Or it could be version or mode (somatic) dependent?
Any feedback or suggestion would be appreciated,
Thank you,
Jane
When using VarScan in somatic mode, I filter out the positions with less than 10 read counts.
I ran VarScan for 7 samples this way:
java -Xmx4g -jar VarScan.v2.2.11.jar somatic /data/patient1/fibros.pileup /data/patient1/296.pileup --output-snp /data/patient1/output_varscan.snp --output-indel /data/patient1/output_varscan.indel --min-coverage-normal 10 --min-coverage-tumor 10 --min-var-freq 0.125 --min-freq-for-hom 0.75 --normal-purity 1 --tumor-purity 1 --p-value 0.05 --somatic-p-value 0.05 --strand-filter 0 --min-avg-qual 20 --min-strands2 0 --min-reads2 0
Then, I reanalyse my data: for 7 samples, from the output files (.indel & .snp), I had to remove several thousands of positions having less than 10 read counts.
Do someone notice that too?
I wonder why these positions are still here... Do you think there might be conflicts between several parameters? Or it could be version or mode (somatic) dependent?
Any feedback or suggestion would be appreciated,
Thank you,
Jane
Comment