R and PARAFAC

R

If you want to get ANYWHERE using this package, I (-- GeorgeSteinhardt) would STROGNLY suggest searching this site: http://tolstoy.newcastle.edu.au/R/

R on Z

• R is an open source statistical package. This is how you can access R from Z:
  • You need Cygwin in order for the X11 graphics to work properly. It is available here: http://cygwin.com/ After you install, run cygwin and type the command startx. This enables you to run graphs.
  • PUTTY: Make sure you enable X11 forwarding by clicking on the X11 tab under the SSH category
  • Log into z-login. Type the following command to get yourself into the proper direcotry: cd ../../usr/local/R-2.3.2/bin
  • Type ./R to begin running R from Z.
• MACINTOSH INSTRUCTIONS:
  • You need X11 on your computer if you expect to run graphics for R. Install X11, either from your system DVD or by downloading it. If you are running mac os 10.4, this MUST BE DONE FROM INSTALL DVD.
  • When you login to z, type the following using a capital X at the end: username@z-login.cti.depaul.edu -X *if anyone has any questions about running R on Z, they can talk to GeorgeSteinhardt email archaeopteryx2006 at gmail dot com

Functions available in R

This is a list of functions (called packages) that are available in R that Dave thought would be useful to our own work. This is really obvious, but the full list is available at http://cran.r-project.org/src/contrib/PACKAGES.html

Phylogenetics

• ape Analyses of Phylogenetics and Evolution
• apTreeshape Analyses of Phylogenetic Treeshape

Mass Spectrometry

• caMassClass Processing & Classification of Protein Mass Spectra (SELDI) Data

Other Bioinformatics

• clac Clust Along Chromosomes, a method to call gains/losses in CGH arr
• clusterRepro Reproducibility of gene expression clusters
• DICOM Import medical imaging data using the DICOM Standard
• gap Genetic analysis package
• Geneland Simulation and MCMC inference in landscape genetics
• GeneNT Relevance or Dependency network and signaling pathway discovery
• LDheatmap Graphical display of pairwise linkage disequilibria between SNPs
• limma Linear Models for Microarray Data
• lodplot Plot a genome scan
• maanova Tools for analyzing Micro Array experiments
• pamr Pam: prediction analysis for microarrays
• PHYLOGR Functions for phylogenetically based statistical analyses
• ProbeR Reliability for gene expression from Affymetrix chip
• qtl Tools for analyzing QTL experiments
• qtlDesign Design of QTL experiments
• RJaCGH Reversible Jump MCMC for the analysis of CGH arrays.
• RLMM A Genotype Calling Algorithm for Affymetrix SNP Arrays
• samr SAM: Significance Analysis of Microarrays
• seqinr Biological Sequences Retrieval and Analysis
• supclust Supervised Clustering of Genes
• varmixt Mixture model on the Variance for the analysis of gene expression

Datamining (or possibly datamining), AI, Monte Carlo, etc.

• arules Mining Association Rules and Frequent Itemsets
• classPP Projection Pursuit for supervised classification
• clue Cluster ensembles
• cluster Cluster Analysis Extended Rousseeuw et al.
• clustvarsel Variable Selection for Model-Based Clustering
• FactoMineR Factor Analysis and Data Mining with R
• flexclust Flexible Cluster Algorithms
• rattle A graphical user interface for data mining in R using GTK
• kknn Weighted k-Nearest Neighbors
• knncat Nearest-neighbor classification with categorical variables
• knnFinder Fast Near Neighbour Search
• knnTree k-nn classification with variable selection inside leaves of a tr
• RWeka R/Weka interface
• genalg R Based Genetic Algorithm
• grnnR A Generalized Regression Neural Network
• neural Neural Networks
• Geneland Simulation and MCMC inference in landscape genetics
• boa Bayesian Output Analysis Program (BOA) for MCMC
• coda Output analysis and diagnostics for MCMC
• mcmc Markov Chain Monte Carlo
• MCMCpack Markov chain Monte Carlo (MCMC) Package
• scapeMCMC MCMC diagnostic plots
• exactLoglinTest Monte Carlo Exact Tests for Log-linear models

Grid, parallel, web programming

• biopara Self-contained parallel system for R
• fork R functions for handling multiple processes.
• Rmpi Interface (Wrapper) to MPI (Message-Passing Interface)
• rpvm R interface to PVM (Parallel Virtual Machine)
• CGIwithR CGI Programming in R
• taskPR Task-Parallel R Package
• httpRequest Basic HTTP Request
• R.rsp R Server Pages
• R2HTML HTML exportation for R objects
• XML Tools for parsing and generating XML within R and S-Plus.
• xtable Export tables to LaTeX or HTML

R programming

• butler Unit testing, profiling and benchmarking for R
• debug MVB's debugger for R
• matlab MATLAB emulation package
• R.matlab Read and write of MAT files together with R-to-Matlab connectivity
• RUnit R Unit test framework

Wavelets

• rwt Rice Wavelet Toolbox wrapper
• wavelets A package of funtions for computing wavelet filters, wavelet tran
• waveslim Basic wavelet routines for one-, two- and three-dimensional signa
• wavethresh Software to perform wavelet statistics and transforms

PARAFAC

Instructions to get started

Use the PTAk package. This is installed on z-login and is ready to use

You must load some libraries in order to run a PARAFAC analysis. Type in:

library(tensor)

and then:

library(PTAk)

[Hint: In Windows, you only need to do "library(PTAk)"] Once these libraries are properly loaded, use the CANDPARA function. See REFERENCE for proper usage instructions. In order for the function to work, you must have a tensor. You can create an example tensor by following this general formula: x<- array(1:20, dim=c(4,5,4)). This creates a 4x5x4 array of numbers 1-20. Problem: creating an nxnxn array from something like a text file... any ideas?

Data

• makeTwoVardata.R: Sample theoretical LC-DAD data for two compounds at two wavelengths

-- DaveAngulo - 28 Aug 2006

• PTAk_library.txt: R to load PTAk and tensor packages - load of PTAk will elicit load of tensor package