Mercurial > repos > marie-tremblay-metatoul > nmr_annotation

changeset 0:a8e85c07dc7b draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author marie-tremblay-metatoul
date Tue, 30 Jan 2018 05:37:10 -0500
parents
children b55559a2854f
files nmr_annotation/.shed nmr_annotation/DrawSpec.R nmr_annotation/Library.Rdata nmr_annotation/ReadMe.txt nmr_annotation/asics_wrapper.R nmr_annotation/asics_xml.xml nmr_annotation/static/images/NmrAnnotation_ASICS.png nmr_annotation/test-data/ADG_007.zip nmr_annotation/test-data/MTBLS1_177_NMR_Annotation.pdf nmr_annotation/test-data/MTBLS1_177_NMR_Annotation_proportionEstimation.tabular
diffstat 10 files changed, 559 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/.shed	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,7 @@
+categories: [Metabolomics]
+description: '[Metabolomics][W4M][NMR] NMR Annotation - Annotation of complex mixture NMR spectra and metabolite proportion estimation'
+homepage_url: http://workflow4metabolomics.org
+long_description: 'Part of the W4M project: http://workflow4metabolomics.org'
+name: nmr_annotation
+owner: marie-tremblay-metatoul
+remote_repository_url: https://github.com/workflow4metabolomics/nmr_annotation
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/DrawSpec.R	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,74 @@
+drawSpec <- function (X, startP = -1, endP = -1, groupLabel = NULL, useLog = -1, highBound = -1, lowBound = -1, 
+                      xlab = NULL, ylab = NULL, main = NULL, nAxisPos = 4, offside = 0) 
+{
+  groupLabel_name = groupLabel
+  X = as.data.frame(X)
+#  colnames(X) = c(1:ncol(X))
+  X = as.matrix(X)
+  if (highBound != -1) {
+    for (i in 1:nrow(X)) {
+      myIndex = which(X[i, ] > highBound)
+      X[i, myIndex] = highBound
+    }
+  }
+  if (lowBound != -1) {
+    for (i in 1:nrow(X)) {
+      myIndex = which(X[i, ] < lowBound)
+      X[i, myIndex] = lowBound
+    }
+  }
+  if (is.null(groupLabel)) {
+    groupLabel = c(1:nrow(X))
+    groupLabel = as.factor(groupLabel)
+  }
+  else {
+    levels(groupLabel) = c(1:length(levels(groupLabel)))
+  }
+  if (startP == -1) 
+    startP = 1
+  if (endP == -1) 
+    endP = ncol(X)
+  if (is.null(xlab)) {
+    xlab = "index"
+  }
+  if (is.null(ylab)) {
+    ylab = "intensity"
+  }
+  if (is.null(main)) {
+    main = paste(" ", startP + offside, "-", endP + offside)
+  }
+  GraphRange <- c(startP:endP)
+  yn <- X[, GraphRange]
+  if (useLog != -1) 
+    yn = log(yn)
+  if (length(yn) > ncol(X))
+  {
+    plot(yn[1, ], ylim = c(min(yn), max(yn)), type = "n", ylab = ylab, xlab = xlab, main = main, xaxt = "n")
+    tempVal = trunc(length(GraphRange)/nAxisPos)
+    xPos = c(0:nAxisPos) * tempVal
+    axis(1, at = xPos, labels = colnames(X)[xPos + startP + offside])
+    for (i in 1:length(levels(groupLabel))) 
+    {
+      groupLabelIdx = which(groupLabel == levels(groupLabel)[i])
+      color <- palette(rainbow(length(levels(groupLabel))))
+      for (j in 1:length(groupLabelIdx)) 
+      {
+        lines(yn[groupLabelIdx[j], ], col = color[i])
+      }
+    }
+    if (!is.null(groupLabel_name)) 
+    {
+      legendPos = "topleft"
