vannberg / kea_drugresponse Goto Github PK

Drug response prediction

	R package use for predicting drug response of cell lines/patients.

	This R package analyzes cancer cell lines /patients gene expression file (Affymetrix U133 plus2.0), 
	and predicts response to 7 FDA proved cancer drugs (Gemcitabine, Gefitinib, Cisplatin, Doxorubicin, 
	Docetaxel, Paclitaxel, Carboplatin).




Requirement

	This R package requires user to pre-install affy package from bioconductor 


Install instruction

	R install affy package
	source("http://bioconductor.org/biocLite.R")
	biocLite("affy")

	R install KEADrugResponse package
	Download zip file from build folder and use install.package function in R to install
	R CMD INSTALL --merge-multiarch KEADrugResponse_*.tar.gz
	Unzip file

Usage

	library(KEADrugResponse); library(affy);
	DrugResponse.predict(patient_inputfile, inputfile_format, sample_name, path_to_model)

Arguments

	input_directory			directory of gene expression data for patient.

	inputfile_format		input file format, can be cel (cel file) or exp (gene expression file)

	sample_name				sample name

	path_to_model			where the model stored, default to 'KEA_DrugResponse/data'. 
 
	return a matrix of positive or negative score for each drug

	output pdf files of prediction score for each patient


Example

	input .cel file
	DrugResponse.predict(input_directory,'cel',patient name,'KEA_DrugResponse/data')


Pipeline

	we provide a bash script to run test for cancer samples in pipeline folder.
	Each pipeline test uses the model file in KEA_DrugResponse/data (default $DATAPATH) folder. 
	Change the $DATAPATH variable in pipeline script for different work path

	cd into pipeline folder
	# cd pipeline

	run test for GSE sets
	# ./DrugResponse_sample_test_GSE.sh GSE38069

	run test for GSM samples
	# ./DrugResponse_sample_test_GSM.sh GSM516801

	run test for user samples
	put all samples in a folder
	# ./DrugResponse_sample_test_multi.sh path.to.samples.folder