Scribd
Upload
54 views3 pages

ARTP (Adaptive Rank Truncated Product) Package: Detailed Examples of Computing The Gene and Path-Way P-Values

This document provides instructions for running gene and pathway analysis using the ARTP R package. It shows how to define phenotype and genotype data files, run permutations to generate observed and permuted p-values, and call the ARTP_pathway function to calculate gene and pathway p-values based on those files. Sample data and output from running the analysis on a small dataset is also shown.

Uploaded by

ajquinonesp
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
54 views3 pages

ARTP (Adaptive Rank Truncated Product) Package: Detailed Examples of Computing The Gene and Path-Way P-Values

This document provides instructions for running gene and pathway analysis using the ARTP R package. It shows how to define phenotype and genotype data files, run permutations to generate observed and permuted p-values, and call the ARTP_pathway function to calculate gene and pathway p-values based on those files. Sample data and output from running the analysis on a small dataset is also shown.

Uploaded by

ajquinonesp
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

ARTP(Adaptive Rank Truncated Product)

Package

February 20, 2014

> library(ARTP)

Detailed examples of computing the gene and path-


way p-values
We will start with the sample data of SNPs and sample phenotype data to
generate the observed and permutation p-values for each SNP in the pathway.
First, lets get the paths to the phenotype and genotype data
> pheno_file <- system.file("sampleData", "pheno_data.txt", package="ARTP")
> geno_file <- system.file("sampleData", "geno_data.txt", package="ARTP")
> print(pheno_file)

[1] "/tmp/RtmpqzSrr9/Rinst1a73f3a1816d1/ARTP/sampleData/pheno_data.txt"

> print(geno_file)

[1] "/tmp/RtmpqzSrr9/Rinst1a73f3a1816d1/ARTP/sampleData/geno_data.txt"

The phenotype file is tab-delimited text file and has columns, ”ID”, ”Y”,
”X1”, and ”X2”, where ”ID” is the subject id, ”Y” is the case-control status, ”X1”
and ”X2” are continuous variables. Define the list that describes the phenotype
data:
> pheno.list <- list(file=pheno_file, delimiter="\t", header=1, id.var="ID",
+ response.var="Y", main.vars=c("X1", "X2"))
The genotype file is also a tab-delimited text file of type 2 where row 1 has
the string ”ldat” followed by the subject ids. The first column of this file has
the SNP ids. Define the list that describes the genotype data:
> geno.list <- list(file=geno_file, delimiter="\t", file.type=2)
We need to choose a directory that has write access to serve as the directory
where the output files will be created. For this example, let this directory be
the working directory.
> out.dir <- getwd()
> print(out.dir)

1
[1] "/tmp/RtmpqzSrr9/Rbuild1a73f30334c9c/ARTP/vignettes"

We also need a file that gives the SNPs belonging to each gene. Let us use
the sample gene-SNP file which is a tab-delimited text file with columns ”SNP”
and ”Gene”.
> gs_file <- system.file("sampleData", "gene_SNP_data.txt", package="ARTP")
> print(gs_file)

[1] "/tmp/RtmpqzSrr9/Rinst1a73f3a1816d1/ARTP/sampleData/gene_SNP_data.txt"

Define the list that describes this file:


> gs.list <- list(file=gs_file, snp.var="SNP", gene.var="Gene", delimiter="\t", header=1)

Calling the runPermutations and ARTP pathway


functions
Define the names of the 2 output files that will store the observed p-values and
permutated p-values.
> obs.outfile <- paste(out.dir, "/", "obs.txt", sep="")
> perm.outfile <- paste(out.dir, "/", "perm.txt", sep="")
Set up the options list. Let us run 50 permutations and choose to generate
a new response vector for each permutation (perm.method=2).
> nperm <- 50
> op.list <- list(nperm=nperm, obs.outfile=obs.outfile, perm.outfile=perm.outfile, perm.met
Run the permutations. The base (NULL) model summary will be printed
to the console.
> runPermutations(geno.list, pheno.list, 1, op=op.list)

Call:
glm(formula = response0 ~ phenoData0[, -snpcol] - 1, family = family,
model = FALSE, x = TRUE, y = TRUE)

Deviance Residuals:
Min 1Q Median 3Q Max
-1.168 -1.128 -1.097 1.227 1.272

Coefficients:
Estimate Std. Error z value Pr(>|z|)
phenoData0[, -snpcol]x1 -0.1115 0.2396 -0.465 0.642
phenoData0[, -snpcol]x2 0.1024 0.3087 0.332 0.740
phenoData0[, -snpcol]x3 -0.1211 0.3168 -0.382 0.702

(Dispersion parameter for binomial family taken to be 1)

Null deviance: 693.15 on 500 degrees of freedom

2
Residual deviance: 691.08 on 497 degrees of freedom
AIC: 697.08

Number of Fisher Scoring iterations: 3

NULL

Now we have the observed p-values and permutated p-values stored in the
files obs.outfile and perm.outfile so that we can compute the gene and pathway
p-values by using the default parameters for op (see the manual for details).
> set.seed(76523)
> ret <- ARTP_pathway(obs.outfile, perm.outfile, nperm, out.dir, gene.list=gs.list)
> print(ret)

$pathway.pvalue
[1] 0.07843137

$gene.table
Gene N.SNP Pvalue
1 Gene_3 17 0.9803920
2 Gene_4 12 0.0784314
3 Gene_1 9 0.0196078
4 Gene_2 12 0.1176470

$nperm
[1] 50

Now compute the pathway p-value assuming all the SNPs belong to the same
gene. Note that if gene.list is NULL, then the program assumes all SNPs belong
to the same gene.
> set.seed(76523)
> ret <- ARTP_pathway(obs.outfile, perm.outfile, nperm, out.dir)
> print(ret)

$pathway.pvalue
[1] 0.09803922

$gene.table
Gene N.SNP Pvalue
1 gene 50 0.0980392

$nperm
[1] 50

You might also like