Package diggit. R topics documented: January 11, Version Date

Transcription

1 Version Date Package diggit January 11, 2019 Title Inference of Genetic Variants Driving Cellular Phenotypes Author Mariano J Alvarez <reef103@gmail.com> Maintainer Mariano J Alvarez <reef103@gmail.com> Depends R (>= 3.0.2), Biobase, methods Imports ks, viper(>= 1.3.1), parallel Suggests diggitdata Inference of Genetic Variants Driving Cellullar Phenotypes by the DIGGIT algorithm License file LICENSE biocviews SystemsBiology, NetworkEnrichment, GeneEpression, FunctionalPrediction, GeneRegulation git_url git_branch RELEASE_3_8 git_last_commit 038aa4d git_last_commit_date Date/Publication R topics documented: aecdf aqtl conditional correlation diggit-class fcnv marina mutualinfo plot,diggit-method print,diggit-method Inde 13 1

2 2 aqtl aecdf Approimate empirical commulative distribution function This function generates an empirical null model that computes a normalized statistics and p-value aecdf(dnull, symmetric = FALSE) dnull symmetric Numerical vector representing the null model Logical, whether the distribution should be treated as symmetric around zero and only one tail should be approimated function with two parameters, and alternative aqtl Inference of aqtl This function infers aqtls from F-CNVs and VIPER activity aqtl(,...) aqtl(, mr = 0.01, mr.adjust = c("none", "fdr", "bonferroni"), fcnv = 0.01, fcnv.adjust = c("none", "fdr", "bonferroni"), method = c("spearman", "mi", "pearson", "kendall"), mindy = FALSE, cores = 1, verbose = TRUE) Object of class diggit... Additional parameters to pass to the function mr mr.adjust fcnv Either a numerical value between 0 and 1 indicating the p-value threshold for the Master Regulator (MR) selection, or a vector of character strings listing the MRs Character string indicating the multiple hypothesis test correction for the MRs Either a numerical value between 0 and 1 indicating the p-value threshold for the F-CNV, or a vector of character strings listing the F-CNVs

3 conditional 3 fcnv.adjust Character string indicating the multiple hypothesis test correction for the F- CNVs method mindy cores verbose Character string indicating the method for computing the association between F-CNV and regulator activity (aqtl analysis) Logical, whether only post-translational modulators of each evaluated TF should be considered as putative genetic driver Integer indicating the number of cores to use (1 for Windows-based systems) Logical, whether progress should be reported Updated diggit object with viper and aqtl slots data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") data(gbm.aracne, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, cnv=gbmcnv, regulon=gbmtfregulon) dobj <- fcnv(dobj) dobj <- aqtl(dobj, mr=c("cebpd", "STAT3"), fcnv.adjust="fdr") dobj diggitaqtl(dobj)[, 1:4] conditional Conditional analysis of CNVs This function performs the conditional analysis of fcnvs conditional(,...) conditional(, pheno = "cond", group1, group2 = NULL, cnv = 0.2, mr = 0.01, mr.adjust = c("none", "fdr", "bonferroni"), modul = 0.01, modul.adjust = c("none", "fdr", "bonferroni"), fet.pval = 0.05, cores = 1, verbose = TRUE) Object of class diggit... Additional parameters to pass to the function pheno group1 group2 cnv Character string indicating the feature for sample groups Character string indicating the treatment group Optional character string indicating the reference group Single number or vector of two numbers indicating the thresholds for CNVs

