Classify: classify

In Oshlack/Baller: Classification of RNA-seq data from ALL samples

Description

Classifies the streamlined log FPKM samples into one of the four classes: Phlike, ETV, ERG, Other or returns Unclassified if none of the classes have a probability of classification above the threshold (default: ERG:0.25, ETV:0.25, Phlike:0.5, Other:0.75)

Usage

classify(FPKM,thresh)

Arguments

FPKM	A matrix of fragments/reads per kilobase per million. Rows are genes and columns are samples.
thresh	A character vector of four floats (between 0 and 1) which are the thresholds for ERG,ETV,Phlike and Other respectively.

Value

A data frame where each row represents the probability for given a sample to fit each of the four classes as well as its overall classification based on the chosen thresholds.

Examples

cf <- system.file("data","test_data.txt",package="AllSorts") #Get path to raw text file (a tsv)
counts <- read.table(file=cf,sep=' ',stringsAsFactors = FALSE,header=TRUE)
FPKM <- streamline(counts[,c(1:6)],counts$Gene_Length)
thresh <- c(0.2,0.2,0.8,0.9)
classed <- classify(FPKM,thresh)

Oshlack/Baller documentation built on Jan. 23, 2021, 6:50 p.m.