Accelerating genome-scale model optimisation.

AI assisted tool designed to turn repository structure, docstring coverage, and code context into a publishable documentation site.

Software suite for quantitative prediction of cellular and multicellular biochemical networks with constraint-based modelling.

Collection of tutorials explaining how to use COBRA tools

An AI-assisted, fully transparent clinical decision support tool that delivers expert-level rare disease diagnosis suggestions from unstructured patient descriptions

Open-source interface for metabolic network reaction-by-reaction curation and validation

Automated GitHub & GitLab repository review using LLM evaluation against FAIR, reproducibility, security, and quality standards.

Secure feedback management backend for collecting, tracking, and managing user feedback through APIs and threaded conversations.

Data management platform that supports research in Inherited Metabolic Diseases (IMDs)

Privately hosted open source identity and access management solution

KinForm is a machine learning framework for predicting enzyme kinetic parameters using optimised protein representations that combine multiple protein language models, binding-site probability weighted pooling, PCA, and similarity-based oversampling. KinForm outperforms baseline methods, particularly for proteins with low sequence similarity to the training data, and shows that sequence-aware data splitting provides a more realistic measure of model generalisation.

Open-source interface for data validation and generation of kinetic parameter predictions of enzyme-catalysed reactions using state of the art methods

OpenKinetics Data provides downloadable CatLog-derived enzyme kinetic data enriched with UniProt protein sequences and PubChem substrate structures. It also provides associated protein representations, including ESM2, ESM Cambrian and ProtT5 residue embeddings, alongside Pseq2Sites binding-site probability scores, with metadata linking each artefact to its sequence.

OptArrow is an optimization integration engine designed to connect optimization clients and solver backends through a stable, high-performance transport layer.

Structural approach to reconciling VMH and Rhea metabolic reaction databases. Uses atom-atom mapping and conserved/reacting moiety identification (extended from COBRA Toolbox) to build resource-independent chemical fingerprints for reactions, aiming to compare signatures rather than labels to surface cross-resource matches naming conventions miss.

StereoMapper is a stereochemistry-aware pipeline that compares molecular structures to identify relationships such as equivalent molecules, enantiomers, and diastereomers etc. It generates explicit stereochemical mappings between metabolites to support improved cross-resource reconciliation and annotation.

A comprehensive database of curated human and gut microbial metabolic models linking metabolism with disease and nutrition

Privately hosted open source monitoring tool for tracking availability and website performance

AI-driven tool that drafts ethics application responses and performs structured reviews for consistency with study documents and regulatory standards.