Menu ▾ ▴

MAP

Mi-kyoung Seo

MAP

Microsatellite instability Absolute single sample Predictor

MAP overview

We present a transcriptomic data-based, absolute, single sample predictor (MAP) that integrates profiles on molecular mechanisms, cancer signatures, immune signatures, and tumor microenvironments (TME) to predict MSI status. To do so, we used 466 samples from The Cancer Genome Atlas and validated the predictive power of the MAP, with 96% accuracy, using an additional 1118 samples, including 250 MSI tumors.

How to install

Download the source and install it as shown below.

install.packages("MAP_0.1.0.tar.gz", repos = NULL, type = "source")
library(MAP)

How to use

The Input should be a log2 transformed expression matrix.
You need expression data for all genes of whole transcriptome data that contain a list of 31 genes.

data(marker) # MAP signaure for MSI, list of 31 genes used in MAPgene model

LY6G6D CYP2W1 TNNC2 CTTNBP2 NKD1 CAB39L MLH1 EPM2AIP1 SHROOM4 RNF43
PRR15 ATP9A H2AFJ FARP1 TCF7 MAPRE3 ZMYND8 DDX27 TGFBR2 PIWIL4 FECH
DOCK5 TYMS HPSE ASPHD2 AGR2 GFI1 RPL22L1 RAB27B GNLY DUSP4

expData # RNA-seq expression data (eg. log2(RSEM+1)) file ("/path/to/expression.txt")

  • Set the folder where the output file will be created.

testP = "/path/to/test"
expData = "/path/to/expression.txt"
runMAP(expData, testP)

A description of the function can be found by entering :

help(package="MAP")