Bioinformatics and computational biology
GWAS and exome analysis, bulk and single-cell transcriptomics, proteomics, multi-omics integration, pathway and network analysis.
MatGen puts computational biologists, chemists and software engineers inside biotech and pharma teams — and builds the pipelines, models and tools those teams run on.
An overview of our capabilities across probabilistic modeling and AI/ML, computational biology and chemistry, and scientific software development, stage by stage from target biology to the clinic.
See the pipelineExample projects we have built for our clients, together with the internal platforms we have developed to support them: a molecular engine, a data integration app, a target explorer, and an assistant that reads a company's own data.
See what we builtSetting up a contract and getting our computational experts engaged with your projects is straightforward, and we guide you through every step. Here is an overview.
See the processWe turn complex biological and chemical data into results you can act on.
MatGen partners with biotech and pharmaceutical companies of every size, from a two-person startup with a single dataset to major pharmaceutical companies with established computational departments of their own. We are actively engaged at both ends of that range today.
The work spans custom analysis pipelines, statistical and machine learning modeling, computational biology and chemistry, and the software and data infrastructure that keeps all of it reproducible. Most engagements are one expert or a small team, working inside your meetings and your repositories.
Everything generated during a collaboration is yours. Analyses, models, and any code we write for you — MatGen keeps no rights over any of it.
Who we work with
Most projects draw on several of these at once. A target shortlist needs genomics, statistics and software; a selectivity assessment needs chemistry, machine learning and a well-maintained database behind it.
GWAS and exome analysis, bulk and single-cell transcriptomics, proteomics, multi-omics integration, pathway and network analysis.
Docking, molecular dynamics, structure prediction, ADME/T and side-effect modeling, generative design of small molecules, peptides and oligonucleotides.
Bayesian data integration, hierarchical models, survival analysis, experimental design, and honest treatment of uncertainty.
Property and structure prediction, representation learning, transfer learning on small datasets, and retrieval-augmented assistants over a company's own documents.
Analysis pipelines, bespoke databases, dashboards and internal tools, plus the infrastructure that keeps results reproducible a year later.
Connecting variants, expression, structure and pathways into a shortlist you can defend to a committee.
The three below are the arrangements we set up most often, and they are examples rather than a fixed menu — they can be combined, or adapted to whatever your project needs. The full process, including what you sign and how people are matched, is on the HowTo page.
An expert joins your team half or full time, for a minimum of three months and typically longer, operating as an extension of your staff.
A defined piece of work — a pipeline, an analysis, a tool — scoped against your objectives and delivered against them.
Senior experts provide strategic guidance and project leadership, while additional experts deliver execution-level support as needed.
An introductory call runs 30 to 45 minutes. By the end of it you will know whether this is something we can help with, and who from our team would be a good match for the work.