+      legend(legendPos, levels(groupLabel_name), col = as.integer(levels(groupLabel)), text.col = "black", pch = c(19, 19), bg = "gray90")
+    }
+  }
+  if (length(yn) == ncol(X))
+  {
+    plot(yn, ylim = c(min(yn), max(yn)), type = "n", ylab = ylab, xlab = xlab, main = main, xaxt = "n")
+    tempVal = trunc(length(GraphRange)/nAxisPos)
+    xPos = c(0:nAxisPos) * tempVal
+#    axis(1, at = xPos, labels = xPos + startP + offside)
+    axis(1, at = xPos, labels = colnames(X)[xPos + startP + offside])
+    lines(yn)
+  }
+}
\ No newline at end of file
Binary file nmr_annotation/Library.Rdata has changed
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/ReadMe.txt	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,35 @@
+Instructions to integrate the ""NMR Annotation" tool into a local instance of Galaxy
+Version December 2017 M Tremblay-Franco
+
+
+## --- R bin and Packages : --- ##
+R version 3.0.2 (2013-09-25) -- "Frisbee Sailing
+Platform: x86_64-redhat-linux-gnu (64-bit)
+
+Install the "batch" library, necessary for parseCommandArgs function and the "ASICS" library necessary for NMR spectra annotation:
+ - Download package source (*.tar.gz file) from your favorite CRAN (http://www.r-project.org/)
+For example: http://cran.univ-lyon1.fr/
+
+ - Install package in your R session
+install.packages("path/package_name.tar.gz",lib="path",repos=NULL)
+For Example: install.packages("/usr/lib64/R/library/batch_1.1-4.tar",lib="/usr/lib64/R/library",repos=NULL)
+
+ - Finally, load the packages into your R session
+library(batch)
+library(ASICS)
+
+
+## --- Config : --- ##
+ - Edit the file "/galaxy/dist/galaxy-dist/tool_conf.xml" and add 
+<section id="id_name" name="Name">
+  <tool file="path/asics_xml.xml" />
+</section>
+to create a new section containing the asics_xml tool
+or add
+  <tool file="path/asics_xml.xml" />
+in an existing section
+
+ - Put the two files asics_xml.xml, asics_wrapper.R, all the needeed R functions and the Library.RData (including compound reference spectra) in a same directory
+For example, path=/galaxy/dist/galaxy-dist/tools/annotation
+
+Finally, restart Galaxy
\ No newline at end of file
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/asics_wrapper.R	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,191 @@
+#!/usr/local/public/bin/Rscript --vanilla --slave --no-site-file
+
+## 29122017_asics_wrapper.R
+## Remi Servien, Patrick Tardivel, Marie Tremblay-Franco and Gaelle Lefort
+## marie.tremblay-franco@inra.fr
+
+runExampleL <- FALSE
+
+##------------------------------
+## Options
+##------------------------------
+strAsFacL <- options()$stringsAsFactors
+options(stringsAsFactors=FALSE)
+
+
+##------------------------------
+## Libraries loading
+##------------------------------
+# ParseCommandArgs function
+library(batch) 
+library(ASICS) 
+
+
+
+# R script call
+source_local <- function(fname)
+{
+argv <- commandArgs(trailingOnly=FALSE)
+base_dir <- dirname(substring(argv[grep("--file=", argv)], 8))
+source(paste(base_dir, fname, sep="/"))
+}
+#Import the different functions
+source_local("DrawSpec.R")
+
+
+##------------------------------
+## Errors ?????????????????????
