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

############################### ## 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 shangjunliang110@163.com

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

Results

R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(CINOEDV)
Loading required package: R.matlab
R.matlab v3.5.1 (2016-03-27) successfully loaded. See ?R.matlab for help.

Attaching package: 'R.matlab'

The following objects are masked from 'package:base':

    getOption, isOpen

Loading required package: igraph

Attaching package: 'igraph'

The following objects are masked from 'package:stats':

    decompose, spectrum

The following object is masked from 'package:base':

    union

Loading required package: ggplot2
Loading required package: reshape2
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/CINOEDV/CINOEDV-package.Rd_%03d_medium.png", width=480, height=480)
> ### Name: CINOEDV-package
> ### Title: Co-Information based N-Order Epistasis Detector and Visualizer
> ### Aliases: CINOEDV-package CINOEDV
> ### Keywords: package
> 
> ### ** 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)
    The Maximum order : 2 

   The RatioThreshold : 1 1 

   The NumberThreshold : 10 10 

 The input file is : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test.mat 
    Samples: 4000 
      Cases: 2000 
      Controls: 2000 
    SNPs: 10 

    The SNP Name File : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test_Name.mat 

#### Search Results ####

 Please waiting for the search ...
    Exhaustive Search !

   1-order effect computing ... 
elapsed 
  0.018 
   2-order effect computing ... 
elapsed 
  0.216 
#### Normalization ####
dev.new(): using pdf(file="Rplots27.pdf")
#### SNP Name Notation ####
#### Collect Vertices and Edges ####
#### Construct Complete Graph ####
dev.new(): using pdf(file="Rplots28.pdf")
#### Plot Top Effects ####
dev.new(): using pdf(file="Rplots30.pdf")
#### Degree Analysis ####
dev.new(): using pdf(file="Rplots31.pdf")
      SNP     Degree
 [1,] "SNP5"  "4"   
 [2,] "SNP4"  "3"   
 [3,] "SNP7"  "3"   
 [4,] "SNP10" "3"   
 [5,] "SNP8"  "2"   
 [6,] "SNP1"  "2"   
 [7,] "SNP6"  "2"   
 [8,] "SNP3"  "1"   
 [9,] "SNP2"  "0"   
[10,] "SNP9"  "0"   
#### Split subgraphs ####
dev.new(): using pdf(file="Rplots34.pdf")
 The input file is : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test1.mat 
    Samples: 4000 
      Cases: 2000 
      Controls: 2000 
    SNPs: 20 

    The SNP Name File : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test1_Name.mat 

#### Search Results ####

 Please waiting for the search ...
    Exhaustive Search !

   1-order effect computing ... 
elapsed 
   0.05 
   2-order effect computing ... 
elapsed 
  0.944 
#### Normalization ####
dev.new(): using pdf(file="Rplots37.pdf")
#### SNP Name Notation ####
#### Collect Vertices and Edges ####
#### Construct Complete Graph ####
dev.new(): using pdf(file="Rplots38.pdf")
#### Plot Top Effects ####
dev.new(): using pdf(file="Rplots39.pdf")
#### Degree Analysis ####
dev.new(): using pdf(file="Rplots40.pdf")
      SNP       Degree
 [1,] "rs00013" "4"   
 [2,] "rs00005" "3"   
 [3,] "rs00015" "2"   
 [4,] "rs00002" "2"   
 [5,] "rs00019" "2"   
 [6,] "rs00016" "2"   
 [7,] "rs00008" "1"   
 [8,] "rs00011" "1"   
 [9,] "rs00017" "1"   
[10,] "rs00020" "1"   
[11,] "rs00014" "1"   
[12,] "rs00010" "0"   
[13,] "rs00006" "0"   
[14,] "rs00001" "0"   
[15,] "rs00007" "0"   
#### Split subgraphs ####
dev.new(): using pdf(file="Rplots41.pdf")
############################
#####      Finish      #####
############################
> 
> 
> ###################
> ##     Second    ##
> ###################
> 
> # Install Packages
> # InstallPackage()
> 
> # Check SNP data
> FileName <- system.file("extdata","test.mat",package="CINOEDV")
> Data <- InputData(FileName)
 The input file is : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test.mat 
    Samples: 4000 
      Cases: 2000 
      Controls: 2000 
    SNPs: 10 

> pts <- Data$pts
> class <- Data$class
> 
> # Check MaxOrder
> MaxOrder <- 2
> TestMaxOrder(as.character(MaxOrder))
    The Maximum order : 2 

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

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

> 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))
    The SNP Name File : /home/ddbj/local/lib64/R/library/CINOEDV/extdata/test_Name.mat 

> SNPNames <- SNPNames$SNPNames
> 
> # Define file name which is used for saving results. 
> SaveFileName <- "TEST_RESULT_"
> 
> # Search Strategies
> Effect <- ExhaustiveSearch(pts,class,MaxOrder,measure,0)
    Exhaustive Search !

   1-order effect computing ... 
elapsed 
  0.013 
   2-order effect computing ... 
elapsed 
  0.149 
> SingleEffect <- Effect$SingleEffect
> TwoEffect <- Effect$TwoEffect
> ThreeEffect <- Effect$ThreeEffect
> FourEffect <- Effect$FourEffect
> FiveEffect <- Effect$FiveEffect
> 
> # Normalization
> Effect <- NormalizationEffect(MaxOrder,SingleEffect,TwoEffect,ThreeEffect,FourEffect
+                                 ,FiveEffect,SaveFileName)
dev.new(): using pdf(file="Rplots42.pdf")
>   
> SingleEffect <- Effect$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$vertices
> 
> # Construct Complete Graph
> ConstructCompleteGraph(Vertices,Edges,6,SaveFileName)
dev.new(): using pdf(file="Rplots43.pdf")
> 
> # Plot Top Effects
> TpEffect <- PlotTopEffects(Vertices,20,SaveFileName)
dev.new(): using pdf(file="Rplots44.pdf")
> TopEffect <- TpEffect$TopEffect
> CombinationEffect <- TpEffect$CombinationEffect
> 
> # Degree Analysis
> Degrees <- DegreeAnalysis(Vertices,Edges,SaveFileName)
dev.new(): using pdf(file="Rplots45.pdf")
> Degrees <- Degrees$Degrees
> print(Degrees)
      SNP     Degree
 [1,] "SNP5"  "4"   
 [2,] "SNP4"  "3"   
 [3,] "SNP7"  "3"   
 [4,] "SNP10" "3"   
 [5,] "SNP8"  "2"   
 [6,] "SNP1"  "2"   
 [7,] "SNP6"  "2"   
 [8,] "SNP3"  "1"   
 [9,] "SNP2"  "0"   
[10,] "SNP9"  "0"   
> 
> # Split subgraphs
> SubgraphSNPs <- SubgraphSplit(Vertices,Edges)
> SubgraphSNPs <- SubgraphSNPs$SubgroupSNPs
> 
> # heatmap Factor  
> HeatMapFactors <- HeatMapFactor(pts,class,factor = c(5, 8),SaveFileName,Title ="")
dev.new(): using pdf(file="Rplots46.pdf")
> HeatMapFactors <- HeatMapFactors$HeatMapFactors
> 
> 
> 
> 
> 
> 
> dev.off()
pdf 
  2 
>