TissueForge: a million pounds of laboratory, or seventy-two hours of compute
Complex wound research is one of the slowest and most expensive corners of biomedical science. The arithmetic behind that sentence is worse than most people assume.
A single laboratory experiment on a bioprinted multi-tissue construct, run over seventy-two hours, costs somewhere between eight and twenty-five thousand pounds once consumables, facility time and skilled labour are counted. Fifty parametric variations — roughly the minimum needed for statistically meaningful data — comes to over a million pounds and somewhere between eighteen and thirty-six months of work.
That arithmetic is why wound research moves slowly, and it is getting harder rather than easier. Regulatory pressure to reduce animal use in preclinical research is increasing across the UK, the EU and the US, and there is no validated computational alternative operating at scale. Meanwhile the three pathogens that dominate complex wound infection are developing resistance faster than conventional drug screening pipelines can respond.
TissueForge is a computational platform that simulates how a complex wound — skin, muscle and bone together — progresses under multi-drug-resistant infection across a timeline of up to seven days. That window matters more than any other detail: it is the period in which most therapeutic interventions either work or fail, and it is missing from existing computational wound models, which capture tissue mechanics in isolation and stop long before the biology gets interesting.
The intended use is screening. Test wound therapies, antibiotic combinations and biomaterials computationally before committing laboratory resources to them — turning a screening campaign that would take a year and a half of bench work into something that runs over a long weekend.
It also has a capability we find genuinely unusual: generating wound profiles that do not exist in any clinical dataset. That allows a therapy to be evaluated against injury patterns before they appear in patients, which is a strange sentence to write and a useful thing to have.
Now the honest part. TissueForge is at technology readiness level three: the concept is formulated and the components are being built, not validated. It is also not a pure simulation — it depends on a small in-vitro calibration layer to keep its outputs anchored to biological reality, because a model that has never been checked against a living system is a very confident guess. The design deliberately minimises the biological resources it consumes, but it does not eliminate them, and claiming otherwise would be the kind of overstatement this field already has too much of.