+##------------------------------
+
+
+##------------------------------
+## Constants
+##------------------------------
+topEnvC <- environment()
+flagC <- "\n"
+
+
+##------------------------------
+## Script
+##------------------------------
+if(!runExampleL)
+    argLs <- parseCommandArgs(evaluate=FALSE)
+
+# Standards loading
+load(argLs[["standards"]])
+
+## Parameters Loading
+##-------------------
+# Inputs
+## Spectrum to annotate
+zipfile= argLs[["zipfile"]]
+directory=unzip(zipfile, list=F)
+directory=paste(getwd(),strsplit(directory[1],"/")[[1]][2],sep="/")
+
+
+##Exclusion zone(s)
+exclusionZones <- argLs[["zone_exclusion_choices.choice"]]
+exclusionZonesBorders <- NULL
+if (!is.null(argLs$zone_exclusion_left))
+{
+   for(i in which(names(argLs)=="zone_exclusion_left"))
+   {
+#     exclusionZonesBorders <- c(exclusionZonesBorders,list(c(argLs[[i]],argLs[[i+1]])))
+     exclusionZonesBorders <- c(exclusionZonesBorders,argLs[[i]],argLs[[i+1]])
+   }
+}
+
+## Maximal allowed shift
+shift <- argLs[["shift"]]
+
+## Graphical zone(s)
+graphicalZones <- argLs[["zone_graphical_choices.choice"]]
+graphicalZonesBorders <- NULL
+if (!is.null(argLs$zone_exclusion_left))
+{
+  for(i in which(names(argLs)=="zone_graphical_left"))
+  {
+    graphicalZonesBorders <- c(graphicalZonesBorders,list(c(argLs[[i]],argLs[[i+1]])))
+  }
+}
+
+# Outputs
+logOut <- argLs[["logOut"]]
+proportionEstimation <- argLs[["proportionEstimation"]]
+graphOut <- argLs[["graphOut"]]
+
+sink(logOut)
+cat("\tPACKAGE INFO\n")
+# pkgs=c("batch", "ASICS")
+pkgs=c("batch", "ASICS")
+for(pkg in pkgs) {
+    suppressPackageStartupMessages( stopifnot( library(pkg, quietly=TRUE, logical.return=TRUE, character.only=TRUE)))
+    cat(pkg,"\t",as.character(packageVersion(pkg)),"\n",sep="")
+}
+cat("\n")
+
+
+## Checking arguments
+##-------------------
+error.stock <- "\n"
+if(length(error.stock) > 1)
+  stop(error.stock)
+  
+  
+## Computation
+##------------
+annotation.Asics <- ASICS(directory, exclusion.areas=matrix(exclusionZonesBorders, byrow=T, ncol=2), 
+                           max.shift=shift, which.spectra="last", library.metabolites=NULL, 
+                           threshold.noise=0.02, seed=1234, nb.iter.signif=400)
+
+
+## Saving
+##-------
+# Identified metabolites
+metabolites.estimation <- present_metabolites(annotation.Asics)
+colnames(metabolites.estimation) <- c("Metabolite",colnames(metabolites.estimation)[-1])
+write.table(metabolites.estimation,file=argLs$proportionEstimation,row.names=FALSE,quote=FALSE,sep="\t") 
+
+
+## Graphical display
+##------------------
+# Raw and annotated spectra comparison
+pdf(graphOut,onefile=TRUE)
+
+## Graphical output: overlay of raw and estimated spectra
+ppm.metabolites.estimation <- data.frame(round(ppm_grid(annotation.Asics),3), 
+                                         original_mixture(annotation.Asics))
+colnames(ppm.metabolites.estimation) <- c("PPM", "EstimatedProportion")
+ppm.metabolites.estimation <- ppm.metabolites.estimation[order(ppm.metabolites.estimation[,1],decreasing=T), ]
+
+mix <- data.frame(t(ppm.metabolites.estimation[,2]))
+colnames(mix) <- ppm.metabolites.estimation[,1]
+ppm <- ppm.metabolites.estimation[,1]
+
+estimatedMix <- data.frame(round(ppm_grid(annotation.Asics),3), reconstituted_mixture(annotation.Asics))
+colnames(estimatedMix) <- c("PPM","EstimatedProportion")
+estimatedMix <- estimatedMix[order(estimatedMix[,1],decreasing=T), ]
+estimatedMix <- estimatedMix[,2]
+
+## Whole spectra
+GraphRange <- 1:ncol(mix)
+tempVal <- trunc(length(GraphRange)/10)
