MetTracer

Introduction

MetTracer is an R package for tracing isotopic labeling experiments. Current source code still requires some in-house packages, which were not released yet. Hence, users are recommanded to use the docker image we build, which is build with all requirements and easy to use.

The docker image zhulab/mettracer-r contains entire envorienment for running MetTracer. Users can pull it and run MetTracer just as following.

What is MetTracer

MetTracer-r is an Docker environment to processing isotope labelled metabolomics data with MetTracer R package. It is based on the r-base docker.

Pulling image

Users can pull the MetTracer-r image with the following script

docker pull zhulab/mettracer-r

Data preparation

MetTracer requires the unlabelled metabolomics data to be pre-processed by xcms (see section 3.1.2 in http://metdna.zhulab.cn/metdna/help) and identified by MetDNA (see http://metdna.zhulab.cn) first. If users obtained metabolite identification results via other software, please modify your result table as required

The following files are required by MetTracer for labeled metabolites extraction:

• A folder named unlabelled containing unlabelled data files (.mzXML), annotation result (.csv) and peak information (.Rda) for targe isotopologue generation
• A folder named labelled containing data files (.mzXML) for isotope labelling data extraction
• A R script named run.R for your data processing

Preparing the labelled folder

The labelled folder contains all the data files requiring extraction of labeled metabolites. We permit multiple subfolders in it, and each subfolder is an independent biological group. The data files in the folder should be in mzXML format.

Preparing the unlabelled folder

The unlabeled folder contains unlabeled data files, annotation result and peak information.

• unlabeled data files
  • unlabeled data files are those used for peak detection and metabolite annotation. These data files should be in mzXML format and put into the subfolder.
• annotation information
  • annotation information is a .csv format table named MRN.annotation.result. The table should be put into the subfolder named MRN_annotation_result under subfolder MetProcess-Result
• peak information
  • peak information is a .Rda file which is generated after peak detection with XCMS.

Note:

if the users use other software tools for metabolite annotation, please adjust the format of annotation table. The annotation table should include columns: name, mz, rt, ID, compound.name, adduct and Formula. If the peak was annotated as several metabolites, separate them with `;`

Preparing the R script

File run.R is an R source code file paralleled with labelled folder and unlabelled folder. We provide a template here. Users only need to change the folder name in general. Other parameters are recommended parameters.

• Parameters
  • adj.label All subfolder names in labelled folder, the data files in these folder will do isotope contamination correction.
  • adj.unlabel A subfolder in labelled folder. The data files in this subfolder are unlabeled, and they will be used as reference files to correct the isotope contamination for data files in all labelled subfolder.

wd <- getwd()
setwd(wd)
library(MetTracer)

isotopologueParam <- IsotopologueParam(rt.extend=15, value="maxo")
experimentParam = ExperimentParam(wd = wd, nSlaves = 6)
extractParam <- ExtractParam(d.extract = "labelled",
                             adj.label = c("Glu_Gln_Ac","Glu_Gln","Gln_Ac","Unlabel"),
                             adj.unlabel = c(rep("Unlabel", 4)), 
                             adj.contaminate = TRUE)
iso.targets <- GenerateIsotopologue(isotopologueParam, experimentParam)
pdParam <- PeakdetectionParam(peakwidth = c(5,30))
iso.peaks <- ExtractIsotopicPeaks(extractParam, pdParam,
                                  iso.targets,check.peaks = TRUE)
ExtractIsotopologue(extractParam, iso.peaks, iso.targets, correct.iso = TRUE)

Overview of the data preparation

Run data processing work with mettracer-r image

• go to your data folder (e.g., data)

 cd data

• run docker using following code

# MUST keep the code exactly as it is!
docker run -it --rm -v "$PWD":/data -u $(id -u ${USER}):$(id -g ${USER}) zhulab/mettracer-r Rscript run.R

• wait till data processing work done

• Explaining docker run arguments

• -v "$PWD":/home/${USER}: mapping current dirctory as home directory in docker container

• -u $(id -u ${USER}):$(id -g ${USER}): using current user to run the container

• Rscript ~/run.R: run run.R in container home directory with Rscript command

updates

v1.0.1

• bugfix for adduct charge defination

v1.0.4

• update adducts info to match adduct types in MetDNA2
• several minor bugfix

License

This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)