Updated changelog and docs; reverted R package installation

CHANGELOG.md

@@ -1,4 +1,10 @@
 ## CARLISLE development version
+- Bug fixes (#127, @epehrsson)
+- Removes single-sample group check for DESeq.
+- Increases memory for DESeq.
+- Ensures control replicate number is an integer.
+- Fixes FDR cutoff misassigned to log2FC cutoff.
+- Fixes `no_dedup` variable names in library normalization scripts.
 
 ## CARLISLE v2.5.0
 - Refactors R packages to a common source location (#118, @slsevilla)

docs/user-guide/preparing-files.md

@@ -158,3 +158,5 @@ An example contrast file:
 | condition1 | condition2 |
 | --- | --- |
 | MOC1_siSmyd3_2m_25_HCHO | MOC1_siNC_2m_25_HCHO |
+
+**Note**: you must have more than one sample per condition in order to perform differential analysis with DESeq2

workflow/scripts/_diff_markdown.Rmd

@@ -39,7 +39,7 @@ source(params$carlisle_functions)
 Rlib_dir=params$Rlib_dir
 Rpkg_config=params$Rpkg_config
 print(paste0("Using the lib.loc location: ",Rlib_dir))
-#assign(".lib.loc", Rlib_dir, envir = environment(.libPaths))
+assign(".lib.loc", Rlib_dir, envir = environment(.libPaths))
 .libPaths()
 
 # read in package info
@@ -51,11 +51,11 @@ pkg_df
 CARLISLE_HANDLE_PACKAGES(pkg_df)
 
 ##handle ELBOW separately - it requires an older version of bioconductor / R
-#if("ELBOW" %in% new.packages){
-#  install.packages(paste0(Rlib_dir,"ELBOW_1.10.0.tar.gz"), 
-#    repos = NULL, type="source", INSTALL_opts = '--no-lock')
-#  new.packages=new.packages[names(new.packages) != "ELBOW"]
-#}
+if("ELBOW" %in% new.packages){
+  install.packages(paste0(Rlib_dir,"ELBOW_1.10.0.tar.gz"), 
+    repos = NULL, type="source", INSTALL_opts = '--no-lock')
+  new.packages=new.packages[names(new.packages) != "ELBOW"]
+}
 ```
 
 ## Loading SampleInfo and Counts