search

niu-lab / msisensor2

Microsatellite instability (MSI) detection for tumor only data.
GNU General Public License v3.0
94 stars 21 forks source link

Msisensor on WES-T vs WGS-C #20

Open Ephedria opened 3 years ago

Ephedria commented 3 years ago

Hello,

First, thank for your tool, it is really useful.

I was wondering if using different technology can include a bias. To be more clear : I have several samples which I analyze using the WGS-T against the WGS-C. I also have the WES-T on these samples. I used the WES-T against the WGS-C. The comparison on these samples show a lot of discrepancy : comparison using WES-T / WGS-C tend to show more Unstable status than the comparison WGS-T / WGS-C .

Since the coverage is different, ( ~200X for the WES-T and ~60X for the WGS-T) can it explain this bias ?

Best,

owehann commented 3 years ago

Hi Ephiedria Sorry for the late feedback. It is not recommended for paired MSI calling using different types of sequences. Besides, there are more microsatellites in WGS, which leading to more valid sites (denominator of MSIscore). We do not think that coverage is the main point, and suggest that you can compare the difference in the number of valid sites and unstable sites obtained by your trails. Thanks Xinyin