+xPos <- c(10:0) * tempVal
+plot(1:ncol(mix), mix, type='l', xlab="", main="", xaxt="n", ylab="")
+axis(1, at=xPos, labels=colnames(mix)[xPos + 1])
+lines(estimatedMix, col="red")
+legend("topleft",legend=c("Real Mixture","Estimated Composition"),lty=c(1,1),col=c("black","red"))
+
+## Zoomed spectral window depending on user-selected zone(s)
+graphical.zone.length <- length(graphicalZonesBorders)
+if (graphical.zone.length != 0)
+
+  #   par(mfrow=c(2,1))
+for (g in 1:graphical.zone.length)
+  {
+     print(g)
+     plot(1:length((which(round(as.numeric(colnames(mix)),2) == graphicalZonesBorders[[g]][1])[1]):(which(round(as.numeric(colnames(mix)),2) == max(graphicalZonesBorders[[g]][2],0.5))[1])), 
+          mix[(which(round(as.numeric(colnames(mix)),2) == graphicalZonesBorders[[g]][1])[1]):(which(round(as.numeric(colnames(mix)),2) == max(graphicalZonesBorders[[g]][2],0.5))[1])], type='l', xlab="", ylab="Intensity", main="", xaxt="n")
+     lines(estimatedMix[(which(round(as.numeric(colnames(mix)),2) == graphicalZonesBorders[[g]][1])[1]):(which(round(as.numeric(colnames(mix)),2) == max(graphicalZonesBorders[[g]][2],0.5))[1])],col="red")
+    
+     xPos <- 1
+     nAxisPos <- 4
+     startP <- length(nAxisPos) 
+     endP <- length((which(round(as.numeric(colnames(mix)),2) == graphicalZonesBorders[[g]][1])[1]):(which(round(as.numeric(colnames(mix)),2) == max(graphicalZonesBorders[[g]][2],0.5))[1]))
+     GraphRange <- c(startP:endP)
+     tempVal <- trunc(length(GraphRange)/nAxisPos)
+     xPos <- c(0:nAxisPos) * tempVal
+     noms <- ppm.metabolites.estimation[xPos + which(ppm == round(graphicalZonesBorders[[g]][1],1))[1],1]
+     axis(1, at=xPos, labels=noms)
+   }
+
+invisible(dev.off())
+
+
+## Ending
+##---------------------
+cat("\nEnd of 'NMR annotation' Galaxy module call: ", as.character(Sys.time()), sep="")
+options(stringsAsFactors=strAsFacL)
+rm(list=ls())
+sink()
+
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/asics_xml.xml	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,212 @@
+<tool id="NmrAnnotation" name="NMR_Annotation" version="1.0.0">
+		
+	<description> Annotation of complex mixture NMR spectra and metabolite proportion estimation </description>
+
+	<command>
+        which Rscript ;
+		## Wrapper
+		Rscript '$__tool_directory__/asics_wrapper.R' standards '$__tool_directory__/Library.Rdata'
+
+		## Bruker files
+		zipfile $zip_file
+
+		## Exclusion zone
+		zone_exclusion_choices.choice ${zone_exclusion_choices.choice}
+		#if str($zone_exclusion_choices.choice) == 'yes':
+			#for $i in $zone_exclusion_choices.conditions:
+				zone_exclusion_left ${i.zone_exclusion_left}
+				zone_exclusion_right ${i.zone_exclusion_right}
+			#end for
+		#end if
+
+		## Maximal shift (spectrum alignement)
+		shift $shift
+
+		## Graphical zone
+        zone_graphical_choices.choice ${zone_graphical_choices.choice}
+        #if str($zone_graphical_choices.choice) == 'yes':
+            #for $i in $zone_graphical_choices.conditions:
+                zone_graphical_left ${i.zone_graphical_left}
+                zone_graphical_right ${i.zone_graphical_right}
+            #end for
+        #end if
+
+		## Outputs
+		logOut $logOut
+		proportionEstimation $proportionEstimation
+		graphOut $graphOut
+	</command>
+		
+	<inputs>
+		<param name="zip_file" type="data" format="no_unzip.zip" label="Zip file" />
+
+	<conditional name="zone_exclusion_choices">
+		<param name="choice" type="select" label="Exclusion zone(s)" help="Choose if you want to exclude particular zone(s)" >
