|
doppelgangR.Rcheck/tests_i386/maintest.Rout
R version 3.6.3 (2020-02-29) -- "Holding the Windsock"
Copyright (C) 2020 The R Foundation for Statistical Computing
Platform: i386-w64-mingw32/i386 (32-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(RUnit)
> library(Biobase)
Loading required package: BiocGenerics
Loading required package: parallel
Attaching package: 'BiocGenerics'
The following objects are masked from 'package:parallel':
clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
clusterExport, clusterMap, parApply, parCapply, parLapply,
parLapplyLB, parRapply, parSapply, parSapplyLB
The following objects are masked from 'package:stats':
IQR, mad, sd, var, xtabs
The following objects are masked from 'package:base':
Filter, Find, Map, Position, Reduce, anyDuplicated, append,
as.data.frame, basename, cbind, colnames, dirname, do.call,
duplicated, eval, evalq, get, grep, grepl, intersect, is.unsorted,
lapply, mapply, match, mget, order, paste, pmax, pmax.int, pmin,
pmin.int, rank, rbind, rownames, sapply, setdiff, sort, table,
tapply, union, unique, unsplit, which, which.max, which.min
Welcome to Bioconductor
Vignettes contain introductory material; view with
'browseVignettes()'. To cite Bioconductor, see
'citation("Biobase")', and for packages 'citation("pkgname")'.
> library(doppelgangR)
Loading required package: BiocParallel
>
> ncor <- 0; npheno <- 0; nsmoking <- 0
> options(stringsAsFactors=FALSE)
> set.seed(1)
> m1 <- matrix(rnorm(1100), ncol=11)
> colnames(m1) <- paste("m", 1:11, sep="")
> rownames(m1) <- make.names(1:nrow(m1))
> n1 <- matrix(rnorm(1000), ncol=10)
> colnames(n1) <- paste("n", 1:10, sep="")
> rownames(n1) <- make.names(1:nrow(n1))
> ##m:1 & n:1 are expression doppelgangers:
> m1[, 1] <- n1[, 1] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##m:2 & m:3 are expression doppelgangers:
> m1[, 2] <- m1[, 3] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##n:2 & n:3 are expression doppelgangers:
> n1[, 2] <- n1[, 3] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##n:8 & n:9 are expression doppelgangers:
> n1[, 8] <- n1[, 9] + rnorm(100, sd=0.25); ncor <- ncor+1
> #n:4 & m:6 are expression doppelgangers:
> n1[, 4] <- m1[, 6] + rnorm(100, sd=0.25); ncor <- ncor+1
> #n:5 & m:4 are expression doppelgangers:
> n1[, 5] <- m1[, 4] + rnorm(100, sd=0.25); ncor <- ncor+1
>
> ##
> ##m:10 and n:10 are phenotype doppelgangers:
> m.pdata <- matrix(letters[sample(1:26, size=110, replace=TRUE)], ncol=10)
> rownames(m.pdata) <- colnames(m1)
> n.pdata <- matrix(letters[sample(1:26, size=100, replace=TRUE)], ncol=10)
> n.pdata[10, ] <- m.pdata[10, ]; npheno <- npheno+1
> rownames(n.pdata) <- colnames(n1)
>
>
>
> ##Create ExpressionSets
> m.eset <- ExpressionSet(assayData=m1, phenoData=AnnotatedDataFrame(data.frame(m.pdata)))
> m.eset$id <- toupper(colnames(m1))
> ##
> n.eset <- ExpressionSet(assayData=n1, phenoData=AnnotatedDataFrame(data.frame(n.pdata)))
> n.eset$id <- toupper(colnames(n1))
> ##m5 and n4 are "smoking gun" doppelgangers:
> n.eset$id[4] <- "gotcha"
> m.eset$id[5] <- "gotcha"
> nsmoking <- nsmoking+1
> ##
> esets <- list(m=m.eset, n=n.eset)
>
> ##------------------------------------------
> ##Check of all three types of doppelgangers:
> ##------------------------------------------
> res1 <- doppelgangR(esets, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df1 <- summary(res1)
>
> checkIdentical(df1[df1$sample1=="m:m1" & df1$sample2=="n:n1", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m2" & df1$sample2=="m:m3", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m6" & df1$sample2=="n:n4", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="n:n2" & df1$sample2=="n:n3", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m6" & df1$sample2=="n:n4", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m10" & df1$sample2=="n:n10", "pheno.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m5" & df1$sample2=="n:n4", "smokinggun.doppel"], TRUE)
[1] TRUE
> checkEquals(nrow(df1), ncor+npheno+nsmoking)
[1] TRUE
> checkEquals(sum(df1$expr.doppel), ncor)
[1] TRUE
> checkEquals(sum(df1$pheno.doppel), npheno)
[1] TRUE
> checkEquals(sum(df1$smokinggun.doppel), nsmoking)
[1] TRUE
> checkEquals(sum(is.na(df1$expr.similarity)), 0)
[1] TRUE
> checkEquals(sum(is.na(df1$pheno.similarity)), 0)
[1] TRUE
> checkEquals(sum(is.na(df1$smokinggun.similarity)), 0)
[1] TRUE
> checkEquals(df1$id, c("M2:M3", "N2:N3", "N8:N9", "M1:N1", "gotcha:gotcha", "M6:gotcha", "M4:N5", "M10:N10"))
[1] TRUE
