Hello fellow Bioinformaticians,

I'm a fresher and currently working in Matched Tumor-Normal samples (Specifically Lung cancer Tumor and the blood from the same patient). I want to know the somatic mutation in each patient. I have built a pretty good pipeline.

Tumor-Normal (4 fastq files) -> MultiQC -> Fastp -> MultiQC ->BWA-MEM2 ->Sortsam-> MarkDuplicates->BQSR->Mutect2->gatkvariantfilter->SNPEff eff. (Please suggest me if this pipeline is good enough.)

Recently I was told to incorporate Panel of Normal (PON) into my pipeline. I read about PON, and have a few doubts. I would be grateful if anyone can help me clarify.

Do I have to make my own PON? Or can I use the one that is available publicly? Is it ok to use that? (I do not have PON and have no source to make it)

If I have a PON, in the pipeline where will I incorporate it, like at what step?

I would be grateful for all your suggestions. Thank you!!