search

OSU-SRLab / MANTIS

Microsatellite Analysis for Normal-Tumor InStability
GNU General Public License v3.0
69 stars 27 forks source link

MANTIS on Mouse samples #38

Closed Rashesh7 closed 5 years ago

Rashesh7 commented 6 years ago

Hi,

I was wondering if you have tested MANTIS on mouse samples? Also I was thinking since Mantis uses the MS regions generated by RepeatFinder and gets the distribution of the repeats at those locations in the input tumor and normal samples, it should behave similarly for Mouse also, right?

Please let me know if I am assuming something incorrectly. Your support and suggestions are much appreciated.

Thank you!

rbonneville commented 6 years ago

Hello Rashesh7,

We have not tested MANTIS on mouse samples, however MANTIS should in theory work for mouse data as well as human. Note that RepeatFinder, given a fasta file, will identify all microsatellites in a genome (according to MANTIS's definition of a microsatellite). The resulting BED file should further be filtered to remove microsatellites outside of your capture region.

Rashesh7 commented 6 years ago

Hi @rbonneville , Yeah, that is what I was thinking. Thank you for your quick reply.

Also, do you think the threshold of 0.4 should still be fine to determine the instability?

rbonneville commented 6 years ago

This I am less sure about; we determined this threshold through testing with TCGA data from cohorts which had known MSI status by MSI-PCR. However, MANTIS should still be able to quantify the relative instability among a population of mice, which may well segregate into clearly defined subpopulations.