> for (i in match(paste("X", 1:10, sep=""), colnames(df1))){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(all(grepl("[a-z]:[a-z]", df1[[i]])), TRUE)
+ }
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
>
> ##------------------------------------------
> cat("\n")
> cat("Check without smoking guns: \n")
Check without smoking guns:
> ##------------------------------------------
> res2 <- doppelgangR(esets, smokingGunFinder.args=NULL, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df2 <- summary(res2)
> for (i in grep("pheno.similarity|smokinggun.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df2[, i], df1[!df1$smokinggun.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 3
Checking column 4
Checking column 6
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
>
> ##------------------------------------------
> cat("\n")
> cat("Check without phenotype: \n")
Check without phenotype:
> ##------------------------------------------
> res3 <- doppelgangR(esets, phenoFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Finalizing...
> df3 <- summary(res3)
> for (i in grep("pheno.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df3[, i], df1[!df1$pheno.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 3
Checking column 4
Checking column 6
Checking column 7
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
> ##------------------------------------------
> cat("\n")
> cat("Check without expression: \n")
Check without expression:
> ##------------------------------------------
> res4 <- doppelgangR(esets, corFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df4 <- summary(res4)
> for (i in grep("expr.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df4[, i], df1[!df1$expr.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 4
Checking column 5
Checking column 6
Checking column 7
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
> ##------------------------------------------
> cat("\n")
> cat("Check smoking guns only: \n")
Check smoking guns only:
> ##------------------------------------------
> res4b <- doppelgangR(esets, corFinder.args=NULL, phenoFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Finalizing...
> df4b <- summary(res4b); rownames(df4b) <- NULL
> df4b.compare <- df1[df1$smokinggun.doppel, ]; rownames(df4b.compare) <- NULL
> checkIdentical(df4b.compare[, -3:-6], df4b[, -3:-6]) ##don't check expr and pheno columns
[1] TRUE
>
>
> ##------------------------------------------
> cat("\n")
> cat("Check pruning: \n")
Check pruning:
> ##------------------------------------------
> res5 <- doppelgangR(esets, manual.smokingguns="id", automatic.smokingguns=FALSE, intermediate.pruning=TRUE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df5 <- summary(res5)
> checkEquals(df1, df5)
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Check caching, with a third ExpressionSet that is almost identical to the first: \n")
Check caching, with a third ExpressionSet that is almost identical to the first:
> ##------------------------------------------
> esets2 <- c(esets, esets[[1]])
> #newmat <- exprs(esets[[1]])
> #newmat <- newmat + rnorm(nrow(newmat) * ncol(newmat), sd=0.1)
> #ExpressionSet(assayData=exprs(esets[[1]]), phenoData=phenoData(esets[[1]]))
>
> names(esets2)[3] <- "o"
> exprs(esets2[[3]]) <- exprs(esets2[[3]]) + rnorm(nrow(esets2[[3]]) * ncol(esets2[[3]]), sd=0.1)
> esets2[[3]]$X10 <- "a"
> sampleNames(esets2[[3]]) <- paste("X", sampleNames(esets2[[3]]), sep="")
>
> tmpcachedir <- tempdir()
> ##Do this twice, so the second time the cache will be used:
> for (i in 1:2){
+ res6 <- doppelgangR(esets2, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=tmpcachedir)
+ ##Make sure comparison of m to n is the same as res1:
+ df6a <- summary(res6)
+ df6a <- df6a[grepl("^[mn]", df6a$sample1) & grepl("^[mn]", df6a$sample2), ]
+ rownames(df6a) <- 1:nrow(df6a)
+ checkEquals(df1, df6a)
+ }
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets o and o
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Working on datasets m and m
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets o and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
>
> df6b <- summary(res6)
> df6b <- df6b[grepl("^[mo]", df6b$sample1) & grepl("^[mo]", df6b$sample2), ]
> checkIdentical(df6b[df6b$sample1=="o:Xm2" & df6b$sample2=="o:Xm3", "expr.doppel"], TRUE)
[1] TRUE
> df6b <- df6b[-1:-2, ]
> ##checkTrue(all(df6b$expr.doppel)) ## not a bug, but a shortcoming in the outlier detection that these are not all identified as expression doppelgangers.
