Skip to content

Best Practices

Jump to bottom
Peng Jia edited this page Oct 15, 2021 · 10 revisions

Best Practices for MSI classification using MSIsensor-pro

(a) For tumor only samples:

1. scan : scan the reference genome to get microsatellites information

      msisensor-pro scan -d /path/to/reference.fa -o /path/to/reference.list

This module scans the reference genome to get microsatellites information. You need to input (-d) a reference file (*.fa or *.fasta), and you will get a microsatellites file (-o) for following analysis.

2. baseline : build baseline for tumor only detection

      msisensor-pro baseline -d /path/to/reference.list -i /path/to/configure.txt -o /path/to/baseline/directory

This module builds baseline for the input microsatellites (-d) from the scan module output or our github. You also need to offer some normal sample sequence data (-i，click here for more detail about configure file) from the sample sequencing center or platform and the output directory (-o).

3. pro : evaluate MSI using single (tumor) sample sequencing data

      msisensor-pro pro -d /path/to/baseline/directory/reference_baseline.list -t /path/to/tumor/case1_sorted.bam -o /path/to/output

This module scores the MSI using the tumor only sequence data. You need to offer the microsatellites with baseline (-d) from the baseline module, the aligned sequencing file (-t) and the output prefix (-o).

(b) For tumor-normal paired samples:

1. scan : scan the reference genome to get microsatellites information

      msisensor-pro scan -d /path/to/reference.fa -o /path/to/reference.site

This module scans the reference genome to get microsatellites information. You need to input (-d) a reference file (*.fa or *.fasta), and you will get a microsatellites file (-o) for following analysis.

2. msi : evaluate MSI using paired tumor-normal sequencing data

      msisensor-pro msi -d /path/to/reference.site -n /path/to/case1_normal_sorted.bam -t /path/to/case1_tumor_sorted.bam -o /path/to/output

This module scores the MSI using the tumor-normal paired sequeuence data. You need to offer the microsatellites (-d) from the scan module or our github (GRCh38.d1.vd1 ), the aligned sequencing file (-t ,-o ) and the output prefix (-o).

Clone this wiki locally