Project layout

A project is a MATLAB hierarchical structure. Click on each field to discover the associated subfields.

Field format differences

Field formats may differ between what is required to initialize a project and how they are stored downstream in the project. Before building the pipeline, check the Project Initialization page for field requirements.

Project structure overview

project
├── models
│   └── <modelName>
│       ├── model                    (COBRA struct)
│       ├── sampleMetadata
│       ├── discretizedData
│       ├── expressionData
│       ├── mappedDiscretizedRxnsAllSamples
│       ├── mappedDiscretizedRxns
│       ├── coreReactions
│       ├── settings
│       │   ├── medium
│       │   │   ├── mediumComposition
│       │   │   └── manuallySetBoundaries
│       │   │       ├── closedImports
│       │   │       ├── closedExports
│       │   │       ├── unconstrainedImports
│       │   │       └── unconstrainedExports
│       │   ├── scriptParameters
│       │   │   ├── consensusProportion
│       │   │   └── sampleLabeling
│       │   ├── dico
│       │   ├── objFunction
│       │   ├── referenceModel
│       │   ├── mapping
│       │   └── optionalSettings
│       │       ├── medium
│       │       ├── notMediumConstrained
│       │       └── func
│       └── analysis
│           ├── <analysisId>
│           │   ├── parameters
│           │   ├── FBA
│           │   ├── FVA
│           │   │   ├── minMaxFluxes
│           │   │   └── loopStatus
│           │   ├── singleGeneDeletion
│           │   ├── doubleGeneDeletion
│           │   ├── sampling
│           │   │   ├── samples
│           │   │   └── cycleFreeFlux
│           │   │       ├── samplesLl
│           │   │       ├── thermoFeas
│           │   │       ├── sampleStatusAfterCorrection
│           │   │       ├── neededAttempts
│           │   │       └── looplessStatus
│           │   └── kld
│           └── active
└── comparisons
    └── <Name1_vs_Name2__date>
        ├── modelNames
        ├── referenceModel
        ├── analysisIds
        ├── structuralComparison
        │   ├── rxnMappingTable
        │   └── plots
        ├── functionalComparison
        │   └── plots
        └── samplingComparison
            ├── orderedSamples
            ├── sampleModelLabels
            ├── orderedFba
            └── plots

project (struct)

Field	Type	Status	Description
`models`	struct	Required	One or more models
`comparisons`	struct	Optional	Comparisons between models

project.models (struct)

Each field is a model name (<modelName>).

project.models.<modelName> (struct)

Field	Type	Status	Dimensions	Description
`model`	struct (COBRA)	Required	—	Metabolic model
`sampleMetadata`	table	Optional	nbSamples × N	Sample metadata
`discretizedData`	table	Optional	nbGenes × 2+	Discretized data (`geneIds`, `value`)
`expressionData`	table	Optional	nbGenes × 2+	Raw expression data (`geneIds`, `expression`)
`mappedDiscretizedRxnsAllSamples`	int8	Optional	nbRxns × N	Mapping across all samples
`mappedDiscretizedRxns`	int8	Optional	nbRxns × N	Consensus mapping
`coreReactions`	vector	Optional	—	Core reaction indices
`settings`	struct	Optional	—	Configuration parameters
`analysis`	struct	Optional	—	Analysis results

settings (struct)

Field	Type	Status	Description
`medium`	struct	Optional	Medium composition
`scriptParameters`	struct	Optional	Script parameters
`dico`	table	Conditional*	Gene ID mapping (data ↔ model)
`objFunction`	char	Optional	Objective reaction
`referenceModel`	char	Optional	Reference model
`mapping`	sparse double	Optional	Mapping matrix
`optionalSettings`	struct	Optional	Additional parameters

*Required if discretizedData or expressionData is present.

settings.dico (table)

Column	Status	Description
`geneIdsInModel`	Required	Model gene IDs (same order as `model.genes`)
`geneIdsInData`	Conditional	Data gene IDs (same order as `discretizedData.geneIds`)

settings.medium (struct)

Field	Type	Description
`mediumComposition`	table	Medium composition
`manuallySetBoundaries`	struct	Manually set boundaries

medium.manuallySetBoundaries (struct)

Field	Description
`closedImports`	Closed imports
`closedExports`	Closed exports
`unconstrainedImports`	Unconstrained imports
`unconstrainedExports`	Unconstrained exports

settings.scriptParameters (struct)

Field	Type	Description
`consensusProportion`	numeric	Consensus proportion (default: 0.9)
`sampleLabeling`	string	Column in `sampleMetadata` defining sample groups

settings.optionalSettings (struct)

Field	Description
`medium`	Optional medium
`notMediumConstrained`	Reactions not constrained by medium
`func`	Forced reaction(s)

analysis (struct)

One entry per analysis (analysis_<id>) plus an active entry.

analysis.<id> (struct)

Field	Type	Status	Description
`parameters`	table	Optional	Parameters (`Parameter`, `Analysis`, `Value`)
`FBA`	struct	Optional	Flux Balance Analysis
`FVA`	struct	Optional	Flux Variability Analysis
`singleGeneDeletion`	struct	Optional	Single gene deletion
`doubleGeneDeletion`	struct	Optional	Double gene deletion
`sampling`	struct	Optional	Flux sampling
`kld`	struct	Optional	Kullback-Leibler Divergence

analysisId field

The active entry contains an additional analysisId (char) field in each sub-struct.