> checkTrue(all(df6b$pheno.doppel))
[1] TRUE
> checkTrue(all(df6b$smokinggun.doppel))
[1] TRUE
>
>
> res7 <- doppelgangR(esets2, phenoFinder.args=NULL, smokingGunFinder.args=NULL,
+ outlierFinder.expr.args=list(bonf.prob = 1.0, transFun = atanh,
+ tail = "upper"), cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets o and o
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets m and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Working on datasets n and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Finalizing...
> df7 <- summary(res7)
> has.o <- grepl("^o", df7$sample1) | grepl("^o", df7$sample2)
>
> df7a <- df7[has.o, ]
> df7a$sample1 <- sub("o", "m", df7a$sample1)
> df7a$sample1 <- sub("X", "", df7a$sample1)
> df7a$sample2 <- sub("o", "m", df7a$sample2)
> df7a$sample2 <- sub("X", "", df7a$sample2)
>
> df7b <- df7[!has.o, ]
> df7b <- df7b[(!grepl("^n", df7b$sample1) | !grepl("^n", df7b$sample2)), ]
> for (i in 1:nrow(df7a)){
+ df7a[i, 1:2] <- sort(df7a[i, 1:2])
+ df7b[i, 1:2] <- sort(df7b[i, 1:2])
+ }
> df7a <- df7a[order(df7a$sample1, df7a$sample2), ]
> df7b <- df7b[order(df7b$sample1, df7b$sample2), ]
> checkTrue(all(df7a$expr.doppel == df7b$expr.doppel))
[1] TRUE
> checkTrue(all(df7a$pheno.doppel == df7b$pheno.doppel))
[1] TRUE
> checkTrue(all(df7a$smokinggun.doppel == df7b$smokinggun.doppel))
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Check corFinder function: \n")
Check corFinder function:
> ##------------------------------------------
> cor1 <- corFinder(eset.pair=esets)
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
> cor2 <- corFinder(eset.pair=esets[c(2, 1)])
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
> checkEquals(cor1, t(cor2))
[1] TRUE
>
> cor1 <- corFinder(eset.pair=esets, use.ComBat=FALSE)
> cor2 <- corFinder(eset.pair=esets[c(2, 1)], use.ComBat=FALSE)
> checkEquals(cor1, t(cor2))
[1] TRUE
>
> cor1 <- corFinder(eset.pair=esets[c(1, 1)])
> cor2 <- corFinder(eset.pair=esets[c(1, 1)], use.ComBat=FALSE)
> checkEquals(cor1, cor2)
[1] TRUE
>
> ##Check missing values:
> exprs(esets[[1]])[1:10, 1:5] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
> ## More missing values:
> exprs(esets[[1]])[1:10, 1:8] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning message:
In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> ## More missing values:
> exprs(esets[[1]])[1:10, 1:11] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning message:
In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> ## infinite values:
> exprs(esets[[1]])[14, 1] <- -Inf
> exprs(esets[[1]])[15, 2] <- Inf
> obs <- tryCatch(doppelgangR(esets[1:2]), warning=conditionMessage)
> checkIdentical("Replacing -+Inf with min/max expression values for dataset m", obs)
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Smoking guns only with cache=TRUE: \n")
Smoking guns only with cache=TRUE:
> ##------------------------------------------
> dop <- doppelgangR(esets, corFinder.args=NULL, phenoFinder.args=NULL, manual.smokingguns="id")
KNN imputation for m
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets, corFinder.args = NULL, phenoFinder.args = NULL, :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkEquals(summary(dop)[, 1], "m:m5")
[1] TRUE
> checkEquals(summary(dop)[, 2], "n:n4")
[1] TRUE
>
>
> ##------------------------------------------
> cat("\n")
> cat("Identical ExpressionSets:\n")
Identical ExpressionSets:
> ##------------------------------------------
> df1 <- summary(doppelgangR(esets[[1]], cache.dir=NULL))
KNN imputation for ExpressionSet1
KNN imputation for ExpressionSet2
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets[[1]], cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
3: In doppelgangR(esets[[1]], cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet2
4: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkTrue(df1$sample1 == "m2")
[1] TRUE
> checkTrue(df1$sample2 == "m3")
[1] TRUE
> ##
> df2 <- summary(doppelgangR(esets[[2]], cache.dir=NULL))
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> checkTrue(all(df2$sample1 == "n2"))
[1] TRUE
