.Rhistory

intersect(MVM_DEGs$U,VSV_DEGs$D)
MVM_resist <- union(MVM_DEGs$U,VSV_DEGs$D)
MVM_resist <- intersect(MVM_DEGs$U,VSV_DEGs$D)
debugSource('~/Desktop/FDA meeting/MVM_susceptible.R')
debugSource('~/Desktop/FDA meeting/MVM_susceptible.R')
RNA_resist <- intersect(VSV_DEGs$D,EMCV_DEGs$D)
log2(1.5)
log2(-1.5)
log2(0.5)
load('~/Desktop/16_FDA_virus/BatchEffect_Results/MVM_DEG.rda')
rm(list=ls())
load('~/Desktop/16_FDA_virus/BatchEffect_Results/VSV_DEG.rda')
rm(list=ls())
load('~/Desktop/16_FDA_virus/BatchEffect_Results/VSV_DEG.rda')
VSV_DEG <- DEG_IDs$DEseq2_7_mouse
load('~/Desktop/16_FDA_virus/BatchEffect_Results/EMCV_DEG.rda')
EMCV_DEG <- DEG_IDs$DEseq2_7_mouse
load('~/Desktop/16_FDA_virus/BatchEffect_Results/REO_DEG.rda')
REO_DEG <- DEG_IDs$DEseq2_7_mouse
load('~/Desktop/16_FDA_virus/BatchEffect_Results/MVM_DEG.rda')
MVM_DEG <- DEG_IDs$DEseq2_7_mouse
debugSource('~/Desktop/FDA meeting/MVM_susceptible.R')
RNA_resist <- intersect(VSV_DEGs$D,EMCV_DEGs$D)
RNA_resist <- intersect(RNA_resist,REO_DEGs$D)
source('~/Desktop/FDA meeting/MVM_susceptible.R')
source('~/Desktop/FDA meeting/MVM_susceptible.R')
source('~/Desktop/FDA meeting/MVM_susceptible.R')
load('~/Desktop/16_FDA_virus/BatchEffect_Results/FDA_DEG.rda')
load('~/Desktop/16_FDA_virus/BatchEffect_Results/FDA_DEG.rda')
load('~/Desktop/16_FDA_virus/BatchEffect_Results/FDA_DE.rda')
rm(list=ls())
load('~/Desktop/16_FDA_virus/BatchEffect_Results/FDA_DE.rda')
WriteXLS(DiffExp,'~/Desktop/FDA_DE.xlsx')
library(WriteXLS)
WriteXLS(DiffExp,'~/Desktop/FDA_DE.xlsx')
?WriteXLS
WriteXLS(DiffExp,'~/Desktop/FDA_DE.xls')
FDA_DE <- DiffExp
FDA_DE$VSV <- as.data.frame(FDA_DE$VSV)
FDA_DE$EMCV <- as.data.frame(FDA_DE$EMCV)
FDA_DE$REO <- as.data.frame(FDA_DE$REO)
FDA_DE$MVM <- as.data.frame(FDA_DE$MVM)
F_vsv <- FDA_DE$VSV
View(F_vsv)
FDA_DE <- list()
FDA_DE$VSV_FC <- as.data.frame(DiffExp$VSV$FoldChange)
FDA_DE$EMCV_FC <- as.data.frame(DiffExp$EMCV$FoldChange)
FDA_DE$REO_FC <- as.data.frame(DiffExp$REO$FoldChange)
FDA_DE$MVM_FC <- as.data.frame(DiffExp$MVM$FoldChange)
FDA_DE$VSV_P <- as.data.frame(DiffExp$VSV$Sigp)
FDA_DE$EMCV_P <- as.data.frame(DiffExp$EMCV$Sigp)
FDA_DE$REO_P <- as.data.frame(DiffExp$REO$Sigp)
FDA_DE$MVM_P <- as.data.frame(DiffExp$MVM$Sigp)
F_vsv <- FDA_DE$VSV_FC
F_vsv <- FDA_DE$VSV_P
WriteXLS(FDA_DE,'~/Desktop/FDA_DE.xls')
WriteXLS(FDA_DE,'~/Desktop/FDA_DE.xlsx')
WriteXLS(FDA_DE,'~/Desktop/FDA_DE.xlsx', row.names = TRUE, col.names = TRUE)
q()
q()
q()
-log10(0.05)
install.packages("clValid")
library(clValid)
data(mouse)
express <- mouse[1:25,c("M1","M2","M3","NC1","NC2","NC3")]
rownames(express) <- mouse$ID[1:25]
intern <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="internal")
summary(intern)
optimalScores(intern)
plot(intern)
stab <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="stability")
optimalScores(stab)
plot(stab)
fc <- fc[-match( c("EST","Unknown"), names(fc))]
bio <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="biological", annotation=fc)
optimalScores(bio)
plot(bio)
fc <- tapply(rownames(express),mouse$FC[1:25], c)
fc <- fc[-match( c("EST","Unknown"), names(fc))]
bio <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="biological", annotation=fc)
optimalScores(bio)
plot(bio)
if(require("Biobase") && require("annotate") && require("GO.db") && require("moe430a.db")) {
bio2 <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="biological",annotation="moe430a.db",GOcategory="all")
