deseq cwl works (without docker)

runs/deseq2/README.md → runs/deseq2-run/README.md

@@ -2,7 +2,7 @@
 
 
 ```bash
-cd runs/deseq2
+cd runs/deseq2-run
 ```
 
 ```bash

runs/deseq2/job.yml → runs/deseq2-run/job.yml

-arcPath: "../../"
-inKallistoResults: "runs/kallisto/kallisto_results"
-inMetadataFile: "runs/merged_isa_metadata/out/merged_isa.tsv"
+inKallistoResults:
+    class: Directory
+    path: ../../runs/kallisto/kallisto_results
+inMetadataFile:
+    class: File
+    path: ../../runs/merged_isa_metadata/out/merged_isa.tsv
 inMetadataSample: "Source.Name"
 inMetadataFactor:
   - "Factor..Photosynthesis.mode."
\ No newline at end of file

workflows/deseq2/README.md

@@ -11,11 +11,10 @@ Workflow used for **differential gene expression analysis**
 - https://bioconductor.org/packages/release/bioc/vignettes/tximport/inst/doc/tximport.html#kallisto
 
 
-
 ## Run pure script
 
 ```bash
-RScript deseq2.R "../../" "runs/kallisto/kallisto_results" "runs/merged_isa_metadata/out/merged_isa.tsv" "Source.Name" "Factor..Photosynthesis.mode."
+RScript deseq2.R "../../runs/kallisto/kallisto_results" "../../runs/merged_isa_metadata/out/merged_isa.tsv" "Source.Name" "Factor..Photosynthesis.mode."
 ```
 
 ## Run CWL

workflows/deseq2/dependencies.Rmd → workflows/deseq2/dependencies.R

----
-title: "Install dependencies"
-author: "Dominik Brilhaus"
-date: "`r Sys.Date()`"
-output: html_document
----
 
+# Install dependencies for deseq2
 
-```{r}
 if (!require("BiocManager", quietly = TRUE))
     install.packages("BiocManager")
 
-
 BiocManager::install("DESeq2")
 library("DESeq2")
 
@@ -19,6 +12,3 @@ library("tximport")
 
 BiocManager::install("rhdf5")
 library("rhdf5")
-
-```
-

workflows/deseq2/deseq2.R

@@ -9,26 +9,24 @@ library("ggplot2")
 
 ## In-and-out
 
-# arcPath <- "../../"
-# inKallistoResults <- "runs/kallisto/kallisto_results"
-# inMetadataFile <- "runs/merged_isa_metadata/out/merged_isa.tsv"
-# inMetadataSample <- "Source.Name"
-# inMetadataFactor <- "Factor..Photosynthesis.mode."
+inKallistoResults <- "../../runs/kallisto/kallisto_results"
+inMetadataFile <- "../../runs/merged_isa_metadata/out/merged_isa.tsv"
+inMetadataSample <- "Source.Name"
+inMetadataFactor <- "Factor..Photosynthesis.mode."
 
 ### Read arguments from CLI
 
 args <- commandArgs(trailingOnly = T)
 
-arcPath <- args[1]
-inKallistoResults <- args[2]
-inMetadataFile <- args[3]
-inMetadataSample <- args[4]
-inMetadataFactor <- args[5]
+inKallistoResults <- args[1]
+inMetadataFile <- args[2]
+inMetadataSample <- args[3]
+inMetadataFactor <- args[4]
 
 ## Import kallisto count data
 
-files <- dir(file.path(arcPath, inKallistoResults) , recursive = T, full.names = T ,"abundance.h5")
-names(files) <- dir(file.path(arcPath, inKallistoResults))
+files <- dir(inKallistoResults, recursive = T, full.names = T ,"abundance.h5")
+names(files) <- dir(inKallistoResults)
 
 txi <- tximport(files, type = "kallisto", txOut = TRUE)
 
@@ -36,7 +34,7 @@ head(txi$counts)
 
 ## Read sample metadata
 
-samples_metadata <- read.table(file = file.path(arcPath, inMetadataFile), sep = "\t")
+samples_metadata <- read.table(file = inMetadataFile, sep = "\t")
 
 samples <- samples_metadata[order(samples_metadata[[inMetadataSample]]), c(inMetadataSample, inMetadataFactor)]
 colnames(samples)[1:2] <- c("sampleID", "condition")
@@ -49,16 +47,16 @@ dds <- DESeqDataSetFromTximport(txi, colData = samples, design = ~ condition)
 
 dds <- DESeq(dds)
 
-## Extract results
+## Outputs
 
-res <- results(dds)
-res
+### Extract results
 
-## Outputs
+res <- results(dds)
+write.csv(res, file = "results_stats.csv", append = FALSE, quote = TRUE)
 
 ### Generate and save default plots
 
-png("ma-plot.png")
+png("results_ma-plot.png")
   plotMA(res, ylim=c(-2,2))
 dev.off()
 
@@ -72,7 +70,7 @@ p2 <- ggplot(pcaData, aes(PC1, PC2, color=condition)) +
   ylab(paste0("PC2: ",percentVar[2],"% variance")) + 
   coord_fixed()
 
-png("pca-plot.png")
+png("results_pca-plot.png")
   print(p2)
 dev.off()
   

workflows/deseq2/deseq2.Rmd

----
-title: "deseq2"
-author: "Dominik Brilhaus"
-date: "`r Sys.Date()`"
-output: html_document
----
-
-
-## Libraries
-
-```{r}
-
-library("DESeq2")
-library("tximport")
-library("rhdf5")
-library("ggplot2")
-
-```
-
-
-
-
-## In-and-out
-
-```{r}
-
-arc <- "../../"
-
-inKallistoResults <- file.path(arc, "runs/kallisto/kallisto_results")
-inMetadataFile <- file.path(arc, "runs/merged_isa_metadata/out/merged_isa.tsv")
-inMetadataSample <- "Source.Name"
-inMetadataFactor <- "Factor..Photosynthesis.mode."
-
-
-```
-
-## Import kallisto count data
-
-```{r}
-
-files <- dir(inKallistoResults, recursive = T, full.names = T ,"abundance.h5")
-names(files) <- dir(inKallistoResults)
-
-txi <- tximport(files, type = "kallisto", txOut = TRUE)
-
-head(txi$counts)
-
-```
-
-
-## Read sample metadata
-
-```{r}
-
-
-samples_metadata <- read.table(file = inMetadataFile, sep = "\t")
-
-samples <- samples_metadata[order(samples_metadata[[inMetadataSample]]), c(inMetadataSample, inMetadataFactor)]
-colnames(samples)[1:2] <- c("sampleID", "condition")
-
-rownames(samples) <- samples$sampleID
-
-
-```
-
-## DESeq
-
-```{r}
-dds <- DESeqDataSetFromTximport(txi,
-                                   colData = samples,
-                                   design = ~ condition)
-
-dds <- DESeq(dds)
-
-res <- results(dds)
-res
-
-plotMA(res, ylim=c(-2,2))
-
-
-```
-
-
-```{r}
-vsd <- vst(dds, blind=FALSE)
-
-pcaData <- plotPCA(vsd, intgroup=c("condition"), returnData=TRUE)
-percentVar <- round(100 * attr(pcaData, "percentVar"))
-ggplot(pcaData, aes(PC1, PC2, color=condition)) +
-  geom_point(size=3) +
-  xlab(paste0("PC1: ",percentVar[1],"% variance")) +
-  ylab(paste0("PC2: ",percentVar[2],"% variance")) + 
-  coord_fixed()
-  
-```
-
-
-

workflows/deseq2/deseq2.cwl

@@ -13,30 +13,27 @@ requirements:
     networkAccess: true
 baseCommand: [RScript, deseq2.R]
 inputs:
-  arcPath:
-    type: string
-    inputBinding:
-      position: 1
   inKallistoResults:
-    type: string
+    type: Directory
     inputBinding:
-      position: 2
+      position: 1
   inMetadataFile:
-    type: string
+    type: File
     inputBinding:
-      position: 3
+      position: 2
   inMetadataSample:
     type: string
     inputBinding:
-      position: 4
+      position: 3
   inMetadataFactor:
     type: string[]
     inputBinding:
-      position: 5
+      position: 4
 
 outputs:
   output:
     type: File[]
     outputBinding:
       glob:
-        - "*"
+        - "*.png"
+        - "*.csv"

workflows/isaSampleToRawDataSeq/README.md

 
 