+			<option value="yes" > yes </option>
+			<option value="no" selected="true"> no </option>
+		</param>
+		<when value="yes">
+			<repeat name="conditions" title="exclusion zones">
+				<param name="zone_exclusion_left" label="Left exclusion zone border" type="float" value="10.0" />
+				<param name="zone_exclusion_right" label="Right exclusion zone border" type="float" value="10.0" />
+			</repeat>
+		</when>
+	</conditional>
+	
+	<param name="shift" type="float" value="0.01" help="Maximal allowed shift for spectra alignment. Default value is 0.01 ppm" />
+
+	<conditional name="zone_graphical_choices">
+		<param name="choice" type="select" label="Graphical zone(s)" help="Choose if you want to display particular zone(s)" >
+			<option value="yes" > yes </option>
+			<option value="no" selected="true"> no </option>
+		</param>
+		<when value="yes">
+			<repeat name="conditions" title="grapical zones">
+				<param name="zone_graphical_left" label="Left graphical zone border" type="float" value="10.0" />
+				<param name="zone_graphical_right" label="Right graphical zone border" type="float" value="10.0" />
+			</repeat>
+		</when>
+		<when value="no" />
+	</conditional>
+	</inputs>
+
+	<outputs>
+		<data format="txt" name="logOut" label="${tool.name}_log" />
+		<data format="tabular" name="proportionEstimation" label="${tool.name}_proportionEstimation" />
+		<data format="pdf" name="graphOut" label="${tool.name}_graph" />
+	</outputs>
+
+	<stdio>
+		<exit_code range="1:" level="fatal" />
+	</stdio>
+<help>
+
+.. class:: infomark
+
+**Authors** Marie Tremblay-Franco (marie.tremblay-franco@inra.fr), Patrick Tardivel (patrick.tardivel@inra.fr), Rémi Servien (remi.servien@inra.fr) and Gaelle Lefort (gaelle.lefort@inra.fr)
+
+.. class:: infomark
+
+**Please cite**
+
+Tardivel P., Servien R. and Concordet D. Non asymptotic active set properties of lasso-type estimators in small-dimension (submitted)
+Tardivel P., Servien R., Canlet C., Tremblay-Franco M., Debrauwer L. and Concordet D. ASICS: an automatic method for identification and quantification of metabolites in NMR 1D 1H spectra (in preparation)
+
+---------------------------------------------------
+
+==============
+NMR Annotation
+==============
+
+-----------
+Description
+-----------
+
+ASICS, based on a strong statistical theory, handles automatically the metabolite identification and quantification
+ 
+-----------------
+Workflow position
+-----------------
+
+**Upstream tools**
+
+========================= ================= ======= =========
+Name                      output file       format  parameter
+========================= ================= ======= =========
+NA                        NA                NA      NA      
+========================= ================= ======= =========
+
+
+**Downstream tools**
+
+========================= ================= ======= =========
+Name                      output file       format  parameter
+========================= ================= ======= =========
+NA                        NA                NA      NA      
+========================= ================= ======= =========
+
+
+
+-----------
+Input files
+-----------
+
++---------------------------+------------+
+| Parameter : num + label   |   Format   |
++===========================+============+
+| 1 : Choose your inputs    |   zip      |
++---------------------------+------------+
+
+**Choose your inputs**
+
+You have two methods for your inputs:
+
+    | Zip file : You can put a zip file containing one condition (sample): myinputs.zip.