optimalScores(bio2)
plot(bio2)
}
if(require("Biobase") && require("annotate") && require("GO.db") && require("moe430a.db")) {
bio2 <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="biological",annotation="moe430a.db",GOcategory="all")
optimalScores(bio2)
plot(bio2)
}
plot(bio2)
bio2 <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="biological",annotation="moe430a.db",GOcategory="all")
plot(bio2)
?clValid
summary(intern)
rownames(express) <- mouse$ID[1:25]
intern <- clValid(express, 2:6, clMethods=c("hierarchical","kmeans","pam"),
validation="internal")
intern@measures
intern@measNames
intern@clMethods
intern@metric
intern@call
View(express)
intern@nClust
intern@clusterObjs
ff <- clValid(express, nClust, clMethods = "hierarchical", validation =
"stability", maxitems = 600, metric = "euclidean", method = "average",
neighbSize = 10, annotation = NULL, GOcategory = "all",
goTermFreq=0.05, dropEvidence=NULL, verbose=FALSE, ...)
ff <- clValid(express, nClust, clMethods = "hierarchical", validation =
"stability", maxitems = 600, metric = "euclidean", method = "average",
neighbSize = 10, annotation = NULL, GOcategory = "all",
goTermFreq=0.05, dropEvidence=NULL, verbose=FALSE)
ff <- clValid(express, 2:6, clMethods = "hierarchical", validation =
"stability", maxitems = 600, metric = "euclidean", method = "average",
neighbSize = 10, annotation = NULL, GOcategory = "all",
goTermFreq=0.05, dropEvidence=NULL, verbose=FALSE)
summary(ff)
log10(0.05)
install.packages("devtools")
library(devtools)
install_github("bkellman/RegressionModelPipeline")
install_github("bkellman/RegressionModelPipeline")
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
library(RegressionModelPipeline)
setwd("~/GitHub/RegressionModelPipeline/")
iris_test<-function(){
print(str(iris))
}
mtcars_tests <- function(){
#	source('~/Desktop/modeling_functions.r')
K=5
inter=NULL
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm)
out=vis(mod)
out[[1]]
out[[2]]
dev.off()
sig_vars_thresh = list(model_sel_interaction=6,model_sel_additive=7,glmnet_interaction=200,glmnet_additive=2000) # see regularization output (lower upper bound for additive model selection)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm,sig_vars_thresh=sig_vars_thresh)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',family='poisson',K=K,thresh_screen = .05)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='Wald',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',only_return_selected=FALSE,K=K)
mtcars$log_mpg = log(mtcars$mpg)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'log_mpg',interactions=inter,test='LRT',K=K)
}
glmnet_test<-function(){
sig_vars_thresh = list(model_sel_interaction=1,model_sel_additive=1,glmnet_interaction=1,glmnet_additive=1e5)
robust=TRUE
}
### main
#' model_selection
#'
#' Main function for model selection from a set of many variables
#' @param df, a data.frame containing response and observations variables. Factors with more than 2 levels have only been implimented for test='LRT'
#' @param observations, a character vector of the names of independent/observations variables in df
#' @param response, a character vector of the names of dependent/response variables in df
#' @param family, a character string indicating the family associated with the submitted model c('gaussian','binomial','poisson'...)