4 4 correlation mr mr.adjust modul modul.adjust fet.pval cores verbose Either vector of character strings indicating the MR genes, or number indicating the corrected p-value threshold for selecting the MRs Character string indicating the multiple-hypothesis correction to apply to the MR p-values Number indicating the p-value threshold for a modulator to be considered associated with the MR activity Character string indicating the multiple-hypothesis correction to apply to the aqtl results Number indicating the FET p-value threshold for the association between CNVs and sample groups Integer indicating the number of cores to use (1 for Windows-based systems) Logical, whether progress should be reported Object of class diggit with conditional analysis results data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") data(gbm.aracne, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, cnv=gbmcnv, regulon=gbmtfregulon) dobj <- fcnv(dobj) dobj <- aqtl(dobj, mr=c("cebpd", "STAT3"), fcnv.adjust="fdr", verbose=false) dobj <- conditional(dobj, pheno="subtype", group1="mes", group2="pn", mr="stat3", verbose=false) dobj correlation Correlation test This function computes the correlation between and y given both are numeric vectors, between the columns of if it is a numeric matri, or between the columns of and y if both are numeric matries correlation(, y = NULL, method = c("pearson", "spearman", "kendall"), pairwise = FALSE) y method pairwise Numeric vector or matri Optional numeric vector or matri Character string indicating the correlation method Logical, wether columns of and y should be compared in a pairwise manner. and y must have the same number of columns

5 diggit-class 5 Details This function computes correlation and associated p-values Numeric value, vector or matri of results <- seq(0, 10, length=50) y <- +rnorm(length(), sd=2) correlation(, y) diggit-class The diggit class This class stores parameters and results of the diggit algorithm This function generates diggit class objects diggitclass(epset = NULL, cnv = NULL, regulon = NULL, mindy = NULL, fcnv = NULL, mr = NULL, viper = NULL, aqtl = NULL, conditional = NULL) epset cnv regulon mindy fcnv mr viper aqtl conditional EpressionSet object or numeric matri of epression data, with features in rows and samples in columns Numeric matri of CNV data Regulon class object containing the transcriptional interactome Regulon class object containing the post-translational interactome Vector of F-CNV p-values Vector of master regulator Z-score (NES) Numeric matri of VIPER results Numeric matri of aqtl p-values List containing the conditional analysis results Details see diggit-methods for related methods Object of class diggit

6 6 fcnv Slots epset: EpressionSet object containing the gene epression data cnv: Matrri containing the CNV data regulon: Regulon object containing the transcriptional interactome mindy: Regulon object containing the post-translational interactome fcnv: Numeric vector containing the p-values for functional CNVs mr: Numeric vector of normalized enrichment scores for the MARINa analysis viper: Numeric matri of normalized enrichment scores for the VIPER analysis aqtl: Numeric matri of association p-values for the aqtl analysis conditional: List containing the conditional analysis results data(gbm.epression, package="diggitdata") data(gbm.aracne, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, regulon=gbmtfregulon) print(dobj) fcnv Inference of functional CNVs This function infers functional CNVs by computing their association with gene epression fcnv(,...) fcnv(, epset = NULL, cnv = NULL, method = c("spearman", "mi", "pearson", "kendall"), cores = 1, verbose = TRUE) ## S4 method for signature 'EpressionSet' fcnv(, cnv, method = c("spearman", "mi", "pearson", "kendall"), cores = 1, verbose = TRUE) ## S4 method for signature 'matri' fcnv(, cnv, method = c("spearman", "mi", "pearson", "kendall"), cores = 1, verbose = TRUE) ## S4 method for signature 'data.frame' fcnv(, cnv, method = c("spearman", "mi", "pearson", "kendall"), cores = 1, verbose = TRUE)