> checkTrue(all(df2$sample2 == "n3"))
[1] TRUE
> ##
> df5 <- summary(doppelgangR(list(eset1=esets[[1]], eset2=esets[[2]]), cache.dir=NULL))
KNN imputation for eset1
Working on datasets eset1 and eset1
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets eset2 and eset2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets eset1 and eset2
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(list(eset1 = esets[[1]], eset2 = esets[[2]]), cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset eset1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> df6 <- summary(doppelgangR(esets, cache.dir=NULL))
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets, cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkIdentical(df5[, -1:-2], df6[, -1:-2])
[1] TRUE
> checkIdentical(sub("eset2", "n", sub("eset1", "m", df5$sample1)), df6$sample1)
[1] TRUE
> checkIdentical(sub("eset2", "n", sub("eset1", "m", df5$sample2)), df6$sample2)
[1] TRUE
>
> ## with zero-column pData:
> withpheno <- summary(doppelgangR(esets))
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> esets3 <- esets
> pData(esets3[[1]]) <- pData(esets3[[1]])[, 0]
> withoutpheno <- summary(doppelgangR(esets3))
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Finalizing...
Warning message:
In FUN(...) :
m and n have different column names in phenoData. Skipping phenotype checking for this pair. Set phenoFinger.args=NULL to disable phenotype checking altogether.
Warning messages:
1: In doppelgangR(esets3) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
>
> pData(esets3[[2]]) <- pData(esets3[[2]])[, 0]
> withoutpheno2 <- summary(doppelgangR(esets3))
KNN imputation for m
Working on datasets m and m
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets3) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
>
> withoutpheno3 <- withoutpheno[!withoutpheno$pheno.doppel, ]
> withoutpheno4 <- withoutpheno2[!withoutpheno2$pheno.doppel, ]
>
> checkIdentical(withoutpheno[, 1:4], withoutpheno3[, 1:4])
[1] TRUE
> checkIdentical(withoutpheno2[, 1:4], withoutpheno4[, 1:4])
[1] TRUE
>
> esets4 <- esets
> for (i in 1:length(esets4))
+ pData(esets4[[i]]) <- pData(esets4[[i]])[1]
> doppelgangR(esets4[[1]])
KNN imputation for ExpressionSet1
KNN imputation for ExpressionSet2
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 1 : 1 expression, 0 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning messages:
1: In doppelgangR(esets4[[1]]) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
3: In doppelgangR(esets4[[1]]) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet2
4: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> doppelgangR(esets4[[2]])
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 1 : 1 expression, 0 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
>
>
> proc.time()
user system elapsed
20.51 0.87 254.35
|
doppelgangR.Rcheck/tests_x64/maintest.Rout
R version 3.6.3 (2020-02-29) -- "Holding the Windsock"
Copyright (C) 2020 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (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(RUnit)
> library(Biobase)
Loading required package: BiocGenerics
Loading required package: parallel
Attaching package: 'BiocGenerics'
The following objects are masked from 'package:parallel':
clusterApply, clusterApplyLB, clusterCall, clusterEvalQ,
clusterExport, clusterMap, parApply, parCapply, parLapply,
parLapplyLB, parRapply, parSapply, parSapplyLB
The following objects are masked from 'package:stats':
IQR, mad, sd, var, xtabs
The following objects are masked from 'package:base':
Filter, Find, Map, Position, Reduce, anyDuplicated, append,
as.data.frame, basename, cbind, colnames, dirname, do.call,
duplicated, eval, evalq, get, grep, grepl, intersect, is.unsorted,
lapply, mapply, match, mget, order, paste, pmax, pmax.int, pmin,
pmin.int, rank, rbind, rownames, sapply, setdiff, sort, table,
tapply, union, unique, unsplit, which, which.max, which.min
Welcome to Bioconductor
Vignettes contain introductory material; view with
'browseVignettes()'. To cite Bioconductor, see
'citation("Biobase")', and for packages 'citation("pkgname")'.