 ```bash
-dotnet fsi isaSampleToRawDataSeq.fsx ../../ Talinum_RNASeq_minimal 1 rnaseq-samples
-
 
+dotnet fsi isaSampleToRawDataSeq.fsx ../../ Talinum_RNASeq_minimal 1 rnaseq-samples
 
 ```

workflows/isaSampleToRawDataSeq/isaSampleToRawDataSeq.fsx

@@ -5,8 +5,6 @@
 #r "nuget: ARCtrl.NET, 2.0.2"
 #r "nuget: ARCtrl.QueryModel, 2.0.2"
 
-// open FsSpreadsheet.CsvIO
-open FsSpreadsheet.Net
 open System.IO
 open ARCtrl.NET
 open ARCtrl
@@ -17,19 +15,17 @@ open ARCtrl.QueryModel
 let args : string array = fsi.CommandLineArgs |> Array.tail
 let arcPath = args.[0]
 let assayName = args.[1]
-let outName = args.[2]
-
-let startingNodeNum = args.[3] |> int
+let startingNodeNum = args.[2] |> int
+let outName = args.[3]
 
 
 // test parameters
+let source = __SOURCE_DIRECTORY__
+let arcPath = Path.Combine(source, "../../")
+let assayName = "Talinum_RNASeq_minimal"
+let startingNodeNum = 1
+let outName = "rnaseq-samples"
 
-// let source = __SOURCE_DIRECTORY__
-// let arcPath = Path.Combine(source, "../../")
-
-// let assayName = "pick2012_illumina_rnaseq"
-
-// let outName = "out.csv"
 
 // Load ARC
 
@@ -55,7 +51,7 @@ let headers = [
             CompositeHeader.Component v.Category
         else failwithf "what the f is %O" v
 
-    CompositeHeader.Output IOType.RawDataFile
+    CompositeHeader.Output IOType.Data
 ]
 
 // Create rows