7 marina 7 Object of class diggit, epressionset object or numeric matri of epression data, with features in rows and samples in columns... Additional arguments epset cnv method cores verbose Optional numeric matri of epression data Optional numeric matri of CNVs Character string indicating the method for computing the association between CNVs and epression Integer indicating the number of cores to use (1 for Windows-based systems) Logical, whether to report analysis progress Objet of class diggit with updated fcnv slot data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") genes <- intersect(rownames(gbmeprs), rownames(gbmcnv))[1:100] gbmcnv <- gbmcnv[match(genes, rownames(gbmcnv)), ] dgo <- diggitclass(epset=gbmeprs, cnv=gbmcnv) dgo <- fcnv(dgo) dgo diggitfcnv(dgo)[1:5] dgo <- fcnv(gbmeprs, gbmcnv) print(dgo) diggitfcnv(dgo)[1:5] dgo <- fcnv(eprs(gbmeprs), gbmcnv) dgo diggitfcnv(dgo)[1:5] dgo <- fcnv(as.data.frame(eprs(gbmeprs)), gbmcnv) dgo diggitfcnv(dgo)[1:5] marina Inference of Master Regulators This function infers the master regulators for the transition between two phenotypes marina(,...) ## S4 method for signature 'matri' marina(, y = NULL, mu = 0, regulon, per = 1000, cores = 1, verbose = TRUE)

8 8 marina ## S4 method for signature 'EpressionSet' marina(, pheno = "cond", group1, group2 = NULL, mu = 0, regulon, per = 1000, cores = 1, verbose = TRUE) marina(, pheno, group1, group2 = NULL, mu = 0, regulon = NULL, per = 1000, cores = 1, verbose = TRUE) Object of class diggit, epressionset object or numerical matri containing the test samples... Additional arguments y mu regulon per cores verbose pheno group1 group2 Numerical matri containing the control samples Number indicating the control mean when y is ommited Transcriptional interactome Interger indicating the number of permutations to compute the marina null model Integer indicating the number of cores to use (1 for Windows-based systems) Logical, whether progress should be reported Character string indicating the phenotype data to use Vector of character strings indicating the category from phenotype pheno to use as test group Vector of character strings indicating the category from phenotype pheno to use as control group Updated diggit object with Master Regulator results cores <- 3*(Sys.info()[1]!= "Windows")+1 data(gbm.epression, package="diggitdata") data(gbm.aracne, package="diggitdata") eset <- eprs(gbmeprs) samples <- pdata(gbmeprs)[["subtype"]] <- eset[, samples=="mes"] y <- eset[, samples=="pn"] dgo <- marina(, y, regulon=gbmtfregulon, per=100, cores=cores) dgo diggitmr(dgo)[1:5] dgo <- marina(gbmeprs, pheno="subtype", group1="mes", group2="pn", regulon=gbmtfregulon, per=100, cores=cor dgo diggitmr(dgo)[1:5] <- diggitclass(epset=gbmeprs, regulon=gbmtfregulon) dgo <- marina(, pheno="subtype", group1="mes", group2="pn", per=100, cores=cores) dgo diggitmr(dgo)[1:5]

9 mutualinfo 9 mutualinfo Mutual information This function estimates the mutual information between and y given both are numeric vectors, between the columns of if it is a numeric matri, or between the columns of and y if both are numeric matries mutualinfo(, y = NULL, per = 0, pairwise = FALSE, bw = 100, cores = 1, verbose = TRUE) y per pairwise bw cores verbose Numeric vector or matri Optional numeric vector or matri Integer indicating the number of permutations to compute p-values Logical, wether columns of and y should be compared in a pairwise maner. and y must have the same number of columns Integer indicating the grid size for integrating the joint probability density Integer indicating the number of cores to use (1 for Windows-based systems) Logical, whether progression bars should be shown Details This function estimates the mutual information between continuous variables using a fi bandwidth implementation Numeric value, vector or matri of results <- seq(0, pi, length=100) y <- 5*sin()+rnorm(100) cor.test(, y) mutualinfo(, y, per=100)

10 10 print,diggit-method plot,diggit-method Diggit plot This function generate plots for the diggit conditional analysis plot(, mr = NULL, cluster = NULL, sub = NULL,...) Diggit class object mr Optional vector of character strings indicating the MR names cluster Optional vector of cluster names sub Optional sub-title for the plot... Additional parameters to pass to the plot function Nothing, plots are generated in the default output device data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") data(gbm.aracne, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, cnv=gbmcnv, regulon=gbmtfregulon) dobj <- fcnv(dobj) dobj <- aqtl(dobj, mr=c("cebpd", "STAT3"), fcnv.adjust="fdr", verbose=false) dobj <- conditional(dobj, pheno="subtype", group1="mes", group2="pn", mr="stat3", verbose=false) plot(dobj, cluster="3") print,diggit-method Basic methods for class diggit This document lists a series of basic methods for the class diggit

