Network optimization, for biology.

An open-source Python framework for expressing and solving biological network optimization problems, from network inference to general-purpose graph-based modeling.

Read the documentation Read the paper
—GitHub stars Nat. Mach. Intell.2025 BSD-3-Clause open source

§ 01 Unified framework

For computational biology

Infer mechanistic networks from your omics data.

Move from gene expression, phosphoproteomics, or metabolomics to condition-specific signaling and metabolic networks, using biological knowledge.

  • Intracellular signaling from transcriptomics
  • Context-specific metabolic networks from omics
  • Multi-condition network inference
  • Compatible with Decoupler, Omnipath, COBRApy, and more.

For optimization & modeling

A composable API for optimization on graphs.

Build network optimization problems with modular formulations for flows, paths, trees, and acyclic networks, then solve them with the backend and solver of your choice.

  • Reusable constraint & objective building blocks
  • Backend-agnostic (CVXPY / PICOS; +20 solvers)
  • Supports directed, undirected and mixed (hyper)graphs
  • Graphviz and WebAssembly plotting support

§ 02 Applications

CARNIVAL signaling network

Signaling

Intracellular Signaling Networks

Causal reasoning over prior knowledge signaling networks from perturbation data.
Flux balance analysis map

Metabolism

Flux Balance Analysis

Genome-scale metabolic modeling with multi-sample joint formulations.
Network optimization feasible space

Graph optimization

Paths, Trees, DAGs and more

Build formulations for finding optimal network structures.

§ 03 Part of a broader ecosystem

NetworkCommons

CORNETO is part of NetworkCommons.

NetworkCommons helps you build context-specific biological networks by combining omics data, prior knowledge, and network inference methods in one platform. It makes it easier to access data, run different methods, and compare results across workflows. Many of its optimization-based methods are powered by CORNETO, the underlying engine for flexible network inference.

Paton V, Türei D, Ivanova O, et al. NetworkCommons: bridging data, knowledge, and methods to build and evaluate context-specific biological networks. Bioinformatics, 2025.

Explore NetworkCommons View on GitHub
NetworkCommons graphical abstract

§ 04 Peer-reviewed research

Nature Machine Intelligence  ·  2025

Unifying multi-sample network inference from prior knowledge and omics data with CORNETO.

Rodriguez-Mier, P., Garrido-Rodriguez, M., Gabor, A., & Saez-Rodriguez, J.

Read on nature.com →
BibTeX 
@article{rodriguez-mier2025,
  title   = {Unifying multi-sample network inference from prior knowledge
             and omics data with CORNETO},
  author  = {Rodriguez-Mier, Pablo and Garrido-Rodriguez, Martin
             and Gabor, Attila and Saez-Rodriguez, Julio},
  journal = {Nature Machine Intelligence},
  year    = {2025},
  doi     = {10.1038/s42256-025-01069-9},
  url     = {https://www.nature.com/articles/s42256-025-01069-9}
}