COVRECON

COVRECON reconstructs genome- and metabolome-scale Jacobian structures and infers biochemical regulatory interactions from large-scale OMICS and metabolomics data using regression loss–based inverse Jacobian algorithms.

Key Features:

• Automated Network Reconstruction: Generates organism-specific enzyme and reaction datasets and reconstructs Jacobian structures using databases such as Bigg and KEGG tailored to targeted metabolomics datasets.
• Inverse Differential Jacobian Evaluation: Implements a regression loss–based inverse Jacobian algorithm that rates biochemical interactions by relevance from large-scale metabolomics data to enhance numerical stability and robustness.
• Improved Stability and Efficiency: Applies an algorithmic framework that reduces instability in regression problems and improves computational efficiency for large-scale metabolic networks.
• Data-Driven Superpathway Model Reconstruction: Automatically reconstructs data-driven superpathway models to explore generalized network structures beyond traditional pathways.

Scientific Applications:

• Inferring Biochemical Regulations: Identifies regulatory checkpoints and biochemical regulations within metabolic networks by analyzing large-scale OMICS and metabolomics data.
• Stochastic Analysis and Real-World Application: Enables in silico stochastic analysis on various-sized metabolic networks from the BioModels database and can be applied to real-world metabolomics examples.

Methodology:

Sim-Network: automates creation of organism-specific enzyme and reaction datasets and reconstructs Jacobian structure using databases such as Bigg and KEGG for targeted metabolomics analysis. Inverse Differential Jacobian Evaluation: applies a regression loss–based inverse Jacobian algorithm to rate biochemical interactions by relevance, improving numerical stability for large-scale network modeling.