> library(doppelgangR)
Loading required package: BiocParallel
>
> ncor <- 0; npheno <- 0; nsmoking <- 0
> options(stringsAsFactors=FALSE)
> set.seed(1)
> m1 <- matrix(rnorm(1100), ncol=11)
> colnames(m1) <- paste("m", 1:11, sep="")
> rownames(m1) <- make.names(1:nrow(m1))
> n1 <- matrix(rnorm(1000), ncol=10)
> colnames(n1) <- paste("n", 1:10, sep="")
> rownames(n1) <- make.names(1:nrow(n1))
> ##m:1 & n:1 are expression doppelgangers:
> m1[, 1] <- n1[, 1] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##m:2 & m:3 are expression doppelgangers:
> m1[, 2] <- m1[, 3] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##n:2 & n:3 are expression doppelgangers:
> n1[, 2] <- n1[, 3] + rnorm(100, sd=0.25); ncor <- ncor+1
> ##n:8 & n:9 are expression doppelgangers:
> n1[, 8] <- n1[, 9] + rnorm(100, sd=0.25); ncor <- ncor+1
> #n:4 & m:6 are expression doppelgangers:
> n1[, 4] <- m1[, 6] + rnorm(100, sd=0.25); ncor <- ncor+1
> #n:5 & m:4 are expression doppelgangers:
> n1[, 5] <- m1[, 4] + rnorm(100, sd=0.25); ncor <- ncor+1
>
> ##
> ##m:10 and n:10 are phenotype doppelgangers:
> m.pdata <- matrix(letters[sample(1:26, size=110, replace=TRUE)], ncol=10)
> rownames(m.pdata) <- colnames(m1)
> n.pdata <- matrix(letters[sample(1:26, size=100, replace=TRUE)], ncol=10)
> n.pdata[10, ] <- m.pdata[10, ]; npheno <- npheno+1
> rownames(n.pdata) <- colnames(n1)
>
>
>
> ##Create ExpressionSets
> m.eset <- ExpressionSet(assayData=m1, phenoData=AnnotatedDataFrame(data.frame(m.pdata)))
> m.eset$id <- toupper(colnames(m1))
> ##
> n.eset <- ExpressionSet(assayData=n1, phenoData=AnnotatedDataFrame(data.frame(n.pdata)))
> n.eset$id <- toupper(colnames(n1))
> ##m5 and n4 are "smoking gun" doppelgangers:
> n.eset$id[4] <- "gotcha"
> m.eset$id[5] <- "gotcha"
> nsmoking <- nsmoking+1
> ##
> esets <- list(m=m.eset, n=n.eset)
>
> ##------------------------------------------
> ##Check of all three types of doppelgangers:
> ##------------------------------------------
> res1 <- doppelgangR(esets, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df1 <- summary(res1)
>
> checkIdentical(df1[df1$sample1=="m:m1" & df1$sample2=="n:n1", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m2" & df1$sample2=="m:m3", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m6" & df1$sample2=="n:n4", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="n:n2" & df1$sample2=="n:n3", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m6" & df1$sample2=="n:n4", "expr.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m10" & df1$sample2=="n:n10", "pheno.doppel"], TRUE)
[1] TRUE
> checkIdentical(df1[df1$sample1=="m:m5" & df1$sample2=="n:n4", "smokinggun.doppel"], TRUE)
[1] TRUE
> checkEquals(nrow(df1), ncor+npheno+nsmoking)
[1] TRUE
> checkEquals(sum(df1$expr.doppel), ncor)
[1] TRUE
> checkEquals(sum(df1$pheno.doppel), npheno)
[1] TRUE
> checkEquals(sum(df1$smokinggun.doppel), nsmoking)
[1] TRUE
> checkEquals(sum(is.na(df1$expr.similarity)), 0)
[1] TRUE
> checkEquals(sum(is.na(df1$pheno.similarity)), 0)
[1] TRUE
> checkEquals(sum(is.na(df1$smokinggun.similarity)), 0)
[1] TRUE
> checkEquals(df1$id, c("M2:M3", "N2:N3", "N8:N9", "M1:N1", "gotcha:gotcha", "M6:gotcha", "M4:N5", "M10:N10"))
[1] TRUE
> for (i in match(paste("X", 1:10, sep=""), colnames(df1))){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(all(grepl("[a-z]:[a-z]", df1[[i]])), TRUE)
+ }
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
>
> ##------------------------------------------
> cat("\n")
> cat("Check without smoking guns: \n")
Check without smoking guns:
> ##------------------------------------------
> res2 <- doppelgangR(esets, smokingGunFinder.args=NULL, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df2 <- summary(res2)
> for (i in grep("pheno.similarity|smokinggun.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df2[, i], df1[!df1$smokinggun.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 3
Checking column 4
Checking column 6
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
>
> ##------------------------------------------
> cat("\n")
> cat("Check without phenotype: \n")
Check without phenotype:
> ##------------------------------------------
> res3 <- doppelgangR(esets, phenoFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Finalizing...