#' @param model, a model associated for testing the variables c(glm,lm)
#' @param interactions, a boolean indicating if interactions should be assessed. Default is NULL, by default, interactions will be examined according to the constraints set by sig_vars_thresh. If a value is set for interactions (T/F) this will override the recomendations of sig_vars_thresh
#' @param test, a character string indicating Likelihood Ratio Test ('LRT') testing likelihood improvement of a model or Wald test ('Wald') testing coefficient > 0
#' @param thresh_screen, a numeric value indicating the p-value cutoff for the univariate screening
#' @param only_return_selected, a boolean value. If true, only models with p-value less than the threshold will be returned. Otherwise, all models will be returned.
#' @param K, a numeric value indicating the number of folds to use for k-fold cross-validation. K=10 by default. K=0 to skip k-fold validation.
#' @param sig_vars_thresh a list specifying the maximal number of significant variables allowed for each final model generating method. NULL (self initializing) by default.
#' @param robust boolean indicating if regularization will be run multiple times to get a robust indication of the underlying structure
#' @param N, a numeric value, default N=1, indicating the number of cross validation iterations to perform
#' @return a list containing: univariate models, the final selected model, and crossvalidation stats.
#' @export
#' @examples
#' mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=F,test='LRT',K=5,family = 'gaussian',model=glm)
#' out=vis(mod)
#' print(out[[1]])
#' print(out[[2]])
#' @import glmnet
#' @import glinternet
model_selection <- function(df,observations,response,family='gaussian',model=glm,interactions=FALSE,test=c('Wald','LRT'),thresh_screen=.2,only_return_selected=FALSE,K=10,sig_vars_thresh=NULL,robust=FALSE,N=1){
if(length(response)!=1){stop('use multiresponse_model_selection()')}
if(!test%in%c('Wald','LRT')){stop("test is not in c(Wald,LRT)")}
#if(!interactions%in%c('signif','none','all')){stop("interactions is not in c(signif,none,all)")}
if(is.null(sig_vars_thresh)){
sig_vars_thresh = list(model_sel_interaction=6,model_sel_additive=25,glmnet_interaction=200,glmnet_additive=2000)
}
# Univariate Screen
observationsL <<- univariate_screen(df,observations,response,family,model,interactions,test,thresh=thresh_screen,only_return_selected=only_return_selected)
# Multivariate Model
obs_sign = na.omit( names(observationsL)[attr(observationsL,'Pr')<thresh_screen] )
if( length( obs_sign ) == 0 ){ stop("There are no observations that pass the threshold set by thresh_screen. Consider increasing thresh_screen.")}
# set interaction based on the number of variables passing the screening threshold
if(is.null(interactions)){
if(length(obs_sign) < sig_vars_thresh$model_sel_interaction){
interactions=TRUE
print('interactive model: selection')
}else if(length(obs_sign) < sig_vars_thresh$model_sel_additive){
interactions=FALSE
print('additive model: selection')
}else if(length(obs_sign) < sig_vars_thresh$glmnet_interaction){
#glm_reg = glinternet ## not yet implimented
glm_reg = cv.glmnet
print('interactive model: regularization')
}else if(length(obs_sign) < sig_vars_thresh$glmnet_additive){
glm_reg = cv.glmnet
print('additive model: regularization')
}
}
# construct multivariate models
if(length(obs_sign) < sig_vars_thresh$model_sel_additive){
selected_model = stepwise_multivariate_model_selection(df,obs_sign,response,family,model,interactions)
