Major fixes (see NEWS) for new version 0.2.0

.gitignore

@@ -3,3 +3,5 @@ data/tx_hg38.RData
 .Rproj.user
 *.RData
 *.Rproj
+
+.DS_Store

DESCRIPTION

@@ -7,26 +7,27 @@ Maintainer: Maurits Evers <maurits.evers@anu.edu.au>
 Description: Transcriptome-wide analysis of RNA modifications.
 Imports:
     AnnotationDbi,
-    beanplot,
     Biostrings,
+    beanplot,
     GenomeInfoDb,
     gplots,
     graphics,
     grDevices,
     methods,
     RSQLite,
-    rtracklayer,
     S4Vectors,
     stats,
     utils
 Depends:
     GenomicRanges,
     GenomicFeatures,
-    IRanges
+    IRanges,
+    R (>= 3.5.0)
 Suggests:
     knitr,
     testthat
 License: GPL-3 | file LICENSE
 LazyData: false
 RoxygenNote: 6.1.1
+Encoding: UTF-8
 VignetteBuilder: knitr

NAMESPACE

@@ -8,14 +8,15 @@ export(CheckClassTxLocRef)
 export(CheckPkgDependencies)
 export(CollapseTxBySec)
 export(DedupeBasedOnNearestRef)
+export(DownsampleTxLoc)
 export(EstimateCIFromBS)
 export(FilterTxLoc)
 export(GenerateNull)
 export(GetColPal)
 export(GetEEJunct)
 export(GetGC)
 export(GetGeneIds)
-export(GetRelDistNearest)
+export(GetMotifLoc)
 export(GetSplicingSites)
 export(GetTxBySec)
 export(GetTxDb)
@@ -24,20 +25,17 @@ export(IsEmptyChar)
 export(LoadRefTx)
 export(PerformSanityCheck)
 export(PlotAbundance.generic)
-export(PlotOverlap)
+export(PlotGC)
 export(PlotRelDistDistribution)
 export(PlotRelDistEnrichment)
 export(PlotSectionDistribution)
 export(PlotSectionEnrichment)
 export(PlotSeqLogo)
 export(PlotSpatialDistribution)
 export(PlotSpatialEnrichment)
-export(PlotTxSecLength)
-export(PlotTxSecLength.bean)
 export(ReadBED)
 export(SmartMap)
 export(SubsampleTxLoc)
-export(TxLoc2GRangesList)
 export(Unfactor)
 export(WriteBED)
 export(WriteCSV)
@@ -46,6 +44,7 @@ exportMethods(GetId)
 exportMethods(GetLoci)
 exportMethods(GetNumberOfLoci)
 exportMethods(GetRef)
+exportMethods(GetRegions)
 exportMethods(GetVersion)
 exportMethods(show)
 import(AnnotationDbi)
@@ -56,17 +55,17 @@ import(GenomicRanges)
 import(IRanges)
 import(RSQLite)
 import(methods)
+importFrom(Biostrings,letterFrequency)
 importFrom(Biostrings,vmatchPattern)
 importFrom(S4Vectors,DataFrame)
+importFrom(S4Vectors,queryHits)
+importFrom(S4Vectors,subjectHits)
 importFrom(beanplot,beanplot)
-importFrom(gplots,venn)
 importFrom(grDevices,col2rgb)
-importFrom(grDevices,rainbow)
 importFrom(grDevices,rgb)
 importFrom(graphics,abline)
 importFrom(graphics,axis)
 importFrom(graphics,barplot)
-importFrom(graphics,boxplot)
 importFrom(graphics,hist)
 importFrom(graphics,legend)
 importFrom(graphics,lines)

