MATLAB工具箱大全-基于约束的重构分析工具箱Cobratoolbox

# mgPipe ## Introduction mgPipe is a MATLAB based pipeline to integrate microbial abundances (coming from metagenomic data) with constraint-based modeling, creating individuals' personalized models. The pipeline is divided into 3 parts: [PART 1] Analysis of individuals' specific microbes abundances is computed. Individuals' metabolic diversity in relation to microbiota size and disease presence as well as Classical multidimensional scaling (PCoA) on individuals' reaction repertoire are examples. [PART 2]: 1 Constructing a global metabolic model (setup) containing all the microbes listed in the study. 2 Building individuals' specific models integrating abundance data retrieved from metagenomics. For each organism, reactions are coupled to their objective function. [PART 3] Simulations under different diet regimesand analysis of the predicted metabolic profile (PCoA of computed MNPCs of individuals). **WARNING:** Please take into consideration only the files listed in this document. Everything present in the folder but not listed and explained in this document is to be considered not relevant or obsolete. ## Requirements mgPipe requires `Matrix Laboratory`, the `Parallel Computing Toolbox`, as well as, the COBRA Toolbox to be installed. Please refer to the [installation instructions](https://opencobra.github.io/cobratoolbox/stable/installation.html). The usage of `ILOG CPLEX` solver is strongly advised to obtain the best speed performance (required for the fastFVA.m function). MgPipe was created (and tested) for AGORA 1.0 please first download AGORA version 1.0 from https://www.vmh.life/#downloadview and place the mat files into a dedicated folder. ## Main Folder Structure and Files The following files are essential for the usage of the pipeline and are supplied in the current folder and in `papers/2018_microbiomeModelingToolbox` | Filename | Purpose | | -----------------------------------------------|------------------------------------------------------------------------| | startMgPipe.m | *driver, containing all the input variables, to be modified by the user* | | initMgPipe.m | *function containing all the input variables launching the pipeline* | | loadUncModels.m | *function to load unconstrain and prune the models* | | fastSetupCreator.m | *function to create "global" setup* | | checkNomenConsist.m | *function to check that microbes have the right nomenclature* | | detectOutput.m | *function to check if a specific file was already created and saved* | | getIndividualSizeName.m | *get information on number and ID of organisms and individuals*| | addMicrobeCommunityBiomass.m | *function to add community biomass* | | mgPipe.m | *pipeline* | | parsave.m | *function to allow object saving in parallel loops* | | getMappingInfo.m | *function to extract information from the mapping* | | plotMappingInfo.m | *function plot extracted information from the mapping* | | createPersonalizedModel.m | *function to create personalized models* | | microbiotaModelSimulator.m | *function to simulate under different diets the created models (called from mgPipe)*| | makeDummyModel.m | *function to create a dummy model* | | mgSimResCollect.m | *function to collect and output simulation results* | | extractFullRes.m | *function to retrieve and export all the results (fluxes) computed during the simulations*| | README.md | *this file* | | useDiet.m | *function to impose a specific diet and add essential elements to microbiota models*| | adaptVMHDietToAGORA.m | *function to convert a specific diet from VMH into an AGORA compatible one*| | ***compfile*** | *Results subfolder: contains objects saved in open format* | ## Usage Once installed the necessary dependencies the pipeline is ready to be used at the condition that some input variables are inserted or changed from the default input file `startMgPipe.m` or directly in the input function `initMgPipe.m`. Running the script called `startMgPipe.m` (after having changed the necessary inputs) is the only action required from the user to start the pipeline. The pipeline can be stopped at every moment as all the results are saved as soon as they are computed. In case of accidental or volunteer halt in the execution, the pipeline can be simply restarted without loss of time running again `startMgPipe.m` : already saved results (from previous runs) are automatically detected and not recomputed. ## Inputs Some specific information files need to be loaded by the pipeline. For this reason, they must be formatted and called in a specific way. See the examples folder in `papers/2018_microbiomeModelingToolbox` for more information. The files needed are | File | Description | | -----------------------|----------------------------------------------------------------------------------------------| | normCoverage.csv | table with abundances for each species (normalized to 1, with a minimum value as a threshold to define presence)| | sampInfo.csv | optional: table of the same length of the number of individuals (0 means patient with disease 1 means healthy)| Some variables, in the input file, needs to be created/modified to specify inputs (for example paths of directories containing organisms models). The variables which need to be created or changed from default are | Variables | Description | | ----------------|--------------------------------------| | modPath | path to microbes models | | resPath | path to the directory containing results| | dietFilePath | path to and name of the file with dietary information| | abunFilePath | path to and name of the file with abundance information| | indInfoFilePath | path to csv file for stratification criteria (if empty or not existent no criteria is used)| | objre | name of the objective function of microbes| | figForm | the output is a vectorized picture ('-depsc'), change to '-dpng' for .png| | numWorkers | number of cores dedicated for parallelization| | autoFix | option to automatically solve possible issues (true means on) | | compMod | if outputs in open format should be produced for some sections | | rDiet | if to simulate also a rich diet (rdiet=true)| | extSolve | option to save microbiota models with diet to simulate with a different language (true means yes) | | fvaType | which FVA function to use (fastFVA =true for fastFVA) | The `autorun` variable controls the behavior of the pipeline. The autorun functionality is automatically off. This functionality enables the pipeline to automatically run and detect outputs. By changing `autorun` variable to false, it is possible to enter in manual / debug mode. **WARNING**: concerning the `autorun` v