11 print,diggit-method 11 print(, pval = 0.05) show(object) eprs(object) diggitcnv() diggitregulon() diggitmindy() diggitfcnv() diggitmr() diggitviper() diggitaqtl() diggitconditional() summary(object) head(, rows = 4, cols = 4) mindyfiltering(, mr = 0.01, mr.adjust = c("none", "fdr", "bonferroni")) pval object rows cols Object of class diggit P-value threshold for the conditional analysis Object of class diggit Integer indicating the maimum number of rows to show Integer indicating the maimum number of columns to show

12 12 print,diggit-method mr mr.adjust Either a numerical value between 0 and 1 indicating the p-value threshold for the Master Regulator (MR) selection, or a vector of character strings listing the MRs Character string indicating the multiple hypothesis test correction for the MRs print returns summary information about the diggit object show returns summary information about the object of class diggit eprs returns the EpressionSet object containing the epression profile data diggitcnv returns a matri containing the CNV data diggitregulon returns a regulon object containing the transcriptional interactome diggitmindy returns a regulon object containing the post-translational interactome diggitfcnv returns a vector of p-values for the F-CNVs diggitmr returns a vector of master regulators NES diggitviper returns a matri of VIPER results diggitaqtl returns a matri of aqtls (p-value) diggitconditional returns a list containing the conditional analysis results summary returns the integrated results from the conditional analysis head returns a list containing a reduced view for an object of class diggit mindyfiltering returns a diggit class object with CNV and aqtl slots filtered to contain only MINDy post-translational modulators of the MRs data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") data(gbm.aracne, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, cnv=gbmcnv, regulon=gbmtfregulon) print(dobj) show(dobj) eprs(dobj) diggitcnv(dobj)[1:3, 1:3] diggitregulon(dobj) diggitmindy(dobj) diggitfcnv(dobj) diggitmr(dobj) diggitviper(dobj) diggitaqtl(dobj) diggitconditional(dobj) head(dobj) data(gbm.epression, package="diggitdata") data(gbm.cnv, package="diggitdata") data(gbm.mindy, package="diggitdata") dobj <- diggitclass(epset=gbmeprs, cnv=gbmcnv, mindy=gbmmindy) dobj <- fcnv(dobj) dobj dobj <- mindyfiltering(dobj, mr=c("stat3", "CEBPD")) dobj

13 Inde aecdf, 2 aqtl, 2 aqtl,diggit-method (aqtl), 2 conditional, 3 conditional,diggit-method (conditional), 3 correlation, 4 diggit-class, 5 diggitaqtl diggitaqtl,diggit-method diggitclass (diggit-class), 5 diggitcnv diggitcnv,diggit-method diggitconditional diggitconditional,diggit-method diggitfcnv diggitfcnv,diggit-method diggitmindy diggitmindy,diggit-method diggitmr diggitmr,diggit-method diggitregulon diggitregulon,diggit-method diggitviper diggitviper,diggit-method fcnv,epressionset-method (fcnv), 6 fcnv,matri-method (fcnv), 6 head,diggit-method head.diggit marina, 7 marina,diggit-method (marina), 7 marina,epressionset-method (marina), 7 marina,matri-method (marina), 7 mindyfiltering mindyfiltering,diggit-method mutualinfo, 9 plot (plot,diggit-method), 10 plot,diggit-method, 10 print,diggit-method, 10 print.diggit show,diggit-method show.diggit summary,diggit-method summary.diggit eprs,diggit-method eprs.diggit fcnv, 6 fcnv,data.frame-method (fcnv), 6 fcnv,diggit-method (fcnv), 6 13