FVA (struct)

Field	Type	Dimensions	Description
`minMaxFluxes`	table	nbRxns × 2	Columns `minFlux`, `maxFlux`
`loopStatus`	logical	nbRxns × 1	Loop status
`analysisId`	char	—	(active entry only)

singleGeneDeletion (struct)

Field	Description
`grRatio`	Growth ratio KO/WT
`grRateKO`	Growth rate after deletion
`grRateWT`	Wild-type growth rate
`hasEffect`	Whether deletion has an effect
`delRxns`	Deleted reactions
`fluxSolution`	Flux solution

doubleGeneDeletion (struct)

Field	Description
`grRatioDble`	Double deletion growth ratio
`grRatioKO`	Growth ratio after double deletion
`grRateWT`	Wild-type growth rate

sampling (struct)

Field	Type	Dimensions	Description
`modelSampling`	struct	—	Sampling model
`samples`	single	nbRxns × nSamp	Flux samples
`cycleFreeFlux`	struct	—	Cycle-free results

sampling.cycleFreeFlux (struct)

Field	Type	Dimensions	Description
`samplesLl`	single	nbRxns × nSamp	Loopless samples
`thermoFeas`	uint8	nbRxns × nSamp	Thermodynamic feasibility
`sampleStatusAfterCorrection`	uint8	nbRxns × nSamp	Status after correction
`neededAttempts`	uint8	nbRxns × 1	Number of attempts needed
`looplessStatus`	uint8	nbRxns × nSamp	Loopless status

kld (struct)

Field	Description
`samplingSets`	Compared sample sets
`kldMatrix`	KLD matrix
`pValueKld`	p-value
`fdr`	False Discovery Rate
`setLabels`	Set labels

project.comparisons (struct)

Each field is named Name1_vs_Name2__date.

comparisons.<Name1_vs_Name2__date> (struct)

Field	Type	Dimensions	Description
`modelNames`	string	nbModels × 1	Compared models
`referenceModel`	string	—	Reference model
`analysisIds`	table	nbAnalyses × nbModels	Compared analyses
`structuralComparison`	struct	—	Structural comparison
`functionalComparison`	struct	—	Functional comparison
`samplingComparison`	struct	—	Sampling comparison

structuralComparison (struct)

Field	Type	Dimensions	Description
`rxnMappingTable`	table	nbRxns × nbModels	Reaction mapping
`plots`	struct	—	Generated plots

structuralComparison.plots

Field	Description
`dataDiscretization`	Nested barplot showing the discretization status (−1, 0, 1) of genes and reactions across models and samples, at three stages: after discretization, after GPR mapping, and after consensus proportion
`coreReactions`	Stacked barplots showing core vs non-core reactions per model, percentage of core reactions included, and Venn/heatmap of core reaction presence across models
`coreReactionsIntersections`	Stacked bar chart of core reactions per pathway for each intersection/outersection between models (up to 4 models)
`intersections.genes`	Venn diagram or heatmap of gene presence across models
`intersections.mets`	Venn diagram or heatmap of metabolite presence across models
`intersections.rxns`	Venn diagram or heatmap of reaction presence across models
`jaccardDist.genes`	Jaccard similarity heatmap of gene presence between models
`jaccardDist.mets`	Jaccard similarity heatmap of metabolite presence between models
`jaccardDist.rxns`	Jaccard similarity heatmap of reaction presence between models
`reactionPathwayPresence`	Heatmap of reaction presence per subsystem, relative to the reference model

functionalComparison (struct)

Field	Description
`plots.objValue`	Bar plot of FBA objective function values per model
`plots.import`	FBA flux values for import exchange reactions (upper threshold)
`plots.export`	FBA flux values for export exchange reactions (lower threshold)
`plots.fvaSim.overall`	Clustergram heatmap of overall FVA boundary similarity between models
`plots.fvaSim.hist`	Histograms of per-reaction FVA similarity values for each model pair (lower triangle, values < 1)
`plots.fvaSim.enrich`	Dot plot of pathway enrichment for low FVA similarity reactions (dot size = \|NES\|, color = −log10(FDR))
`plots.fba.heatmapRxnFluxsum`	Heatmap of flux sum per reaction, grouped by pathway (from FBA)
`plots.fba.heatmapRxnActivityFba`	Heatmap of reaction network activity per pathway (from FBA)
`plots.fba.heatmapMetsFluxsum`	Heatmap of flux sum per metabolite, grouped by pathway (from FBA)

samplingComparison (struct)

Field	Type	Dimensions	Description
`orderedSamples`	double	nbRxns × cumSamp	Ordered sampling matrix across all models
`sampleModelLabels`	string	1 × cumSamp	Model label per sample
`orderedllSamples`	double	nbRxns × cumSamp	Ordered loopless sampling matrix (if available)
`orderedFba`	double	nbRxns × nbModels	Ordered FBA solutions
`kld`	struct	—	Inter-model KL divergence results (if KLD was performed)
`plots`	struct	—	Generated plots

samplingComparison.kld (struct)

Field	Type	Description
`orderedkldSets`	double	Ordered KLD sampling sets across models
`modelLabels`	string	Model labels for KLD sets
`interModelKld`	double	Inter-model KLD matrix
`interModelLabels`	string	Labels for inter-model KLD
`orderedkldSetLabels`	string	Set labels for ordered KLD

samplingComparison.plots

Field	Description
`heatmapRxnFluxSum`	Heatmap of flux sum per reaction, grouped by pathway (from sampling)
`heatmapMetsFluxSum`	Heatmap of flux sum per metabolite, grouped by pathway (from sampling)
`heatmapRxnFluxSumSamples`	Heatmap of flux sum per reaction per pathway, broken down by individual samples
`heatmapMetsFluxSumSamples`	Heatmap of flux sum per metabolite per pathway, broken down by individual samples