> df3 <- summary(res3)
> for (i in grep("pheno.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df3[, i], df1[!df1$pheno.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 3
Checking column 4
Checking column 6
Checking column 7
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
> ##------------------------------------------
> cat("\n")
> cat("Check without expression: \n")
Check without expression:
> ##------------------------------------------
> res4 <- doppelgangR(esets, corFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df4 <- summary(res4)
> for (i in grep("expr.similarity", colnames(df1), invert=TRUE)){
+ cat(paste("Checking column", i, "\n"))
+ checkEquals(df4[, i], df1[!df1$expr.doppel, i])
+ }
Checking column 1
Checking column 2
Checking column 4
Checking column 5
Checking column 6
Checking column 7
Checking column 8
Checking column 9
Checking column 10
Checking column 11
Checking column 12
Checking column 13
Checking column 14
Checking column 15
Checking column 16
Checking column 17
Checking column 18
Checking column 19
>
> ##------------------------------------------
> cat("\n")
> cat("Check smoking guns only: \n")
Check smoking guns only:
> ##------------------------------------------
> res4b <- doppelgangR(esets, corFinder.args=NULL, phenoFinder.args=NULL, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=NULL)
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Finalizing...
> df4b <- summary(res4b); rownames(df4b) <- NULL
> df4b.compare <- df1[df1$smokinggun.doppel, ]; rownames(df4b.compare) <- NULL
> checkIdentical(df4b.compare[, -3:-6], df4b[, -3:-6]) ##don't check expr and pheno columns
[1] TRUE
>
>
> ##------------------------------------------
> cat("\n")
> cat("Check pruning: \n")
Check pruning:
> ##------------------------------------------
> res5 <- doppelgangR(esets, manual.smokingguns="id", automatic.smokingguns=FALSE, intermediate.pruning=TRUE, cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> df5 <- summary(res5)
> checkEquals(df1, df5)
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Check caching, with a third ExpressionSet that is almost identical to the first: \n")
Check caching, with a third ExpressionSet that is almost identical to the first:
> ##------------------------------------------
> esets2 <- c(esets, esets[[1]])
> #newmat <- exprs(esets[[1]])
> #newmat <- newmat + rnorm(nrow(newmat) * ncol(newmat), sd=0.1)
> #ExpressionSet(assayData=exprs(esets[[1]]), phenoData=phenoData(esets[[1]]))
>
> names(esets2)[3] <- "o"
> exprs(esets2[[3]]) <- exprs(esets2[[3]]) + rnorm(nrow(esets2[[3]]) * ncol(esets2[[3]]), sd=0.1)
> esets2[[3]]$X10 <- "a"
> sampleNames(esets2[[3]]) <- paste("X", sampleNames(esets2[[3]]), sep="")
>
> tmpcachedir <- tempdir()
> ##Do this twice, so the second time the cache will be used:
> for (i in 1:2){
+ res6 <- doppelgangR(esets2, manual.smokingguns="id", automatic.smokingguns=FALSE, cache.dir=tmpcachedir)
+ ##Make sure comparison of m to n is the same as res1:
+ df6a <- summary(res6)
+ df6a <- df6a[grepl("^[mn]", df6a$sample1) & grepl("^[mn]", df6a$sample2), ]
+ rownames(df6a) <- 1:nrow(df6a)
+ checkEquals(df1, df6a)
+ }
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets o and o
Calculating correlations...
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Working on datasets m and m
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets o and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and o
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Identifying smoking-gun doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
>
> df6b <- summary(res6)
> df6b <- df6b[grepl("^[mo]", df6b$sample1) & grepl("^[mo]", df6b$sample2), ]
> checkIdentical(df6b[df6b$sample1=="o:Xm2" & df6b$sample2=="o:Xm3", "expr.doppel"], TRUE)
[1] TRUE
> df6b <- df6b[-1:-2, ]
> ##checkTrue(all(df6b$expr.doppel)) ## not a bug, but a shortcoming in the outlier detection that these are not all identified as expression doppelgangers.