NEWS

@@ -2,36 +2,75 @@
 
 ## Version 0.2.0
 
-### Major
-
-- Major changes to functions in BuildTx.R to reflect critical changes in
+- RNAModR now requires R >= 3.5.0
+- Major changes to functions in buildTxMethods.R to reflect critical changes in
   `GenomicRanges_1.32.0` and `RMariaDB` that completely broke functionality
   of `BuildTx`
 - Major changes to structure of `txLoc` objects; `txLoc` is now a dressed
   `list` of `DataFrame`s (instead of a dressed `list` of `data.frame`s). This
   has the advantage that we can now store arbitrary R objects in columns, such
   as `GRanges`, `DNAStringSet` etc.
-- In `classes.R`: Added `show` method; this means that we can simply type the
-  name of the `txLoc` object into an R terminal, hit <Enter> and get a summary.
+
+- In `classes.R`: Added `txLoc` S4 class method `show`; this means that we can
+  simply type the name of the `txLoc` object into an R terminal, hit <Enter>
+  and get a summary
 - In `classes.R`: Removed `info` method
+- In `classes.R`: Added `txLoc` S4 class method `GetRegions`
+- In `classes.R`: Tidied roxygen2 documentation; all `txLoc` accessors are now
+  document under `txLoc-accessors` in the reference manual
+
 - In `IO.R`: Renamed `WriteTxLocToBED` to `WriteBED`; chromosome names are
   now always formatted in UCSC style (i.e. chr1, chr2, chrM, ...)
 - In `IO.R`: Renamed `WriteTxLocToCSV` to `WriteCSV`
-- In `IO.R`: Removed `WriteFeatToBED`
-- In `IO.R`: Removed export of `ReadDBN` (which is needed by `GetMFE`, which
-  in turn is also not exported)
+- In `IO.R`: Removed `WriteFeatToBED` and removed `rtracklayer` dependecy
+- In `IO.R`: Removed `ReadDBN` (which was needed by `GetMFE`, which in turn
+  has also been removed)
+
 - In `methods.R`: Removed `method = "perm"` from `GenerateNull`
+- In `methods.R`: Removed `GetMFE`, because such functions should be managed
+  and defined on a per-use and per-user basis
+- In `methods.R`: Removed `filter` argument from `GetMotifLoc`; this is
+  because filtering should be done through `FilterTxLoc`
+- In `methods.R`: Removed `GenerateNull.new`, as it was a function for testing
+  purposes and served no real purpose
+- In `methods.R`: Removed `GetLocus.MapFromTranscripts`, as it served no
+  purpose
+- In `methods.R`: Refactored and tidied `GetGC`
+
 - In `plot.R`: Removed `filter` argument from `PlotSectionDistribution`,
-  `PlotSpatialDistribution`, `PlotSectionEnrichment`, `PlotSpatialEnrichment`;
-  this is because filtering should be done through `FilterTxLoc`
-- In `plot.R`: Removed `PlotTxSecLength` and `PlotTxSecLength.bean`, because
-  such functions should be managed and defined on a per-use and per-user basis.
+  `PlotSpatialDistribution`, `PlotSectionEnrichment`, `PlotSpatialEnrichment`,
+  `PlotGC`; this is because filtering should be done through `FilterTxLoc`
+- In `plot.R`: Removed `PlotTxSecLength`, `PlotTxSecLength.bean` and
+  `PlotOverlap`, because such functions should be managed and defined on a
+  per-use and per-user basis
+- In `plot.R`: Refactored and tidied `PlotSectionDistribution`
+- In `plot.R`: Refactored and tidied `PlotSectionEnrichment`
+- In `plot.R`: Refactored and tidied `PlotSpatialEnrichment`
 
-### Minor
+- In `helper.R`: Added `DownsampleTxLoc`; `DownsampleTxLoc` is a function that
+  downsamples the number of sites from on `txLoc` object to the number of sites
+  from a second `txLoc` object; downsampling is transcript region specific
+- In `helper.R`: Tidied and refactored `SubsampleTxLoc`; `SubsampleTxLoc`
+  generates a subsample of sites per transcript region, based on a vector of
+  fractions
+- In `helper.R`: Removed `internal` keyword from `FilterTxLoc`,
+  `SubsampleTxLoc`
+- In `helper.R`: Refactored and tidied `FilterTxLoc`
+- In `helper.R`: Fixed bug in `CheckClassTxLocConsistency` that prevented the
+  function from producing an error when regions did not match
+- In `helper.R`: Removed `filter` argument from `TxLoc2GRangesList`; this is
+  because filtering should be done through `FilterTxLoc`
+- In `helper.R`: `GetRelDistNearest` and `TxLoc2GRangesList` are now no
+  longer exported; this is because these functions should really never be
+  called directly by user.
 
-- In `plot.R`: Tidied `PlotSectionDistribution`
-- In `plot.R`: Tidied `PlotSectionEnrichment`
 - Renamed "transcript sections" to "transcript regions" in all documentation
+- Fixed "no visible binding for global variable <xyz>" issues that derive from
+  `with(..., x = xyz)` statements by setting `xyz <- NULL` (this is the
+  recommended `data.table` way)
+- DESCRIPTION: Added `Encoding: UTF-8` to silence R CMD check and roxygen2 from
+  complaining about encoding
+
 
 ## Version 0.1.0
 

R/IO.R

@@ -203,36 +203,3 @@ WriteCSV <- function(txLoc,
     cat(sprintf("Output in file %s.\n", file))
 
 }
-
-
-#' Read a DBN file.
-#'
-#' Read a DBN file. See 'Details'.
-#'
-#' The function reads in a DBN (dot-bracket) structure file, and
-#' returns a \code{dataframe} with the following data columns:
-#' \enumerate{
-#' \item Column 1: Sequence ID
-#' \item Column 2: Length of the sequence (in nt)
-#' \item Column 3: Mean free energy (MFE)
-#' }
-#' 
-#' @param file A character string; specifies the input DBN file.
-#'
-#' @return A \code{dataframe} object. See 'Details'.
-#'
-#' @author Maurits Evers, \email{maurits.evers@@anu.edu.au}
-ReadDBN <- function(file) {
-    if (!file.exists(file)) {
-        ss <- sprintf("Could not open %s.", file)
-        stop(ss)
-    }
-    d <- as.data.frame(matrix(readLines(file), ncol = 3, byrow = TRUE),
-                       stringsAsFactors = FALSE)
-    d[, 1] <- gsub("^>", "", d[, 1])
-    d[, 2] <- nchar(d[, 2])
-    d[, 3] <- gsub("^[\\(\\.\\)]+\\s", "", d[, 3])
-    d[ ,3] <- as.numeric(gsub("[\\(\\)]", "", d[, 3]))
-    colnames(d) <- c("id", "siteSeqLength", "siteMFE")
-    return(d)
-}