How to determine the value of captureSize?

[h170607]

hi,
I am using the tmb() function to calculate the TMB value of a set of panel sequencing data:
tmb1 <- tmb(maf, captureSize =1.869664, logScale = T)
Where 1.869664 is the length of the targeted region calculated from the bed file.
Then most of the TMB values obtained are above 100, which are suspicious from experience.
Is the value of captureSize wrong?

[PoisonAlien]

Hi.

Thanks for the issue. If you have estimated the size from the BED file, it is the right way. Note that the capture size should in MBs.
I am not sure why the tmb is so high, but you should check your variant calling and filtering steps.

[h170607]

Hi,
My pipeline is like this:
gatk Mutect2 -L $bed -R $reference -I normal.bam -I tumor.bam -normal $normal -tumor $tumor -O ${sample}.raw.vcf

gatk FilterMutectCalls -R $reference -V ${sample}.raw.vcf -O ${sample}.Filter.vcf

convert2annovar.pl --format vcf4 --includeinfo ${sample}.Filter.vcf > ${sample}.avinput --allsample --withfreq

table_annovar.pl ${sample}.avinput $annovar/humandb/ -buildver hg19 -out ${sample} -protocol refGene,clinvar_20200316,cosmic91_coding,cosmic91_noncoding,exac03,1000g2015aug_eas,1000g2015aug_all,esp6500siv2_all,avsnp150,gnomad_genome,gnomad_exome,mcap13,revel,dbnsfp35a,intervar_20180118 -operation g,f,f,f,f,f,f,f,f,f,f,f,f,f,f -csvout

Then the resulted ${sample}.hg19_multianno.csv was converted to the maf format using annovarToMaf(). Is there anything wrong with this process?