}else{
if(robust){ # run glmnet several times
selected_model_list = list()
for(i in 1:500){
indx = sample(1:nrow(df)) # randomize
indy = sample(1:length(obs_sign)) # randomize
selected_model_list[[i]] = glm_reg(y=df[indx,response],x=data.matrix(df[indx,obs_sign[indy]]),family=family) ## untested
}
}else{
selected_model = glm_reg(y=df[,response],x=data.matrix(df[,obs_sign]),family=family) ## untested
}
}
if(length(obs_sign) > sig_vars_thresh$glmnet_additive){
warning(paste(length(obs_sign),'is a large number of variables that may pose a relative challenge to glmnet'))
}
# Model Diagnostics
# Cross Validation
if(length(obs_sign) < sig_vars_thresh$model_sel_additive){
if(K>1){
cv=cross_assess_wrapper(data=df,formula=selected_model$formula,resp=response,family=family,K,model,cv_function=cross_valid_kfold,N=N)
}else{
cv=NULL
}
}else{
if(robust){
# multi-model characterization
selected_model = vis_reg(selected_model_list)
}else{
# cross validation for regularization
plot(selected_model)
cv = list(auc=selected_model$cvm[selected_model$lambda==selected_model$lambda.min],other_stats=selected_model)
}
}
return(list(screen=observationsL,final=selected_model,cv=cv))
}
#' model_selection_many_to_few
#'
#' runs regularization then model selection to reduce many variables to a small robust model TODO: Austin and Ben w/ vis_reg
#' @param ..., see model_select parameters
#' @return a list containing: univariate models, the final selected model, and crossvalidation stats.
#' @export
model_selection_many_to_few <- function(...){
# run regularization
# (optional) run interaction regularization
# identify regularization prototype models
# for each prototypical model:
# while not overfit:
# select highest magnitude contributors
# univariate screening and backward selection
# return prototypical models
}
#######################################
### Run Tests
iris_test<-function(){
print(str(iris))
}
mtcars_tests <- function(){
#	source('~/Desktop/modeling_functions.r')
K=5
inter=NULL
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm)
out=vis(mod)
out[[1]]
out[[2]]
dev.off()
sig_vars_thresh = list(model_sel_interaction=6,model_sel_additive=7,glmnet_interaction=200,glmnet_additive=2000) # see regularization output (lower upper bound for additive model selection)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm,sig_vars_thresh=sig_vars_thresh)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',family='poisson',K=K,thresh_screen = .05)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='Wald',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',only_return_selected=FALSE,K=K)
mtcars$log_mpg = log(mtcars$mpg)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'log_mpg',interactions=inter,test='LRT',K=K)
}
glmnet_test<-function(){
sig_vars_thresh = list(model_sel_interaction=1,model_sel_additive=1,glmnet_interaction=1,glmnet_additive=1e5)
robust=TRUE
}
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
source("R/univariate_screen.r")
source("R/cross_validation.r")
source("R/multivariate_model.r")
source("R/vis.r")
source("R/performance_assessment.r")
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
source('~/GitHub/RegressionModelPipeline/R/RegressionModelPipeline.R')
source("R/multivariate_model.r")
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
source("R/multivariate_model.r")
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
library(MASS)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',test='LRT',K=5,family = 'gaussian',model=glm)
