stephenturner

rm(list=ls(all=T)); source("~/.Rprofile")
library(SNPath)

# data(simDat)
# str(y); y
# str(gene.info); gene.info
# str(snp.info); head(snp.info)
# str(sim.pathway); sim.pathway
# 
# ?aligator

stephenturner

manhattan <- function(dataframe, colors=c("gray10", "gray50"), ymax="max", xaxis.cex=1, limitchromosomes=1:23, suggestiveline=-log10(1e-5), genomewideline=-log10(5e-8), annotate=NULL, ...) {

    d=dataframe
    
    #throws error if you don't have columns named CHR, BP, and P in your data frame.
	if (!("CHR" %in% names(d) & "BP" %in% names(d) & "P" %in% names(d))) stop("Make sure your data frame contains columns CHR, BP, and P")
    
	# limits chromosomes to plot. (23=x, 24=y, 25=par?, 26=mito?)
    if (any(limitchromosomes)) d=d[d$CHR %in% limitchromosomes, ]
	

stephenturner

# To source this file into an environment to avoid cluttering the global workspace, put this in Rprofile:
# my.env <- new.env(); sys.source("C:/PathTo/THIS_FILE.r", my.env); attach(my.env)

#-----------------------------------------------------------------------
#     Load packages, set options and cwd, set up database connection
#-----------------------------------------------------------------------

## Load packages
# require(ggplot2)  #plotting
# require(plyr)		#data manipulation

stephenturner

rm(list=ls(all=TRUE)); dev.off()
library(randomForest)
library(caret)

###############################################################################
############################## load functions #################################
###############################################################################

# Run randomForest keeping importance measures and the forest
RF <- function(...) randomForest(..., importance=T, keep.forest=T, na.action=na.omit)

stephenturner

#install.packages("RODBC") #Only do this once if you don't have RODBC installed
require(RODBC)
connection <- odbcConnect("localhost")
query <- function(...) sqlQuery(connection, ...)

# The cast function will cast a character string as a numeric vector between 0 and 2, or NA
castletter <- function(x) {
    x<-sub("0",NA,x)
    x<-sub("A",0.0,x)
    x<-sub("B",0.1,x)

stephenturner

# R scripts generated  Tue Nov 15 13:23:06 EST 2011
################################################################
#   Differential expression analysis with limma
library(Biobase)
library(GEOquery)
library(limma)

# load series and platform data from GEO

gset <- getGEO("GSE7442", GSEMatrix =TRUE)

stephenturner

gset <- getGEO("GSE7442", GSEMatrix =TRUE)
if (length(gset) > 1) idx <- grep("GPL5058", attr(gset, "names")) else idx <- 1
gset <- gset[[idx]]

# group names for all samples
sml <- c("X","X","X","X","X","G0","X","X","X","X","G0","X","G0","G1","G1","G0","X","X","X","X","X","X","X","X","X","X","X","X","X","X","X","X","X","G1","X","G0","X","X","X","G1","X","X","G1");

# eliminate samples marked as "X"
sel <- which(sml != "X")
sml <- sml[sel]

stephenturner

     ID logFC AveExpr   t P.Value adj.P.Val    B
7902702  -6.8     7.8 -46 1.0e-09   3.3e-05 11.0
8117594  -4.3    10.0 -33 9.6e-09   1.5e-04 10.0
8168794  -3.6     6.5 -30 1.7e-08   1.9e-04  9.7
8103736  -3.4     7.7 -28 3.1e-08   2.5e-04  9.3
7897426  -4.0     7.2 -26 4.7e-08   3.1e-04  9.0
8094278  -3.7     7.0 -24 7.4e-08   3.7e-04  8.7

stephenturner

     ID    Symbol                                   Name                                                                                             Ensembl     logFC   AveExpr         t      P.Value    adj.P.Val         B
7902702     CLCA2           chloride channel accessory 2 <a href='http://useast.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000137975'>ENSG00000137975</a> -6.832901  7.793344 -46.35621 1.015175e-09 3.281148e-05 11.177075
8117594 HIST1H2BM                histone cluster 1, H2bm <a href='http://useast.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000196374'>ENSG00000196374</a> -4.255756 10.096873 -33.21938 9.579122e-09 1.548034e-04 10.063255
8168794     CENPI                   centromere protein I <a href='http://useast.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000102384'>ENSG00000102384</a> -3.612035  6.486830 -30.41465 1.733246e-08 1.867342e-04  9.698515
8103736     SCRG1         stimulator of chondrogenesis 1 <a href='http://useast.ensembl.org/Homo_sapiens/Gene/Summary?g=ENSG00000164

stephenturner

# Install the necessary bioconductor packages. Use the appropriate annotation.db file
# for your organism/platform. See http://www.bioconductor.org/packages/release/data/annotation/
source("http://www.bioconductor.org/biocLite.R")
biocLite("affy")
biocLite("limma")
biocLite("annotate")
biocLite("hugene10sttranscriptcluster.db")
install.packages("R2HTML")

# Load required packages