Create a context flux table from differential expression gene (DEG) data and a metabolic network.

This task integrates gene expression data with metabolic network information to generate context-specific flux constraints for metabolic modeling. It evaluates gene-reaction rules in the network using fold change values from DEG analysis to determine appropriate flux bounds for reactions.

Input Requirements

DEG Table

• Must contain gene identifiers and log2 fold change values
• Gene IDs should match those used in the metabolic network
• Default expected columns: 'gene_id' and 'avg_log2FC' (configurable)

Network

• Metabolic network in JSON format
• Must contain gene-reaction rules linking genes to reactions
• Reactions without gene rules are not modified

Algorithm

1. Gene ID Validation: Checks compatibility between DEG gene IDs and network gene IDs
2. Boolean Expression Evaluation: For each reaction's gene-reaction rule:
   • Extracts gene IDs from the rule
   • Retrieves corresponding fold change values from DEG data
   • Evaluates boolean logic:
     • OR operations: takes maximum fold change value
     • AND operations: takes minimum fold change value
     • Handles complex nested expressions with parentheses
3. Flux Constraint Assignment: Based on evaluated fold change:
   • Values below -threshold: assigns "low" flux constraints (typically restrictive)
   • Values above +threshold: assigns "high" flux constraints (typically permissive)
   • Values within threshold: no constraints applied

Configuration Parameters

• gene_id_column: Column name containing gene identifiers in DEG table
• log2_fold_change_column: Column name containing fold change values
• threshold: Minimum absolute fold change to trigger constraint
• target_high/low: Target flux values for upregulated/downregulated genes. The target of high fluxes can be for example the mean of fluxes in control conditions.
• lower_bound_high/low: Lower bounds for flux constraints
• upper_bound_high/low: Upper bounds for flux constraints
• confidence: Confidence score for all constraints

Outputs

Context Flux Table

Contains flux constraints with columns:

• id: Reaction identifier
• target: Target flux value
• lower_bound: Lower flux bound
• upper_bound: Upper flux bound
• confidence_score: Constraint confidence

List of Reactions Modified

Summary table showing which reactions were modified and their flux traceability ("high" or "low")

Gene-Reaction Rule Examples

• Simple OR: "gene1 or gene2" → max(fold_change_gene1, fold_change_gene2)
• Simple AND: "gene1 and gene2" → min(fold_change_gene1, fold_change_gene2)
• Complex: "(gene1 and gene2) or gene3" → max(min(fc1, fc2), fc3)

Notes

• Reactions without matching genes in DEG data are not constrained
• Boolean expressions support standard logical operators with proper precedence