v3.1.2

v3.1.2

ChangeLog

@@ -10,12 +10,124 @@
 # changes in development version since last release 
 ################################################################################
 
+################################################################################  
+################################################################################
+
+################################################################################
+### version 3.1.2
+################################################################################  
+
+# IntaRNA
+- `--outSep` = user-defined column separator for tabular CSV output
+- bugfix non-overlapping suboptimal enumeration
+- bugfix noLP optimization (missing case of direct left-stack extension)
+- CSV output
+  - new ensemble energy and partition function output for intra-molecular 
+    structures formed by seq1 and seq2 (`Eall1, Eall2, Zall1, Zall2`)
+  - new total energy output `Etotal` = (E+Eall1+Eall2) and 
+    `EallTotal` = (Eall+Eall1+Eall2)
+  - new `RT` output
+
+# auxiliary R scripts
+- plotRegions.py - visualization of sequence regions covered by IntaRNA 
+  predictions, similar to the IntaRNA webserver output (thanks to @dgelsin)
+
+################################################################################
+
+191030 Martin Raden
+ * IntaRNA/OutputHandlerCsv :
+   + RT
+   * string2list() :
+     + support of '*' encoding to generate full list
+ * bin/CommandLineParsing :
+   + docu and implementation of '*' outCsvCol behaviour  
+ * IntaRNA/InteractionEnergyBasePair :
+   + computeIntraEall() : computes Eall1|2 via NussinovHandler
+   * getEall1|2() : call computeIntraEall if needed
+ * README.md
+   + RT CSV col
+
+191029 Martin Raden
+ * IntaRNA/PredictorMfe :
+   * getNextBest() :
+     * bugfix: energy check was applying duplicated ED values 
+       (thanks to Jens Georg)
+ * IntaRNA/PredictorMfe2dSeedExtension :
+ * IntaRNA/PredictorMfe2dHeuristicSeedExtension :
+ * IntaRNA/PredictorMfeEns2dSeedExtension :
+   * fillHybrid*_left() : 
+   * traceBack() : 
+     * bugfix: missing case in noLP mode (direct left-stack extension)
+ * IntaRNA/AccessibilityVrna :
+   + addConstraints() : dedicated function to add constraint to VrnaFoldCompound
+   * fillByRNAplfold() : using addConstraints()
+ * IntaRNA/InteractionEnergy :
+   + getBoltzmannWeight( Z_type ) : conversion from kcal/mol-based energies
+   + getEall1|2() : ensemble energy for seq1|2
+ * IntaRNA/InteractionEnergyBasePair :
+   + getEall1|2() : NOT IMPLEMENTED YET
+ * IntaRNA/InteractionEnergyIdxOffset :
+   + getEall1|2() : forward to wrapped energy handler
+ * IntaRNA/InteractionEnergyVrna :
+   + Eall1|2 : ensemble energies for seq1|2
+   + getEall1|2() : lazy computation of Eall1|2 using computeIntraEall()
+   + computeIntraEall() : Eall* computation via vrna_pf() using the respective
+     accessibility constraints
+ * IntaRNA/OutputHandlerEnsemble :
+   - no output of Zall
+   + output of RT, Eall1, Eall2, EallTotal 
+ * IntaRNA/OutputHandlerCsv :
+   + Eall1, Eall2, EallTotal, Etotal, Zall1, Zall2
+ * README.md :
+   + docu of new CSV columns (Eall1, Eall2, Zall1, Zall2, Etotal, EallTotal)
+   * ensemble output docu updated
+
+191021 Martin Raden
+ + R/plotRegions.R :  visualization of sequence regions covered by RRI
+ + R/README.md : plotRegions docu
+ * README.md : link to R/README.md
+
+191008 Martin Raden
+ * IntaRNA/OutputHandlerCsv :
+   + bpList output : list of base pairs
+ * README.md :
+   + bpList 
+   + outSep
+
+191008 Martin Raden
+ * bin/CommandLineParsing :
+   + 'outSep' argument to set column separator for tabular CSV output
+   * outSep applied to PredictionTracker*
+ * IntaRNA/PredictionTracker* : 
+   + explicit output column separator
+ * IntaRNA/OutputHandlerCsv :
+   * needsZall() :
+   * needBPs() :
+     - colSep argument : obsolete
+
+191007 Martin Raden
+ * bin/CommandLineParsing :
+   * bugfix accNoLP and accNoGUend checks for energy!=V
+ * IntaRNA/NussinovHandler :
+   * getQ() :
+     * fix: return 1 if (i==j)
+ * IntaRNA/AccessibilityVrna :
+   * construction() : 
+     + bugfix: missing ED initialization for short sequences
+ * tests/AccessibilityBasePair :
+   + short sequence test
+ + tests/AccessibilityVrna  :
+   + short sequence test to validate ED initialization
+
+################################################################################
+### version 3.1.1 
+################################################################################  
+
 # IntaRNA
 - base pairs details only computed if needed for output (speedup for large -n)
 - predefined parameter sets for loading (Turner04, Turner99, Andronescu07)
 - 'tRegion' and 'qRegion' now available for and applied to multi-sequence input
 
