changeset 1:629323b5fc0c draft

update tool

--- a/rmarkdown_deseq2_count_matrix.Rmd	Sat Dec 30 00:25:38 2017 -0500
+++ b/rmarkdown_deseq2_count_matrix.Rmd	Sat Dec 30 16:39:39 2017 -0500
@@ -16,24 +16,92 @@
 ```
 
 
-## User input
+# User input
 
 ```{r 'user input'}
 df = data.frame(name = names(opt)[-1],
                 value = unlist(opt))
-df
+datatable(df, rownames = FALSE)
 ```
 
 
-## Count Matrix
+# Count Matrix
+
+Display the first 100 rows of count data matrix.
 
 ```{r 'count matrix'}
 count_data = read.table(opt$count_data)
-head(count_data, 10)
+col_names = trimws(strsplit(opt$count_matrix_column_names, ',')[[1]])[1:ncol(count_data)]
+col_names = col_names[!is.na(col_names)]
+colnames(count_data)[1:length(col_names)] = col_names
+datatable(head(count_data, 100))
+```
+
+# Column Data
+
+```{r 'column data'}
+col_data = read.table(opt$col_data, 
+                      stringsAsFactors = FALSE, sep=',', header = TRUE, row.names = 1)
+datatable(col_data)
+```
+
+# Match sample names
+
+The goal of this step is to rearrange the rows of the column data matrix so that the samples rows in the count data matrix and the sample columns in the count data matrix are in the same order.
+
+```{r 'match sample names'}
+col_data = col_data[col_names, ]
+datatable(col_data)
 ```
 
-```{r 'ste[ 2'}
+# DESeqDataSet
+
+```{r 'DeseqDataSet'}
+dds = DESeqDataSetFromMatrix(countData = count_data,
+                             colData = col_data,
+                             design = formula(opt$design_formula))
+dds
+```
+
+Pre-filter low count genes
 
+```{r 'pre-filtering'}
+keep = rowSums(counts(dds)) >= 10
+dds = dds[keep, ]
+dds
+```
+
+# Differential expression analysis
+
+```{r 'differential expression analysis'}
+dds = DESeq(dds)
+# res = results(dds, contrast = c(opt$contrast_condition, opt$treatment, opt$control))
+res = results(dds)
+resultsNames(dds)
+if(nrow(res) > 500) {
+  cat('The result table has more than 500 rows. Only 500 rows are randomly selected to dispaly.')
+  datatable(as.data.frame(res)[sample(1:nrow(res), 500), ])
+} else {
+  datatable(as.data.frame(res))
+}
 ```
 
 
+
+```{r 'write results into csv'}
+#Write results into a CSV file.
+write.csv(res, 'differential_genes.csv')
+```
+
+# MAplot
+
+```{r}
+plotMA(res)
+```
+
+
+```{r 'save R objects'}
+# Save R objects to a file
+save(dds, opt, file = 'deseq2.RData')
+```
+

--- a/rmarkdown_deseq2_count_matrix.xml	Sat Dec 30 00:25:38 2017 -0500
+++ b/rmarkdown_deseq2_count_matrix.xml	Sat Dec 30 16:39:39 2017 -0500
@@ -17,11 +17,12 @@
 Rscript '${__tool_directory__}/rmarkdown_deseq2_count_matrix_render.R'
                         -e $echo
                         -c $count_data
+                        -n '$count_matrix_column_names'
                         -C $col_data
                         -D '$design_formula'
                         -k $contrast_condition
                         -T $treatment
-                        -k $control
+                        -K $control
                         -r $report
                         -d $report.files_path
                         -s $sink_message
@@ -31,21 +32,32 @@
         <param type="boolean" name="echo" label="Display analysis code in report?" optional="False" checked="False"
                truevalue="TRUE" falsevalue="FALSE"/>
         <param type="data" name="count_data" label="count data" optional="False" format="txt,csv" multiple="False"/>
-        <param type="data" name="col_data" label="column data" optional="False" format="txt,csv" multiple="False"/>
+        <param type="text" name="count_matrix_column_names" label="column names"
+               help="This field is used to update the column names of count data matrix. Column names are separated by comma (,). For example, 'column_1, column_2, column_3, column_4'"
+               optional="True" area="True" size="5x25"></param>
+        <param type="data" name="col_data" label="column data (from file)" optional="False" format="txt,csv" multiple="False"/>
         <param type="text" name="design_formula" label="design formula"
                help="The simplest design formula for differential expression would be ~ condition, where condition&#13;&#10;                     is a column in colData(dds) that specifies which of two (or more groups) the samples belong to"
-               optional="False" value="~ condition_1 + condition_2"/>
+               optional="False" value="~ condition_1 + condition_2">
+            <sanitizer>
+                <valid>
+                    <add value="~"/>
+                </valid>
+            </sanitizer>
+        </param>
         <param type="text" name="contrast_condition" label="condition"
                help="A condition name specified in the design formula" optional="False" value="condition_1"/>
         <param type="text" name="treatment" label="treatment" help="the name of the numerator level for the fold change"
-               optional="False"/>
+               optional="False" value="treated"/>
         <param type="text" name="control" label="control" help="the name of the denominator level for the fold change"
-               optional="False"/>
+               optional="False" value="untreated"/>
     </inputs>
     <outputs>
         <data name="report" format="html" label="DESeq2 report" hidden="false"/>
         <data name="differential_genes" format="csv" label="differential genes" from_work_dir="differential_genes.csv"
               hidden="false"/>
+        <data name="deseq2" format="rdata" label="DESeq2 R objects" from_work_dir="deseq2.RData"
+              hidden="false"/>
         <data name="sink_message" format="txt" label="Warnings and Errors" from_work_dir="warnings_and_errors.txt"
               hidden="false"/>
     </outputs>

--- a/rmarkdown_deseq2_count_matrix_render.R	Sat Dec 30 00:25:38 2017 -0500
+++ b/rmarkdown_deseq2_count_matrix_render.R	Sat Dec 30 16:39:39 2017 -0500
@@ -2,6 +2,8 @@
 library(rmarkdown)
 library(htmltools)
 library(dplyr)
+library(DT)
+library(DESeq2)
 
 ##============ Sink warnings and errors to a file ==============
 ## use the sink() function to wrap all code within it.
@@ -31,8 +33,10 @@
   #+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
   args_list=list()
   ##------- 1. input data ---------------------
-  args_list$ECHO = c('echo', 'e', '1', 'character')
+  args_list$ECHO = c('echo', 'e', '1', 'logical')
   args_list$c = c('count_data', 'c', '1', 'character')
+  args_list$n = c('count_matrix_column_names', 'n', '1', 'character')
+  args_list$n = c('count_matrix_column_names', 'n', '1', 'character')
   args_list$C = c('col_data', 'C', '1', 'character')
   args_list$D = c('design_formula', 'D', '1', 'character')
   args_list$k = c('contrast_condition', 'k', '1', 'character')