Differential expression analysis–EdgeR log likelihood ratio tests

Last updated: 2019-02-14

• ✖ R Markdown file: uncommitted changes

  The R Markdown file has unstaged changes. To know which version of the R Markdown file created these results, you’ll want to first commit it to the Git repo. If you’re still working on the analysis, you can ignore this warning. When you’re finished, you can run wflow_publish to commit the R Markdown file and build the HTML.

• ✔ Environment: empty

  Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.

• ✔ Seed: set.seed(20181119)

  The command set.seed(20181119) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.

• ✔ Session information: recorded

  Great job! Recording the operating system, R version, and package versions is critical for reproducibility.

• ✔ Repository version: fcaa341

  Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility. The version displayed above was the version of the Git repository at the time these results were generated.
  
  Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated:

  
  Ignored files:
      Ignored:    .Rhistory
      Ignored:    .Rproj.user/
      Ignored:    analysis/Quality_metrics_cache/
  
  Untracked files:
      Untracked:  analysis/enrichment.Rmd
      Untracked:  analysis/figure/
      Untracked:  data/SNP_200000.txt
      Untracked:  data/SNP_50000.txt
      Untracked:  data/SNP_500000.txt
      Untracked:  data/edgeR-qlf-10%filter_res.Rd
  
  Unstaged changes:
      Modified:   .gitignore
      Modified:   analysis/DEseq2-LRT.Rmd
      Modified:   analysis/EdgeR-LRT.Rmd
      Modified:   analysis/EdgeR-QLF.Rmd
      Modified:   analysis/Quality_metrics.Rmd
      Modified:   analysis/index.Rmd
      Modified:   code/summary_functions.R
      Modified:   cropseq.Rproj
      Modified:   data/DE_input.Rd
  
  

  Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

Load data

source("code/summary_functions.R")
library(dplyr)
library(gtools)
library(data.table)
load("data/DE_input.Rd")
glocus <- "VPS45"
Nperm <- 5
dim(dm)[1]

NULL

gcount <- dm[1:(dim(dm)[1]-76), colnames(dm1dfagg)[dm1dfagg[glocus,] >0 & nlocus==1]]
# negative control cells defined as neg gRNA targeted cells
ncount <- dm[1:(dim(dm)[1]-76), colnames(dm1dfagg)[dm1dfagg["neg",] >0 & nlocus==1]]
coldata <- data.frame(row.names = c(colnames(gcount),colnames(ncount)),     
                      condition=c(rep('G',dim(gcount)[2]),rep('N',dim(ncount)[2])))
countall <- cbind(gcount,ncount)
totalcount <- apply(countall,1,sum)
cellpercent <-  apply(countall,1,function(x) length(x[x>0])/length(x))

edgeR log likelihood ratio tests function

library(edgeR)
run_edgeR <- function(y,plotit=T) {
  # y is DGElist object
  y <- calcNormFactors(y)
  group= y$samples[,"group"]
  design <- model.matrix(~group)
  y <- estimateDisp(y,design)

  fitlrt <- glmFit(y,design)
  lrt <- glmLRT(fitlrt,coef=2)
  out <- topTags(lrt, n=Inf, adjust.method = "BH")

  if (plotit==T) {
    outsig <- subset(out$table,FDR <0.1) 
    summ_pvalues(lrt$table$PValue)
    print(paste0("There are ",dim(outsig)[1], " genes passed FDR <0.1 cutoff"))
    print(knitr::kable(signif(as.matrix(head(out$table[order(out$table$PValue),])),digit=2)))
  }
  return(out)
}

y <- DGEList(counts= countall,group=coldata$condition)
resm <- run_edgeR(y)