-################################################################################  
 ################################################################################  
 
 190924 Martin Raden

R/README.md

@@ -0,0 +1,35 @@
+
+# Auxiliary R scripts of the IntaRNA package
+
+
+
+# `plotRegions.R` - Visualization of RRI-covered regions
+
+To visualize sequences' regions covered by RNA-RNA interactions predicted by
+IntaRNA, you can use `plotRegions.R` by providing the following arguments (in 
+the given order)
+
+1. CSV-IntaRNA output file (semicolon separated) covering the columns `start,end,id`
+  with suffix `1` or `2` to plot target or query regions, respectively
+2. `1` or `2` to select whether to plot target or query regions
+3. output file name with a file-format-specific suffix from `.pdf`, `.png`, 
+  `.svg`, `.eps`, `.ps`, `.jpeg`, `.tiff`
+
+An example is given below, when calling
+```bash
+Rscript --vanilla plotRegions.R pred.csv 1 regions.png
+```
+
+with `pred.csv` containing
+```
+id1;start1;end1;id2;start2;end2
+b0001;266;273;query;116;123
+b0002;204;231;query;85;111
+b0003;229;262;query;96;125
+b0004;265;300;query;10;38
+b0005;281;295;query;5;22
+```
+
+will produce the output
+![plotRegions.R example](plotRegions.example.png)
+