> checkTrue(all(df6b$pheno.doppel))
[1] TRUE
> checkTrue(all(df6b$smokinggun.doppel))
[1] TRUE
>
>
> res7 <- doppelgangR(esets2, phenoFinder.args=NULL, smokingGunFinder.args=NULL,
+ outlierFinder.expr.args=list(bonf.prob = 1.0, transFun = atanh,
+ tail = "upper"), cache.dir=NULL)
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets o and o
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Working on datasets m and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Working on datasets n and o
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Finalizing...
> df7 <- summary(res7)
> has.o <- grepl("^o", df7$sample1) | grepl("^o", df7$sample2)
>
> df7a <- df7[has.o, ]
> df7a$sample1 <- sub("o", "m", df7a$sample1)
> df7a$sample1 <- sub("X", "", df7a$sample1)
> df7a$sample2 <- sub("o", "m", df7a$sample2)
> df7a$sample2 <- sub("X", "", df7a$sample2)
>
> df7b <- df7[!has.o, ]
> df7b <- df7b[(!grepl("^n", df7b$sample1) | !grepl("^n", df7b$sample2)), ]
> for (i in 1:nrow(df7a)){
+ df7a[i, 1:2] <- sort(df7a[i, 1:2])
+ df7b[i, 1:2] <- sort(df7b[i, 1:2])
+ }
> df7a <- df7a[order(df7a$sample1, df7a$sample2), ]
> df7b <- df7b[order(df7b$sample1, df7b$sample2), ]
> checkTrue(all(df7a$expr.doppel == df7b$expr.doppel))
[1] TRUE
> checkTrue(all(df7a$pheno.doppel == df7b$pheno.doppel))
[1] TRUE
> checkTrue(all(df7a$smokinggun.doppel == df7b$smokinggun.doppel))
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Check corFinder function: \n")
Check corFinder function:
> ##------------------------------------------
> cor1 <- corFinder(eset.pair=esets)
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
> cor2 <- corFinder(eset.pair=esets[c(2, 1)])
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
> checkEquals(cor1, t(cor2))
[1] TRUE
>
> cor1 <- corFinder(eset.pair=esets, use.ComBat=FALSE)
> cor2 <- corFinder(eset.pair=esets[c(2, 1)], use.ComBat=FALSE)
> checkEquals(cor1, t(cor2))
[1] TRUE
>
> cor1 <- corFinder(eset.pair=esets[c(1, 1)])
> cor2 <- corFinder(eset.pair=esets[c(1, 1)], use.ComBat=FALSE)
> checkEquals(cor1, cor2)
[1] TRUE
>
> ##Check missing values:
> exprs(esets[[1]])[1:10, 1:5] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
> ## More missing values:
> exprs(esets[[1]])[1:10, 1:8] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning message:
In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> ## More missing values:
> exprs(esets[[1]])[1:10, 1:11] <- NA
> doppelgangR(esets[1:2])
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 7 : 6 expression, 1 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning message:
In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> ## infinite values:
> exprs(esets[[1]])[14, 1] <- -Inf
> exprs(esets[[1]])[15, 2] <- Inf
> obs <- tryCatch(doppelgangR(esets[1:2]), warning=conditionMessage)
> checkIdentical("Replacing -+Inf with min/max expression values for dataset m", obs)
[1] TRUE
>
> ##------------------------------------------
> cat("\n")
> cat("Smoking guns only with cache=TRUE: \n")
Smoking guns only with cache=TRUE:
> ##------------------------------------------
> dop <- doppelgangR(esets, corFinder.args=NULL, phenoFinder.args=NULL, manual.smokingguns="id")
KNN imputation for m
Working on datasets m and m
Identifying smoking-gun doppelgangers...
Working on datasets n and n
Identifying smoking-gun doppelgangers...
