How to determine low and high cut-offs?

xjtu-omics / msisensor-pro

Microsatellite Instability (MSI) detection using high-throughput sequencing data.

Other

93 stars 20 forks source link

How to determine low and high cut-offs? #50

Closed komalsrathi closed 1 year ago

komalsrathi commented 1 year ago

Hi,

Here is the distribution of the % Somatic sites from 132 paired tumor-normal msisensor runs (WGS):

Is there a cut-off that can tell me what is considered as low or high instability?

Thanks!

komalsrathi commented 1 year ago

@PengJia6 Hi, any suggestions would be much appreciated.

PengJia6 commented 1 year ago

Did you use whole exom sequencing techonology or target gene sequencing? I think 2% maybe the best cut-off for your solutions! But I suggest you do a PCR test (gold standard) for one negative and one positive samples (2% cut-off) to confirm the cut-off.

komalsrathi commented 1 year ago

Based on our runs, I have checked it says Whole Genome Sequencing.

PengJia6 commented 1 year ago

Based on our runs, I have checked it says Whole Genome Sequencing.

For whole exome sequencing, the cut-off may be 3.5%, you could found it in msisensor paper. For WGS, I suggest you validate two samples with MSI-PCR (gold standard) if you have materials.

komalsrathi commented 1 year ago

Thank you very much for the feedback.