R/plotRegions.R

@@ -0,0 +1,151 @@
+#!/usr/bin/env Rscript
+
+####################################################################
+# Visualization of sequence regions covered by RNA-RNA interactions
+# predicted by IntaRNA.
+#
+# arguments: <IntaRNA-output-CSV> <1|2> <output-plot-file>
+# 1 <IntaRNA-output-CSV> = ";"-separated CSV output of IntaRNA
+# 2 <1|2> = suffix of "start,end,id" CSV cols to plot
+# 3 <output-plot-file> = file name of the otuput figure suffixed
+#     by one of ".pdf",".png",".svg",".eps",".ps",".jpeg",".tiff"
+#
+# example call:
+#
+# Rscript --slave -f plotRegions.R --args pred.csv 1 regions.png
+#
+# This script is part of the IntaRNA source code package. See
+# respective licence and documentation for further information.
+#
+# https://github.com/BackofenLab/IntaRNA
+#
+####################################################################
+# check and load dependencies
+####################################################################
+
+options(warn=-1)
+suppressPackageStartupMessages(require(ggplot2))
+suppressPackageStartupMessages(require(ggalt))
+suppressPackageStartupMessages(require(cowplot)) # cowplot starts with a note
+options(warn=0)
+
+theme_set(theme_cowplot())
+
+####################################################################
+# get command line arguments
+####################################################################
+
+args = commandArgs(trailingOnly=TRUE)
+# check and parse
+if (length(args)!=3) { stop("call with <intarna-csv-output> <1|2> <out-file-of-plot>", call.=FALSE) }
+
+intarnaOutputFile = args[1];
+if (!file.exists(intarnaOutputFile )) { stop("intarna-csv-output file '", intarnaOutputFile, "' does not exist!", call.=FALSE) }
+
+seqNr = args[2];
+if (seqNr != "1" &&  seqNr != "2") { stop("second call argument as to be '1' or '2' to specify which regions to plot"); }
+# compile column names
+id = paste("id",seqNr,sep="");
+start = paste("start",seqNr,sep="");
+end = paste("end",seqNr,sep="");
+
+outFile = args[3];
+fileExtensions = c(".pdf",".png",".svg",".eps",".ps",".jpeg",".tiff");
+outFileExtOk = FALSE;
+for( ext in fileExtensions ) { outFileExtOk = outFileExtOk || endsWith(outFile,ext); }
+if ( !outFileExtOk ) {stop("<out-file-of-plot> has to have one of the following file extensions ",paste(fileExtensions,sep=" "), call.=FALSE);} 
+
+# if set to some x-position, this will trigger the plotting of a vertical line at that location
+xVline = NA;
+#xVline = 250; # DEBUG TEST VALUE
+
+
+####################################################################
+# parse IntaRNA output
+####################################################################
+
+d = read.csv2( intarnaOutputFile )
+# check if all columns present
+for( x in c(id,start,end)) {
+	if (!is.element(x, colnames(d))) { 
+		stop("'",id,"' is not among the column names of '",intarnaOutputFile,"'", call.=FALSE); 
+	}
+}
+
+####################################################################
+# create count plot
+####################################################################
+
+allPos = c();
+for( i in 1:nrow(d) ) {
+	allPos = c( allPos, d[i,start]:d[i,end] );
+}
+allPos = as.data.frame(allPos,ncol=1)
+#allPos # DEBUG OUT
+
+coveragePlot =
+		ggplot( allPos, aes(x=allPos, stat(count))) +
+		geom_density() +
+		ylab("coverage") +
+		xlab("position") +
+		scale_y_continuous(position = "right", expand=expand_scale(mult = c(0, .02))) +
+		scale_x_continuous(expand = c(0, 0)) +
+		theme(	axis.title.x=element_blank() )
+
+if ( ! is.na(xVline)) {
+	coveragePlot = coveragePlot +
+			geom_vline(aes(xintercept=xVline));
+}
+
+####################################################################
+# create region plot
+####################################################################
+
+dRegion = data.frame()
+dRegion[1:nrow(d),1:3] = d[,c(start,end,id)]
+dRegion[1:nrow(d),4] = factor(sprintf("%08d",nrow(d):1))
+colnames(dRegion) = c("start","end","id","idx");
+#dRegion # DEBUG OUT
+
+yLabelScale = 0.6 # if you have to alter for more/less sequence IDs per inch; see <plotHeight> below
+
+regionPlot = 
+		ggplot(dRegion, aes(x=start,xend=end,y=idx)) +
+		geom_dumbbell(color="dodgerblue", size=2) +
+		xlab("position") + 
+		scale_x_continuous( expand = c(0, 0) ) +
+		ylab("") +
+		scale_y_discrete(position = "right", breaks=dRegion$idx, labels=dRegion$id) +
+		geom_vline(aes(xintercept=min(allPos))) +
+		theme(panel.grid.major.y=element_line(size=0.7,color="lightgray")
+				, axis.text.y=element_text(size=rel(yLabelScale))
+		)
+
+if ( ! is.na(xVline)) {
+	regionPlot = regionPlot + 
+			geom_vline(aes(xintercept=xVline));
+}
+
+####################################################################
+# plot to file
+####################################################################
+
+plotWidth = 6
+plotHeightDensity = 2
+plotHeight = plotHeightDensity + max(2,nrow(d)/9)
+plotHeightDensityRel = plotHeightDensity / plotHeight
+
+
+plot_grid( coveragePlot, regionPlot
+		, nrow=2, ncol=1
+		, align = "hv"
+		, axis = "r"
+		, rel_heights= c( plotHeightDensityRel, 1.0-plotHeightDensityRel)
+)
+
+ggsave( outFile
+		, width= plotWidth
+		, height= plotHeight
+);
+
+#############################################################EOF