CEPICS: CEPICS: a comparison and evaluation platform for integration...

In GaoLabXDU/CEPICS: CEPICS: a comparison and evaluation platform for integration methods in cancer subtyping

Description

The main function of CEPICS R package Please refer to examples for details. When running CEPICS, you don't need to initialize the MATLAB server since CEPICS will initialize it automatically.

Usage

CEPICS(datalist, datatype, functionList = list("PINS", "LRA", "SNF",
  "iCluster"), kMax = 5, Res = NULL, KMdata = "default",
  trueLabel = NULL, SNFPara = NULL, LRAPara = NULL,
  iClusterPara = NULL, PINSPara = NULL, filename = "CEPICS",
  title = "Evaluation and Comparison Report", author = "", cores = 0)

Arguments

datalist	A list of different omics data. Each data should be formatted as a data matrix with rows representing features and columns representing samples.
datatype	A list of string means the corresponding datatype of datalist. Each type can choose from "binary", "gaussian" or "poisson".
functionList	A list of functions that need to be ran, choosing from "iCluster" (means iClusterBayes), "SNF", "LRA" (means LRAcluster), "PFA", and "PINS". We set SNF, LRA, iClusterBayes and PINS as default.
kMax	An integer value means the maximize number of clusters we will try.Res A list contains the result of user's method. Please see Res for more details.
Res	User's result.Put your data into Res$data, and put name into Res$name, put the time you cost into Res$time. If this parameter is not null, this data will appear at the top of the report.
KMdata	The survival data of the sample. If this parameter is "default", CEPICS will try to use built-in data including BRCA, COAD, KIRC, LUAD and LUSC downloaded from TCGA. If this parameter is FALSE, CEPICS will not to calculate the KM curve.
trueLabel	A list of the true labels of the samples. If this parameter is null, CEPICS will compare different metrices between each two methods. Otherwise the results of each methods will be compared with the true labels. Please see turelabel for more details.
SNFPara	Parameters of SNF. Should be a list of following parameters, SNFPara$k(default is one tenth of the sample size), SNFPara$t(default is 20). Please see runSNF for more details. If the parameter is null, CEPICS will use default values.
LRAPara	Parameters of LRA. Should be a list of following parameter, LRAPara$maxdimension (default is 10) is the max number of dimension CEPICS will try to reduce. If the parameter is null, CEPICS will use default values.
iClusterPara	Parameters of iClusterBayes. Should be a list of following parameters, iClusterPara$n.burnin(default is 1000), iClusterPara$n.draw(default is 1200), iClusterPara$prior.gamma(default is rep(0.1,6)), iClusterPara$sdev(default is 0.5), iClusterPara$beta.var.scale(default is 1), iClusterPara$thin(default is 1), iClusterPara$pp.cutoff(default is 0.5). Please see iClusterPlus R package for more details. If the parameter is null, CEPICS will use default values.
PINSPara	Parameters of PINS. Should be a list of following parameter, PINSPara$agreementCutoff(default is 0.5) means agreement threshold to be considered consistent. If the parameter is null, CEPICS will use default values.
filename	The file name of the report.
title	The title of the report.
author	The author of thereport.
cores	An integer value means the number of cores for parallel computing, default 0. When using the default value, CEPICS will detect the physical cores and set cores-1 as the number of cores to parallel compute.

Value

Output the evaluation and comparion report and some experimental data files.

Examples

data(COAD_Methy)
data(COAD_miRNA)
data(COAD_mRNA)


datalist <- list(COAD_Methy,COAD_miRNA,COAD_mRNA)


CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'), kMax=5)

#Use Survival data
data(KMSURDATA)
CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'), kMax=5, KMdata = KMSURDATA)

#Use your result
data(Res)

CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'), kMax=5, Res = Res)

#Use gold standard
data(truelabel)
CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'), kMax=5, trueLabel = truelabel)

#Pass parameter to function
CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'),
    kMax=5, SNFPara = list(k = 8, t = 20))

CEPICS(datalist, datatype = c("gaussian", "gaussian", "gaussian"),
    functionList = list('PINS', 'LRA', 'SNF', 'iCluster', 'PFA'),
    kMax=5, LRAPara = list(maxdimension=10))

GaoLabXDU/CEPICS documentation built on June 9, 2020, 2:31 a.m.