+
+.. image:: ./static/images/Mth_Architecture_Repertoire_Bruker.png
+        :width: 800
+
+----------
+Parameters
+----------
+
+Exclusion zone(s)
+| Spectral regions to exclude, water, solvent or contaminant resonances
+| If YES: parameters **Lower exclusion zone** and **Upper exclusion zone** are visible,
+| If NO: no zone to exclude
+| Default value is NO
+|
+
+Left exclusion zone
+| Upper boundary of exclusion zone
+|
+
+Right exclusion zone
+| Lower boundary of exclusion zone
+
+| *Notes:* 
+| - these parameters can be used several times using the "Add new exclusion zones" button
+|
+
+Shift
+| Maximum variation of the chemical shift (due to experimental conditions) of a peak of a fixed metabolite allowed in the warping function
+|
+
+Graphical zone(s)
+    | Spectral regions to display
+    | If YES: parameters **Lower graphical zone** and **Upper graphical zone** are visible,
+    | If NO: no zone to display
+    | Default value is NO
+    |
+
+	
+------------
+Output files
+------------
+
+proportionEstimation.tsv
+| tabular output
+| Array with p rows (corresponding to the identified metabolites) and 1 column containing the estimated relative intensities
+|
+
+spectra.pdf
+| pdf output
+| Graphical chart of real and estimated spectrum (whole spectral width and zooms depending on exclusion zone(s))
+| 
+
+
+---------------------------------------------------
+
+---------------
+Example
+---------------
+
+
+See the W4M00005_mmusculus in the Shared Data/Published Histories menu
+.. image:: ./static/images/NmrAnnotation_ASICS.png
+        :width: 100
+
+
+   </help>
+    
+</tool>
Binary file nmr_annotation/static/images/NmrAnnotation_ASICS.png has changed
Binary file nmr_annotation/test-data/ADG_007.zip has changed
Binary file nmr_annotation/test-data/MTBLS1_177_NMR_Annotation.pdf has changed
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/nmr_annotation/test-data/MTBLS1_177_NMR_Annotation_proportionEstimation.tabular	Tue Jan 30 05:37:10 2018 -0500
@@ -0,0 +1,40 @@
+Metabolite	Relative_Concentration
+Creatinine	0.0569174516124917
+Citrate	0.0285947900676533
+AceticAcid	0.00931139835139612
+L-Glycine	0.00700618802859799
+ThreonicAcid	0.00312228402191229
+Lactose	0.00274471352269867
+Malonate	0.0026239869463794
+Indoxylsulfate	0.00256983041226207
+L-Arabitol	0.00247809838365845
+L-GlutamicAcid	0.00219750986486448
+Threitol	0.00185008716865122
+L-Alanine	0.00176514469439213
+HippuricAcid	0.00173759535227614
+Betaine	0.00163016358478906
+MalicAcid	0.00154536802871288
+1,3-Diaminopropane	0.00150129592201832
+2-oxoglutarate	0.00142533208121273
+Dimethylamine	0.00131575320617744
+L-Aspartate	0.00127266582522283
+AscorbicAcid	0.00100873973025415
+L-Asparagine	0.00099187237086438
+GuanidinoaceticAcid	0.000925290560878349
+Ethanolamine	0.00089279079362688
+Glycerol	0.00088244833185825
+5,6-Dihydro-5-Methyluracil	0.000849548732878672
+Creatine	0.00079525125560122
+MethylmalonicAcid	0.000792506060661937
+Xylitol	0.000787992683508674
+Taurine	0.000787744798563012
+Succinate	0.000527784171596945
+4-HydroxyphenylAceticAcid	0.000522123163032416
+Methylguanidine	0.000486125204031847
+Myo-Inositol	0.0004425385208034
+beta-HydroxyisovalericAcid	0.000419265217514307
+Propionate	0.000392879831159886
+TMAO	0.000391632855219129
+L-Valine	0.000329117139315598
+Isobutyrate	0.000312582908447652
+CholineChloride	0.000306877971629503