11 Large bulk RNA Seq data (TCGA)

How to analyze large bulk RNA-Seq data (TCGA)

In this example the samples are normalized using DESeq2 so no noramalizaion is needed.

sample.file.url = "https://genome.med.nyu.edu/results/external/iCellR/data/TCGA_sample_Normalized_data.tsv.gz"

download.file(url = sample.file.url, 
     destfile = "TCGA_sample_Normalized_data.tsv.gz", 
     method = "auto")  

TCGA.data <- read.table("TCGA_sample_Normalized_data.tsv.gz")
head(TCGA.data)[1:3]
#         Basal_TCGA.A1.A0SK.txt Basal_TCGA.A1.A0SP.txt Basal_TCGA.A2.A04P.txt
#TSPAN6                5823.4300            4318.034382            5265.733258
#TNMD                     0.0000               6.049079               6.763079
#DPM1                  3248.1536            2528.515113            1183.538813
#SCYL3                 1059.7135             965.836315            1109.144945
#C1orf112              1251.3155            1070.687022             485.589067
#FGR                    106.2438             933.574559             512.641383

library(iCellR)
my.obj <- make.obj(TCGA.data)

my.obj@main.data <- my.obj@raw.data

my.obj
###################################
,--. ,-----.       ,--.,--.,------.
`--''  .--./ ,---. |  ||  ||  .--. '
,--.|  |    | .-. :|  ||  ||  '--'.'
|  |'  '--'\   --. |  ||  ||  |
`--' `-----' `----'`--'`--'`--' '--'
###################################
An object of class iCellR version: 1.2.4
Raw/original data dimentions (rows,columns): 69797,882
Data conditions in raw data: Basal,Her2,LumA,LumB,Normal (131,64,404,170,113)
Row names: TSPAN6,TNMD,DPM1 ...
Columns names: Basal_TCGA.A1.A0SK.txt,Basal_TCGA.A1.A0SP.txt,Basal_TCGA.A2.A04P.txt ...
###################################
   QC stats performed:FALSE, PCA performed:FALSE, CCA performed:FALSE
   Clustering performed:FALSE, Number of clusters:0
   tSNE performed:FALSE, UMAP performed:FALSE, DiffMap performed:FALSE
   Main data dimentions (rows,columns):69797,882
   Normalization factors:,...
   Imputed data dimentions (rows,columns):0,0
############## scVDJ-Seq ###########
VDJ data dimentions (rows,columns):0,0
############## CITE-Seq ############
   ADT raw data dimentions (rows,columns):0,0
   ADT main data dimentions (rows,columns):0,0
   ADT columns names:...
   ADT row names:...
########### iCellR object ##########


my.obj <- run.pca(my.obj)

my.obj <- run.clustering(my.obj, 
	clust.method = "kmeans", 
	dist.method = "euclidean",
	index.method = "silhouette",
	max.clust =25,
	min.clust = 2,
	dims = 1:10)

my.obj <- run.pc.tsne(my.obj, dims = 1:10)
my.obj <- run.umap(my.obj, dims = 1:10, method = "umap-learn") 

cluster.plot(my.obj,plot.type = "pca",cell.color = "black",col.by = "conditions",cell.transparency = 0.5,interactive = F)
cluster.plot(my.obj,plot.type = "umap",cell.color = "black",col.by = "conditions",cell.transparency = 0.5,interactive = F)
cluster.plot(my.obj,plot.type = "tsne",cell.color = "black",col.by = "conditions",cell.transparency = 0.5,interactive = F)
cluster.plot(my.obj,plot.type = "umap",cell.color = "black",cell.transparency = 1,interactive = F)