whirlsyu/enmcb

Creation of the correlated blocks using DNA methylation profiles. A stacked ensemble of machine learning models, which combined the support vector machine and elastic-net regression model, can be constructed to predict disease progression.

EnMCB

Package: EnMCB

Type: Package

Title: Predicting Disease Progression Based on Methylation Correlated Blocks using Ensemble Models

Version: 1.11.1

Author: Xin Yu

Maintainer: Xin Yu [email protected] [email protected]

Description: This package is designed to help you to create the methylation correlated blocks using methylation profiles. A stacked ensemble of machine learning models, which combined the Cox regression, support vector regression, Coxboost and elastic-net regression model, can be constructed using this package. You also can choose one of them to build DNA methylation signatures associated with disease progression (survival).

License: GPL-3

Citation: Xin Yu, De-Xin Kong, EnMCB: an R/bioconductor package for predicting disease progression based on methylation correlated blocks using ensemble models, Bioinformatics, 2021, btab415

Followings are brief instructions for using this package:

You can install and our package via BiocManager as following or downloading source from github.

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("EnMCB")

First, you need a methylation data set, currently only most common platform 'Illumina Infinium Human Methylation 450K' is supported. For other platform, for instance, BS sequencing data, users need to introduce the annonation file for other own.

You can use your own datasets:

eset_met<-some_methylation_datamatrix

or use our demo data.

data('demo_data',package = "EnMCB")
methylation_dataset<-demo_data$realdata

Then, you can automatically run following:

library(SummarizedExperiment)

res<-IdentifyMCB(methylation_dataset)

You can extract the MCB information,

MCB<-res$MCBinformation

and select some of MCBs for further modeling.

MCB<-MCB[MCB[,"CpGs_num"]>5,]

In order to get differentially methylated blocks, one may run following:

#simulation for the group data
groups = c(rep("control",200),rep("dis",255))

diffMCB_results <- DiffMCB(methylation_dataset, groups, MCB)

In order to build survival models, one may run following:

data(demo_survival_data)
survival_data <- demo_survival_data
# sample the dataset into training set and testing set
trainingset<-colnames(methylation_dataset) %in% sample(colnames(methylation_dataset),0.6*length(colnames(methylation_dataset)))

testingset<-!trainingset

#build the models

models<-metricMCB(MCB,
                    training_set=methylation_dataset[,training_set],
                    Surv=survival_data)

#select the best
onemodel<-models$best_cox_model

Then, you can predict the risk by the model you build:

predict(onemodel,methylation_dataset[,testing_set])

In order to build ensemble model, one may run following:

# You can choose one of MCBs:
select_single_one=1

em<-ensemble_model(MCB[select_single_one,],
                    training_set=methylation_dataset[,training_set],
                    Surv=survival_data)
                    

Note that this function only can be used for single MCB only, otherwise the precessing time could be very long.

Then, you can predict the risk by the model you build:

ensemble_prediction(ensemble_model = em,
                    predition_data = methylation_dataset[,testing_set])

For detailed information, you can find at our references.