out[[1]]
out[[2]]
K=5
inter=NULL
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm)
out=vis(mod)
sig_vars_thresh = list(model_sel_interaction=6,model_sel_additive=7,glmnet_interaction=200,glmnet_additive=2000) # see regularization output (lower upper bound for additive model selection)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm,sig_vars_thresh=sig_vars_thresh)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',family='poisson',K=K,thresh_screen = .05)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='Wald',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=K)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',only_return_selected=FALSE,K=K)
mtcars$log_mpg = log(mtcars$mpg)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'log_mpg',interactions=inter,test='LRT',K=K)
sig_vars_thresh = list(model_sel_interaction=6,model_sel_additive=7,glmnet_interaction=200,glmnet_additive=2000) # see regularization output (lower upper bound for additive model selection)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',K=5,family = 'gaussian',model=glm,sig_vars_thresh=sig_vars_thresh)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='LRT',family='poisson',K=K,thresh_screen = .05)
mod=model_selection(df=mtcars,colnames(mtcars)[-1],response = 'mpg',interactions=inter,test='Wald',K=K)
rm(list=ls())
setwd("~/GitHub/RegressionModelPipeline/")
library(devtools)
library(MASS)
library(ggplot2)
library(fastcluster)
library(glmnet)
library(reshape)
library(data.table)
library(easyGgplot2)
source("R/vis.r")
source("R/RegressionModelPipeline.R")
source("R/univariate_screen.r")
source("R/performance_assessment.r")
source("R/cross_validation.r")
source("R/multivariate_model.r")
load(paste0('~/Desktop/FDA_Austin/VirusRegModels_run2.Rda'))
load(paste0('~/Desktop/2017 UCSD/Project02-FDAvirus/FDA_Austin/VirusRegModels_run2.Rda'))
mod <- PredResult[['mod']]
visResults <- vis_reg(mod,k=4)
visResults$p1
debugSource('~/GitHub/RegressionModelPipeline/R/test_vis.R')
p1
p2
p3
p4
visResults$p1
ggplot2.multiplot(visResults$p2,visResults$p3, cols=2)
ggplot2.multiplot(visResults$p3,visResults$p4, cols=2)
if(!require(devtools)) install.packages("devtools")
devtools::install_github("kassambara/factoextra")
pkgs <- c("cluster", "fpc", "NbClust")
install.packages(pkgs)
install.packages(pkgs)
library(factoextra)
library(cluster)
library(fpc)
library(NbClust)
data(iris)
head(iris)
nb <- NbClust(iris.scaled, distance = "euclidean", min.nc = 2,
max.nc = 10, method = "complete", index ="all")
iris.scaled <- scale(iris[, -5])
nb <- NbClust(iris.scaled, distance = "euclidean", min.nc = 2,
max.nc = 10, method = "complete", index ="all")
fviz_nbclust(nb) + theme_minimal()
km.res <- eclust(iris.scaled, "kmeans", k = 3,
nstart = 25, graph = FALSE)
km.res$cluster
View(iris.scaled)
fviz_cluster(km.res, geom = "point", frame.type = "norm")
pam.res <- eclust(iris.scaled, "pam", k = 3, graph = FALSE)
pam.res$cluster
fviz_cluster(pam.res, geom = "point", frame.type = "norm")
res.hc <- eclust(iris.scaled, "hclust", k = 3,
method = "complete", graph = FALSE)
head(res.hc$cluster, 15)
fviz_dend(res.hc, rect = TRUE, show_labels = FALSE)
fviz_dend(res.hc, rect = TRUE, show_labels = TRUE)
plot(sil, main ="Silhouette plot - K-means")
sil <- silhouette(km.res$cluster, dist(iris.scaled))
head(sil[, 1:3], 10)
plot(sil, main ="Silhouette plot - K-means")
fviz_silhouette(sil)
si.sum <- summary(sil)
si.sum$clus.avg.widths
si.sum$avg.width
si.sum$clus.sizes