[1] "There are 0 genes passed FDR <0.1 cutoff"

          logFC   logCPM   LR    PValue    FDR
-------  ------  -------  ---  --------  -----
A2M       -1.50      6.9   20   8.6e-06   0.12
LY6H      -2.60      6.6   19   1.2e-05   0.12
LGALS1    -2.20      6.6   19   1.2e-05   0.12
TSPO      -1.50      6.4   14   1.5e-04   0.97
SLF2       1.10      6.4   14   1.9e-04   0.97
POMK      -0.79      6.3   14   2.2e-04   0.97

permreslist <- list()
permreslist[[1]] <- data.table(gene=rownames(resm$table), p=resm$table$PValue, fdr=resm$table$FDR, key="gene")
for (n in 2:(Nperm+1)){
  y <- DGEList(counts= countall,group=permute(coldata$condition))
  res <- run_edgeR(y,plotit = F)
  resp <- data.table(gene=rownames(res$table), p=res$table$PValue, fdr=res$table$FDR, key="gene")
  colnames(resp) <- c("gene", paste0("perm.p_",n-1), paste0("perm.fdr_",n-1))
  permreslist[[n]] <- resp
}
mergedres <- Reduce(merge,permreslist)
knitr::kable(mergedres[fdr <0.1,],digits = 2)

y <- DGEList(counts= countall[totalcount>0,],group=coldata$condition)
resm <- run_edgeR(y)

[1] "There are 3 genes passed FDR <0.1 cutoff"

          logFC   logCPM   LR    PValue     FDR
-------  ------  -------  ---  --------  ------
A2M       -1.50      6.9   20   8.6e-06   0.063
LY6H      -2.60      6.6   19   1.2e-05   0.063
LGALS1    -2.20      6.6   19   1.2e-05   0.063
TSPO      -1.50      6.4   14   1.5e-04   0.510
SLF2       1.10      6.4   14   1.9e-04   0.510
POMK      -0.79      6.3   14   2.2e-04   0.510

permreslist <- list()
permreslist[[1]] <- data.table(gene=rownames(resm$table), p=resm$table$PValue, fdr=resm$table$FDR, key="gene")
for (n in 2:(Nperm+1)){
  y <- DGEList(counts= countall[totalcount>0,],group=permute(coldata$condition))
  res <- run_edgeR(y,plotit = F)
  resp <- data.table(gene=rownames(res$table), p=res$table$PValue, fdr=res$table$FDR, key="gene")
  colnames(resp) <- c("gene", paste0("perm.p_",n-1), paste0("perm.fdr_",n-1))
  permreslist[[n]] <- resp
}
mergedres <- Reduce(merge,permreslist)
knitr::kable(mergedres[fdr <0.1,],digits = 2)

y <- DGEList(counts= countall[cellpercent > 0.03,],group=coldata$condition)
resm <- run_edgeR(y)

[1] "There are 3 genes passed FDR <0.1 cutoff"

           logFC   logCPM   LR    PValue     FDR
--------  ------  -------  ---  --------  ------
LY6H       -2.70      6.6   21   4.1e-06   0.023
LGALS1     -2.30      6.6   21   5.3e-06   0.023
A2M        -1.50      6.9   21   5.9e-06   0.023
TSPO       -1.50      6.4   15   1.1e-04   0.340
FAM228B    -1.00      6.5   14   2.0e-04   0.360
POMK       -0.79      6.3   14   2.1e-04   0.360

permreslist <- list()
permreslist[[1]] <- data.table(gene=rownames(resm$table), p=resm$table$PValue, fdr=resm$table$FDR, key="gene")
for (n in 2:(Nperm+1)){
  y <- DGEList(counts= countall[cellpercent > 0.03,],group=permute(coldata$condition))
  res <- run_edgeR(y,plotit = F)
  resp <- data.table(gene=rownames(res$table), p=res$table$PValue, fdr=res$table$FDR, key="gene")
  colnames(resp) <- c("gene", paste0("perm.p_",n-1), paste0("perm.fdr_",n-1))
  permreslist[[n]] <- resp
}
mergedres <- Reduce(merge,permreslist)
knitr::kable(mergedres[fdr <0.1,],digits = 2)

y <- DGEList(counts= countall[cellpercent > 0.1,],group=coldata$condition)
resm <- run_edgeR(y)

