deco-package: DECO: DEcomposing heterogeneous Cohorts from Omic profiling.

Description Details Author(s) References See Also Examples

Description

DECO integrates a two-step methodology to find out outlier behavior of features which could determine a new subclass of samples. Thus, a subsampling method with LIMMA (Stratified Differential Analysis or RDA) will be applied to a omic matrix data, previously normalized. After that, Non-Symmetrical Correspondence Analysis (NSCA) will be computed on frequency matrix of differential events generated by subsampling.

The feature-sample relationships will define stratified profiles with statistical significance. Any new subclass defined could be associated to hidden phenotypic variables.

Details

Package: deco
Type: Package
Version: 0.99
Date: 2018-11-15
License: GPL 3.0

Author(s)

Francisco Jose Campos Laborie, Jose Manuel Sanchez Santos and Javier De las Rivas.

Maintainer: Francisco Jose Campos Laborie <fjcamlab@usal.es>, Cancer Research Centre (Salamanca) <jrivas@usal.es>

References

Campos-Laborie, FJ et al. DECO: decompose heterogeneous population cohorts for patient stratification and discovery of sample biomarkers using omic data profiling. (2018)

Scarfo, Irene et al. Identification of a new subclass of ALK negative ALCL expressing aberrant levels of ERBB4 transcripts. Blood (2015). http://dx.doi.org/10.1182/blood-2014-12-614503.

See Also

decoRDA, decoNSCA, decoReport, voom

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
#### ALCL EXAMPLE (Scarfo et al., 2015. Blood) ####
## Group-VS-group comparison

########################
# Loading example data #
########################
# Data from two subtypes (ALK+ and ALK-) of Anaplastic Large Cell Leukemia (ALCL).
data(ALCLdata)

## Classes vector to run a supervised analysis to compare both classes.
classes.ALCL <- colData(ALCL)[,"Alk.positivity"]
names(classes.ALCL) <- colnames(ALCL)

####################################
# Parallelization via BiocParallel #
####################################
# Non-parallel computing
bpparam <- SerialParam()

# Computing in shared memory
# all cores by deault
bpparam <- MulticoreParam()

#######################################################################
# RUNNING SUBSAMPLING OF DATA: BINARY design (two classes of samples) #
#######################################################################
# if annotation and rm.xy == TRUE, then
# library(Homo.sapiens)

# Not run as example
# sub.ma.3r.1K <- decoRDA(data = assay(ALCL), classes = classes.ALCL, q.val = 0.01,
#                rm.xy = TRUE, r = NULL, control = "pos", annot = FALSE, bpparam = bpparam,
#                id.type = "ENSEMBL", iterations = 10000, pack.db = "Homo.sapiens")


#########################################################################################
# RUNNING NSCA STEP: Looking for subclasses within a category/class of samples compared #
#########################################################################################
# Not run as example
# deco.results.ma <- decoNSCA(sub = sub.ma.3r.1K, v = 80, method = "ward.D", bpparam = bpparam,
#                        k.control = 3, k.case = 3, samp.perc = 0.05, rep.thr = 10)


# Phenotypical data from TCGA RNAseq samples.
colData(ALCL)

########################################################
# PDF report with feature-sample patterns or subgroups #
########################################################
## Generate PDF report with relevant information and several plots.

## Binary example (ALK+ vs ALK-) -not run as example-
# decoReport(deco.results.ma, sub.ma.3r.1K,
#           pdf.file = "report_example_microarray_binary.pdf",
#           info.sample = as.data.frame(colData(ALCL)[,8:10]),
#           cex.names = 0.3, print.annot = TRUE)

deco documentation built on Nov. 8, 2020, 7:45 p.m.