CINOEDV-package: Co-Information based N-Order Epistasis Detector and...

In CINOEDV: Co-Information based N-Order Epistasis Detector and Visualizer

Description

Detecting and visualizing nonlinear interaction effects of single nucleotide polymorphisms or epistatic interactions, especially high-order epistatic interactions, are important topics in bioinformatics because of their significant mathematical and computational challenges. We present CINOEDV (Co-Information based N-Order Epistasis Detector and Visualizer) for detecting, visualizing, and analyzing high-order epistatic interactions by introducing virtual vertices into the construction of a hypergraph. CINOEDV was developed as an alternative to existing software to build a global picture of epistatic interactions and unexpected high-order epistatic interactions, which might provide useful clues for understanding the underlying genetic architecture of complex diseases.

Details

Package:	CINOEDV
Type:	Package
Version:	1.0
Date:	2014-05-02
License:	GPL-2

############################### ## Normal Using type ## ###############################

result <- CINOEDV_Main()

#### Please input the file name with its format (.mat) that saves SNP data. # test.mat

#### Please input the maximum order (1/2/3/4/5), and 3 is the Recommendation Option. # 3

#### Please input the 1 ratio threshold. # 1

#### Please input the 2 ratio threshold. # 0.5

#### Please input the 3 ratio threshold. # 0.8

#### Please input the 1 number threshold. # 10

#### Please input the 2 number threshold. # 20

#### Please input the 3 number threshold. # 8

#### Please select the evaluation measure (1/2/3), and 1 is the Recommendation Option. # 1

#### Please input the name of such file with (.mat) format. # NA

Author(s)

Junliang Shang [e-mail]

Examples

rm(list=ls())

File1 <- system.file("extdata","test.mat",package="CINOEDV")
File2 <- system.file("extdata","test1.mat",package="CINOEDV")
File3 <- system.file("extdata","test_Name.mat",package="CINOEDV")
File4 <- system.file("extdata","test1_Name.mat",package="CINOEDV")

FileName <- c(File1,File2)
MaxOrder <- 2
RThreshold <- c(1,1)
NThreshold <- c(10,10)
measure <- 1
SNFName <- c(File3,File4)
Stra <- 1
Pop <- 100
Iter <- 10

BatCINOEDV(FileName, MaxOrder, RThreshold, NThreshold, measure, Stra, Pop, Iter, SNFName)


###################
##     Second    ##
###################

# Install Packages
# InstallPackage()

# Check SNP data
FileName <- system.file("extdata","test.mat",package="CINOEDV")
Data <- InputData(FileName)
pts <- Data$pts
class <- Data$class

# Check MaxOrder
MaxOrder <- 2
TestMaxOrder(as.character(MaxOrder))
MaxOrder <- as.numeric(MaxOrder)

# Check RatioThreshold
RatioThreshold <- c(1,1)
TestRatioThreshold(MaxOrder,as.character(RatioThreshold))
RatioThreshold <- as.numeric(RatioThreshold)

# Check NumberThreshold
NumberThreshold <- c(10,10)
TestNumberThreshold(MaxOrder,as.character(NumberThreshold))
NumberThreshold <- as.numeric(NumberThreshold)

# Check measure
measure <- 1

# Check SNPNameFileName
SNPNameFileName <- system.file("extdata","test_Name.mat",package="CINOEDV")
SNPNames <- TestSNPNameFile(ncol(pts),as.character(SNPNameFileName))
SNPNames <- SNPNames$SNPNames

# Define file name which is used for saving results. 
SaveFileName <- "TEST_RESULT_"

# Search Strategies
Effect <- ExhaustiveSearch(pts,class,MaxOrder,measure,0)
SingleEffect <- Effect$SingleEffect
TwoEffect <- Effect$TwoEffect
ThreeEffect <- Effect$ThreeEffect
FourEffect <- Effect$FourEffect
FiveEffect <- Effect$FiveEffect

# Normalization
Effect <- NormalizationEffect(MaxOrder,SingleEffect,TwoEffect,ThreeEffect,FourEffect
                                ,FiveEffect,SaveFileName)
  
SingleEffect <- Effect$SingleEffect
TwoEffect <- Effect$TwoEffect
ThreeEffect <- Effect$ThreeEffect
FourEffect <- Effect$FourEffect
FiveEffect <- Effect$FiveEffect

# SNP Name Notation
Effect <- NotationName(MaxOrder,SingleEffect,TwoEffect,ThreeEffect,FourEffect
                         ,FiveEffect,SNPNames)
  
SingleEffect <- Effect$SingleEffect
TwoEffect <- Effect$TwoEffect
ThreeEffect <- Effect$ThreeEffect
FourEffect <- Effect$FourEffect
FiveEffect <- Effect$FiveEffect

# Collect Vertices and Edges
GraphData <- NetworkData(SingleEffect,TwoEffect,ThreeEffect,FourEffect,
                           FiveEffect,RatioThreshold,NumberThreshold)
Edges <- GraphData$edges
Vertices <- GraphData$vertices

# Construct Complete Graph
ConstructCompleteGraph(Vertices,Edges,6,SaveFileName)

# Plot Top Effects
TpEffect <- PlotTopEffects(Vertices,20,SaveFileName)
TopEffect <- TpEffect$TopEffect
CombinationEffect <- TpEffect$CombinationEffect

# Degree Analysis
Degrees <- DegreeAnalysis(Vertices,Edges,SaveFileName)
Degrees <- Degrees$Degrees
print(Degrees)

# Split subgraphs
SubgraphSNPs <- SubgraphSplit(Vertices,Edges)
SubgraphSNPs <- SubgraphSNPs$SubgroupSNPs

# heatmap Factor  
HeatMapFactors <- HeatMapFactor(pts,class,factor = c(5, 8),SaveFileName,Title ="")
HeatMapFactors <- HeatMapFactors$HeatMapFactors

CINOEDV documentation built on May 29, 2017, 9:07 p.m.