[1] "There are 2 genes passed FDR <0.1 cutoff"

           logFC   logCPM   LR    PValue    FDR
--------  ------  -------  ---  --------  -----
A2M        -1.50      6.9   20   7.0e-06   0.04
LGALS1     -2.20      6.6   20   8.4e-06   0.04
TSPO       -1.50      6.4   15   1.3e-04   0.42
SLF2        1.10      6.5   14   1.8e-04   0.44
FAM228B    -0.98      6.5   14   2.3e-04   0.44
ARAF       -0.83      6.6   13   3.5e-04   0.45

save(resm, file="data/edgeR-lrt-10%filter_res.Rd")

permreslist <- list()
permreslist[[1]] <- data.table(gene=rownames(resm$table), p=resm$table$PValue, fdr=resm$table$FDR, key="gene")
for (n in 2:(Nperm+1)){
  y <- DGEList(counts= countall[cellpercent > 0.1,],group=permute(coldata$condition))
  res <- run_edgeR(y,plotit = F)
  resp <- data.table(gene=rownames(res$table), p=res$table$PValue, fdr=res$table$FDR, key="gene")
  colnames(resp) <- c("gene", paste0("perm.p_",n-1), paste0("perm.fdr_",n-1))
  permreslist[[n]] <- resp
}
mergedres <- Reduce(merge,permreslist)
knitr::kable(mergedres[fdr <0.1,],digits = 2)

y <- DGEList(counts= countall[cellpercent > 0.2,],group=coldata$condition)
resm <- run_edgeR(y)

[1] "There are 1 genes passed FDR <0.1 cutoff"

            logFC   logCPM   LR    PValue     FDR
---------  ------  -------  ---  --------  ------
A2M         -1.50      6.9   19   1.2e-05   0.093
SLF2         1.10      6.5   14   1.7e-04   0.490
FAM228B     -0.99      6.5   14   2.2e-04   0.490
ARAF        -0.82      6.6   13   3.4e-04   0.490
NINJ1        0.73      6.9   12   4.5e-04   0.490
C17orf80     0.95      6.4   12   4.6e-04   0.490

permreslist <- list()
permreslist[[1]] <- data.table(gene=rownames(resm$table), p=resm$table$PValue, fdr=resm$table$FDR, key="gene")
for (n in 2:(Nperm+1)){
  y <- DGEList(counts= countall[cellpercent > 0.2,],group=permute(coldata$condition))
  res <- run_edgeR(y,plotit = F)
  resp <- data.table(gene=rownames(res$table), p=res$table$PValue, fdr=res$table$FDR, key="gene")
  colnames(resp) <- c("gene", paste0("perm.p_",n-1), paste0("perm.fdr_",n-1))
  permreslist[[n]] <- resp
}
mergedres <- Reduce(merge,permreslist)
knitr::kable(mergedres[fdr <0.1,],digits = 2)

sessionInfo()

R version 3.5.1 (2018-07-02)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Scientific Linux 7.4 (Nitrogen)

Matrix products: default
BLAS/LAPACK: /software/openblas-0.2.19-el7-x86_64/lib/libopenblas_haswellp-r0.2.19.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
[1] gridExtra_2.3     edgeR_3.24.3      limma_3.38.2      Matrix_1.2-15    
[5] data.table_1.12.0 gtools_3.8.1      dplyr_0.7.8       lattice_0.20-38  

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.0        highr_0.7         compiler_3.5.1   
 [4] pillar_1.3.1      git2r_0.23.0      workflowr_1.1.1  
 [7] bindr_0.1.1       R.methodsS3_1.7.1 R.utils_2.7.0    
[10] tools_3.5.1       digest_0.6.18     gtable_0.2.0     
[13] evaluate_0.12     tibble_2.0.1      pkgconfig_2.0.2  
[16] rlang_0.3.1       yaml_2.2.0        bindrcpp_0.2.2   
[19] stringr_1.4.0     knitr_1.20        locfit_1.5-9.1   
[22] rprojroot_1.3-2   tidyselect_0.2.5  glue_1.3.0       
[25] R6_2.3.0          rmarkdown_1.10    purrr_0.2.5      
[28] magrittr_1.5      whisker_0.3-2     backports_1.1.2  
[31] htmltools_0.3.6   splines_3.5.1     assertthat_0.2.0 
[34] stringi_1.3.1     crayon_1.3.4      R.oo_1.22.0      

This reproducible R Markdown analysis was created with workflowr 1.1.1