head(silinfo$widths[, 1:3], 10)
head(sil$widths[, 1:3], 10)
silinfo <- km.res$silinfo
names(silinfo)
head(silinfo$widths[, 1:3], 10)
dd <- dist(iris.scaled, method ="euclidean")
km_stats <- cluster.stats(dd,  km.res$cluster)
km_stats$within.cluster.ss
km_stats
species <- as.numeric(iris$Species)
clust_stats <- cluster.stats(d = dist(iris.scaled),
species, km.res$cluster)
clust_stats$corrected.rand
clust_stats$vi
table(iris$Species, pam.res$cluster)
table(iris$Species, res.hc$cluster)
cluster.stats(d = dist(iris.scaled),
species, res.hc$cluster)$vi
d = dist(iris.scaled)
?dist
fviz_dist(d)
source("https://bioconductor.org/biocLite.R")
biocLite("TCGAbiolinks")
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
library(TCGAbiolinks)
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
library('TCGAbiolinks')
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
source("https://bioconductor.org/biocLite.R")
biocLite("TCGAbiolinks")
library('TCGAbiolinks')
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
source("https://bioconductor.org/biocLite.R")
biocLite("TCGAbiolinks")
source("https://bioconductor.org/biocLite.R")
biocLite("TCGAbiolinks")
library('TCGAbiolinks')
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
source("https://bioconductor.org/biocLite.R")
source("https://bioconductor.org/biocLite.R")
biocLite()                  ## R version 3.0 or later
biocLite("TCGAbiolinks")
library('TCGAbiolinks')
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
library(TCGAbiolinks)
query <- GDCquery(project = "TARGET-AML",
data.category = "Transcriptome Profiling",
data.type = "Gene Expression Quantification",
workflow.type = "HTSeq - Counts")
devtools::install_github("BioinformaticsFMRP/TCGAbiolinks")
install.packages("ComplexHeatmap")
biocLite("ComplexHeatmap")
devtools::install_github("BioinformaticsFMRP/TCGAbiolinks")
biocLite("ComplexHeatmap")
source("https://bioconductor.org/biocLite.R")
biocLite("ComplexHeatmap")
devtools::install_github("BioinformaticsFMRP/TCGAbiolinks")
source("https://bioconductor.org/biocLite.R")
biocLite("ComplexHeatmap")
library(ComplexHeatmap)
devtools::install_github("BioinformaticsFMRP/TCGAbiolinks")
devtools::install_github(repo = "BioinformaticsFMRP/TCGAbiolinks")
remove.packages("BiocInstaller")
remove.packages("BiocInstaller")
source("http://bioconductor.org/biocLite.R")
biocLite()    ## to update old packages
biocLite("TCGAbiolinks")
library(TCGAbiolinks)
clin <- GDCquery_clinic("TCGA-ACC", type = "clinical", save.csv = TRUE)
mut <- GDCquery_Maf(tumor = "ACC")
library(TCGAbiolinks)
mut <- GDCquery_Maf(tumor = "ACC")
clin <- GDCquery_clinic("TCGA-ACC","clinical")
clin <- clin[,c("bcr_patient_barcode","disease","gender","tumor_stage","race","vital_status")]
TCGAvisualize_oncoprint(mut = mut, genes = mut$Hugo_Symbol[1:20],
filename = "onco.pdf",
annotation = clin,
color=c("background"="#CCCCCC","DEL"="purple","INS"="yellow","SNP"="brown"),
rows.font.size=10,
heatmap.legend.side = "right",
dist.col = 0,
label.font.size = 10)
query <- GDCquery(project = "TCGA-ACC",
data.category = "Copy Number Variation",
data.type = "Copy Number Segment",
barcode = c( "TCGA-OR-A5KU-01A-11D-A29H-01", "TCGA-OR-A5JK-01A-11D-A29H-01"))
GDCdownload(query)
data <- GDCprepare(query)
biocLite("TCGAbiolinks")
library(TCGAbiolinks)
clin <- GDCquery_clinic("TCGA-ACC", type = "clinical", save.csv = TRUE)
devtools::install_github(repo = "BioinformaticsFMRP/TCGAbiolinks")
print(.libPaths())
print(sessionInfo())
print(version)
q()