DeltaMSI: artificial intelligence-based modeling of microsatellite instability scoring on next-generation sequencing data

Swaerts, Koen; Dedeurwaerdere, Franceska; De Smet, Dieter; De Jaeger, Peter; Martens, Geert A.

doi:10.1186/s12859-023-05186-3

Table 1 Overview of published MSI screening scripts on NGS data

From: DeltaMSI: artificial intelligence-based modeling of microsatellite instability scoring on next-generation sequencing data

Tool	Year	Normal required	Data type	Cohort size	Tumor type	Number of MSI loci	Loci selection	Method	Raw data	References
MSIsensor	2014	Yes	Exome/large panels	242	Endometrium	> 1000	Using all	Peak based, chi-square test	Bam	https://doi.org/10.1093/bioinformatics/btt755
MSIseq	2015	Yes	Exome	526	4 types			Variant based, decision tree	Bam	https://doi.org/10.1038/srep13321
Mantis	2016	Yes	Exome	458	6 types	> 100	Using all	Peak based, distance score	Bam	https://doi.org/10.18632/oncotarget.13918
MSIpred	2018		Exome	1432			–	Variant based, support vector machine	Maf	https://doi.org/10.1038/s41598-018-35682-z
MSI-ColonCore	2017	Yes	Capture	91	Colorectal	22	Manual	Peak based, based on MSIsensor	Bam	https://doi.org/10.1016/j.jmoldx,2017.11.007
USCI-msi	2020	Yes	Capture	64	Colorectal	9	Automatic	Location selection optimisation, uses Mantis	Bam	https://doi.org/10.1186/s12967-020-02373-1
NovoPM-MSI	2020	Yes	Capture	113	Colorectal	19	Manual	Peak based, Man-Witney U-test	Bam	https://doi.org/10.3892/ol.2020.11702
MIAmS	2019	No	Amplicon	286		6	Manual	Peak based, mSINGS + support vector machine correction	Fastq	https://doi.org/10.1093/bioinformatics/btz797
MSIsensor-pro	2020	No	Exome	1532	Colorectal	11,666	Automatic	Peak based, multinomial distribution	Bam	https://doi.org/10.1016/j.gpb.2020.02.001
MSIFinder	2021	No	Capture	419		54	Manual	Peak based, random forest	Bam	https://doi.org/10.1186/s12859-021-03986-z
MEM	2021	No	Capture	146	Colorectal	5	Manual	Peak based, expectation–maximisation	Fastq	https://doi.org/10.3390/cancers13164203
mSINGS	2014	No	Capture	324	Colorectal	15-2957	Manual	Peak based, Z-test	Samtools mpileup	https://doi.org/10.1373/clinchem.2014.223677
DeltaMSI		No	Capture/amplicon	331	11 types	28	Automatic	Peak based, logistic regression and support vector machine	Bam	Present report

Overview of MSI screening scripts with year of initial publication and reference. Defining features of these scripts are the type of NGS data used (small gene panels to exomes), the number of microsatellite loci used, the metrics used to score indel distributions as stable/unstable and manual versus automatic selection of loci with diagnostic power. The table also lists the number of tumor samples used for training and validation and the number of tumor types examined

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com

BMC Bioinformatics

Contact us