VisualMSI for MSI Calculation using NGS/WXS

Question

[sruthi]

Hi community. There is another tool "VisualMSI" for MSI calculation and its installation steps are here When I run the command:

visualmsi -i tumor.sorted.bam -r hg19.fasta -t targets/msi.tsv

it says visualmsi command not found.Could anyone tell me how to troubleshoot it?

TIA

Answer 1

[Ranan Jyoti Sarma]

First, install histlib: https://github.com/samtools/htslib/releases/download/1.9/htslib-1.9.tar.bz2 Extract it. ./configure make sudo make install

Download it : git clone https://github.com/OpenGene/VisualMSI.git # build cd VisualMSI make

Install

sudo make install

---

$ sudo make install install visualmsi /usr/local/bin/visualmsi Installed.

---

[sruthi]

Hey Ranan, Thanks will do it. I had saved binary earlier.

[Ranan Jyoti Sarma]

If you have saved then binary then do this:

Suppose you saved the binary in "Download" (~/Home/Downloads) Folder using wget http://opengene.org/VisualMSI/visualmsi

Go to Downloads and provide: sudo chmod 777 visualmsi

sudo cp visualmsi /usr/local/bin

Then Execute from anywhere.

[sruthi]

Can bed file be given instead of tsv?

[Ranan Jyoti Sarma]

Please go through the Target file generation: BED convert to tsv.

https://github.com/OpenGene/VisualMSI/blob/master/README.md

[sruthi]

Okay. Thanks will check it out. The tsv file given in the target folder is based on hg19 coordination, what must I do to get a target file based on hg38 coordination? Is it enough to convert hg38.bed to tsv?

TIA

[Ranan Jyoti Sarma]

It is always better to use one version. Either hg19 or GRCh38. If you have hg38 then convert it to hg19 coordination system.

[sruthi]

Agreed. But I have been using hg38 for all my analysis, hence. Now, how do I convert hg19 coordination to hg38?

[sruthi]

And when the value is 0 for "entropy of tumor data", "supporting reads of tumor data" and "quality control:failed", does it mean, the sample is MSS/ No microsatellites or is something wrong?

TIA

[Ranan Jyoti Sarma]

and yes ... you have to convert hg19 to hg38.

Something wrong. It should not be like that.

Try Using other tool:

https://omictools.com/microsatellite-instability-detection-category

[sruthi]

Thanks Ratan. I have read about all the tools and most of them need matched normal except mSINGS, MSIsensor. For msiNGS: I need to create baseline file with 10-20 normal samples, which I do not have access to. In MSIsensor: I'm getting 0 even after setting -c to maximum and minimum numbers.

Coming back to conversion of hg19 to hg38. What tool should I use for conversion. Tools in UCSC?

[sruthi]

Another doubt, I have bed file for hg38. How do I convert it to tsv? I do not understand the conversion steps mentioned here

Could you please explain it to me? TIA

[Ranan Jyoti Sarma]

It is already well explained in the file.

[vaish01kv]

Hi, README.md file clearly says "higher the EMD, more chances of MSI locus to be instable", and EMD is evaluated against a normal sample. Then how does the algorithm work for case-only mode?

Could anyone just explain this? TIA