Working on datasets m and n
Identifying smoking-gun doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets, corFinder.args = NULL, phenoFinder.args = NULL, :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkEquals(summary(dop)[, 1], "m:m5")
[1] TRUE
> checkEquals(summary(dop)[, 2], "n:n4")
[1] TRUE
>
>
> ##------------------------------------------
> cat("\n")
> cat("Identical ExpressionSets:\n")
Identical ExpressionSets:
> ##------------------------------------------
> df1 <- summary(doppelgangR(esets[[1]], cache.dir=NULL))
KNN imputation for ExpressionSet1
KNN imputation for ExpressionSet2
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets[[1]], cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
3: In doppelgangR(esets[[1]], cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet2
4: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkTrue(df1$sample1 == "m2")
[1] TRUE
> checkTrue(df1$sample2 == "m3")
[1] TRUE
> ##
> df2 <- summary(doppelgangR(esets[[2]], cache.dir=NULL))
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
> checkTrue(all(df2$sample1 == "n2"))
[1] TRUE
> checkTrue(all(df2$sample2 == "n3"))
[1] TRUE
> ##
> df5 <- summary(doppelgangR(list(eset1=esets[[1]], eset2=esets[[2]]), cache.dir=NULL))
KNN imputation for eset1
Working on datasets eset1 and eset1
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets eset2 and eset2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets eset1 and eset2
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(list(eset1 = esets[[1]], eset2 = esets[[2]]), cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset eset1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> df6 <- summary(doppelgangR(esets, cache.dir=NULL))
KNN imputation for m
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets, cache.dir = NULL) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> checkIdentical(df5[, -1:-2], df6[, -1:-2])
[1] TRUE
> checkIdentical(sub("eset2", "n", sub("eset1", "m", df5$sample1)), df6$sample1)
[1] TRUE
> checkIdentical(sub("eset2", "n", sub("eset1", "m", df5$sample2)), df6$sample2)
[1] TRUE
>
> ## with zero-column pData:
> withpheno <- summary(doppelgangR(esets))
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Skipping phenoFinder, loading cached results.
Identifying phenotype doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> esets3 <- esets
> pData(esets3[[1]]) <- pData(esets3[[1]])[, 0]
> withoutpheno <- summary(doppelgangR(esets3))
KNN imputation for m
Working on datasets m and m
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets n and n
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Finalizing...
Warning message:
In FUN(...) :
m and n have different column names in phenoData. Skipping phenotype checking for this pair. Set phenoFinger.args=NULL to disable phenotype checking altogether.
Warning messages:
1: In doppelgangR(esets3) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
>
> pData(esets3[[2]]) <- pData(esets3[[2]])[, 0]
> withoutpheno2 <- summary(doppelgangR(esets3))
KNN imputation for m
Working on datasets m and m
Skipping corFinder, loading cached results.
Identifying correlation doppelgangers...
Working on datasets n and n
Calculating correlations...
Identifying correlation doppelgangers...
Working on datasets m and n
Calculating correlations...
Found2batches
Adjusting for0covariate(s) or covariate level(s)
Standardizing Data across genes
Fitting L/S model and finding priors
Finding parametric adjustments
Adjusting the Data
Identifying correlation doppelgangers...
Finalizing...
Warning messages:
1: In doppelgangR(esets3) :
Replacing -+Inf with min/max expression values for dataset m
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
>
> withoutpheno3 <- withoutpheno[!withoutpheno$pheno.doppel, ]
> withoutpheno4 <- withoutpheno2[!withoutpheno2$pheno.doppel, ]
>
> checkIdentical(withoutpheno[, 1:4], withoutpheno3[, 1:4])
[1] TRUE
> checkIdentical(withoutpheno2[, 1:4], withoutpheno4[, 1:4])
[1] TRUE
>
> esets4 <- esets
> for (i in 1:length(esets4))
+ pData(esets4[[i]]) <- pData(esets4[[i]])[1]
> doppelgangR(esets4[[1]])
KNN imputation for ExpressionSet1
KNN imputation for ExpressionSet2
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 1 : 1 expression, 0 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
Warning messages:
1: In doppelgangR(esets4[[1]]) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet1
2: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
3: In doppelgangR(esets4[[1]]) :
Replacing -+Inf with min/max expression values for dataset ExpressionSet2
4: In knnimp(x, k, maxmiss = rowmax, maxp = maxp) :
10 rows with more than 50 % entries missing;
mean imputation used for these rows
> doppelgangR(esets4[[2]])
Working on datasets ExpressionSet1 and ExpressionSet2
Calculating correlations...
Identifying correlation doppelgangers...
Calculating phenotype similarities...
Identifying phenotype doppelgangers...
Finalizing...
S4 object of class: DoppelGang
Number of potential doppelgangers: 1 : 1 expression, 0 phenotype, 0 smoking gun.
Use summary(object) to obtain a data.frame of potential doppelgangrs.
>
>
> proc.time()
user system elapsed
24.42 0.62 276.93
|
