R语言学数据分析练习项目

################################################## ## R script for MetaboAnalyst ## Description: report generation using Sweave ## Note: most analyses were already performed, only need to embedding ## the results to the right place without rerun the whole analysis ## through Sweave. Only some auxilliary info (i.e. time, version etc need to ## run R throught Sweave ## ## Author: Jeff Xia, jeff.xia@mcgill.ca ## McGill University, Canada ## ## License: GNU GPL (>= 2) ################################################### CreatePDFReport<-function(usrName){ # create the Rnw file file.create("Analysis_Report.Rnw"); # open for write rnwFile <<- file("Analysis_Report.Rnw", "w") # create a global counter to label figures fig.count <<- 0; table.count <<- 0; if(analSet$type == "stat" ){ CreateStatRnwReport(usrName); }else if(analSet$type == "ts"){ CreateTimeSeriesRnwReport(usrName); }else if(substr(analSet$type, 0, 4) == "mset"){ CreateEnrichRnwReport(usrName); }else{ CreatePathRnwReport(usrName); } # close opened files close(rnwFile); Sweave("Analysis_Report.Rnw"); tools::texi2dvi("Analysis_Report.tex", pdf = TRUE, quiet=TRUE); } ################################ ## MetaboAnalyst statistics ############################### # write .Rnw file template CreateStatRnwReport<-function(usrName){ CreateHeader(usrName); CreateStatIntr(); CreateStatIOdoc(); CreateStatNORMdoc(); InitStatAnalMode(); if(exists("analSet")){ CreateUNIVdoc(); CreateANOVAdoc(); CreateCorrDoc(); CreatePCAdoc(); CreatePLSdoc(); CreateSAMdoc(); CreateEBAMdoc(); CreateHCdoc(); CreateKMdoc(); CreateSOMdoc(); CreateRFdoc(); CreateSVMdoc(); }else{ CreateAnalNullMsg(); } CreateStatFooter(); } # create header CreateHeader<-function(usrName){ header <- c("\\documentclass[a4paper]{article}", "\\usepackage[margin=1.0in]{geometry}", "\\usepackage{longtable}", "\\SweaveOpts{eps=FALSE,pdf=TRUE}", "\\title{Metabolomic Data Analysis with MetaboAnalyst 3.0}", paste("\\author{ User ID: ", usrName, " }", sep=""), "\\begin{document}", "\\parskip=.3cm", "\\maketitle"); cat(header, file=rnwFile, sep="\n", append=TRUE); } CreateStatIntr<-function(){ descr <- c("\\section{Data Processing and Normalization}\n"); cat(descr, file=rnwFile, append=TRUE); } # read and process the raw data CreateStatIOdoc<-function(){ descr <- c("\\subsection{Reading and Processing the Raw Data}\n", "MetaboAnalyst accepts a variety of data types generated in metabolomic studies,", "including compound concentration data, binned NMR/MS spectra data, NMR/MS peak", "list data, as well as MS spectra (NetCDF, mzXML, mzDATA).", "Users need to specify the data types when uploading their data in order for", "MetaboAnalyst to select the correct algorithm to process them.\n", paste("Table", table.count<<-table.count+1,"summarizes the result of the data processing steps.\n") ); cat(descr, file=rnwFile, append=TRUE); # error checking if(is.null(dataSet$orig) | is.null(dataSet$proc) | is.null(dataSet$cls)){ errorMsg<- c(descr, "Error occured during reading the raw data ....", "Failed to proceed. Please check if the data format you uploaded is correct.", "For further information, please contact \\texttt{jeff.xia@mcgill.ca} with", "a description of the problem. Your feedback is highly appreciated!\n"); cat(errorMsg, file=rnwFile, append=TRUE); return(); } if(dataSet$type=="conc"){ descr<-c("\\subsubsection{Reading Concentration Data}\n", "The concentration data should be uploaded in comma separated values (.csv) format.", "Samples can be in rows or columns, with class labels immediately following the sample IDs.\n"); cat(descr, file=rnwFile, append=TRUE); cat("\n\n", file=rnwFile, append=TRUE); cat(dataSet$read.msg, file=rnwFile, append=TRUE, sep="\n"); }else if(dataSet$type=="specbin"){ descr<-c("\\subsubsection{Reading Binned Spectral Data}\n", "The binned spectra data should be uploaded in comma seperated values (.csv) format.", "Samples can be in rows or columns, with class labels immediately following the sample IDs.\n"); cat(descr, file=rnwFile, append=TRUE); cat("\n\n", file=rnwFile, append=TRUE); cat(dataSet$read.msg, file=rnwFile, append=TRUE, sep="\n"); }else if(dataSet$type=="pktable"){ descr<-c("\\subsubsection{Reading Peak Intensity Table}\n", "The peak intensity table should be uploaded in comma separated values (.csv) format.", "Samples can be in rows or columns, with class labels immediately following the sample IDs.\n"); cat(descr, file=rnwFile, append=TRUE); cat("\n\n", file=rnwFile, append=TRUE); cat(dataSet$read.msg, file=rnwFile, append=TRUE, sep="\n"); }else if(dataSet$type=="nmrpeak"){ descr<-c("\\subsubsection{Reading NMR Peak List and Intensities Data}\n", "NMR peak list and intensities data should be uploaded as one zip file. It contains subfolders, one for", "each group. Each folder contains peak list files, one per spectrum. The peak list format", "is a a two-column comma separated values - the first column indicates peak position (ppm)", "and the second one for peak intensities. The first line is assumed to be column labels.", "The files should be saved in .csv format. For paired analysis, users need to upload", "a text file specifying the paired information. Each pair is indicated by their sample names", "seperated by a colon \":\" with one pair per line.\n"); cat(descr, file=rnwFile, append=TRUE); cat("\n\n", file=rnwFile, append=TRUE); cat(dataSet$read.msg, file=rnwFile, append=TRUE, sep="\n"); descr<-c("\\subsubsection{Peak List Alignment}\n", "Proximal peaks are first grouped together based on their position using a moving window of 0.03 ppm", "and a step of 0.015 ppm. Peaks of the same group are aligned to their median positions across all samples.", "If more than one peak from the same sample appear in the same group, they will be replaced by their sum.", "Some peaks that are detected in very few samples (less than half in both classes) are excluded.", "The aligned peaks are reorganized into a single data matrix for further analysis. The name of the parent", "folder is used as class label for each sample.\n"); cat(descr, file=rnwFile, append=TRUE); cat("\n\n", file=rnwFile, append=TRUE); cat(dataSet$proc.msg, file=rnwFile, append=TRUE, sep="\n"); }else if(dataSet$type=="mspeak"){ descr<-c("\\subsubsection{Reading MS Peak List and Intensities Data}\n", "MS peak list and intensities data should be uploaded as one zip file. It contains subfoulders with one for", "each group. Each folder contains peak list files, one per spectrum. The MS peak list format", "is either a two-column (mass and intensities) or three-column (mass, retention time, and intensities)", "comma separated values. The first line is assumed to be column labels.",