TriaX Therapeutics develops small molecules that work by holding two proteins together, rather than by blocking one. Our programs arrive already designed, ranked and reasoned about on the TriaX engine. Our job starts where computation stops: synthesis, binding, cells, and the data that decides whether a designed molecule is a real one.
A molecule that holds two proteins together does something a molecule that occupies a pocket cannot: it adds a function rather than subtracting an activity.
Most of the proteome has no pocket worth binding. Induced proximity gets round that by changing the question, and it turns a chemistry problem into a geometry problem, which is a problem that computation can actually attack.
A degrader is not consumed by the event it causes. One molecule can drive many turnovers, which changes the exposure a programme needs and the kind of target it can reach.
Engaging two non-redundant drivers with a single molecule means an escape route has to solve both problems at once, rather than routing around a single blocked node.
Our programs come out of an engine that enumerates the geometry rather than sampling it, so every candidate arrives with the reasoning and the provenance that produced its ranking.
TriaX AI owns the design platform and assigns each program outright to the company that develops it. TriaX Therapeutics owns composition of matter on its compounds and runs the biology, so the programme is not renting a capability it cannot control.
How the molecules are designed →
Described at programme level. Targets, chemistry and candidate structures are held back, and the stage markers say where each programme actually is rather than where it is headed.
A single molecule that drives degradation of two independent nodes a tumour cell depends on. Chosen for non-redundancy, so compensation has to happen twice.
The same design logic applied to a second, orthogonal pairing. A second programme that costs a fraction of the first is the difference between a platform and a one-off.
Stated plainly: both programmes have a complete designed landscape with positive controls passing under the same settings. What is in progress is the ranking at decision-grade resolution, which is what turns an enumerated landscape into a shortlist worth synthesising. Binding data is the gating milestone for both. We would rather show that boundary than imply a maturity neither programme has yet reached.
A designed molecule is a hypothesis. These are the steps that test it, in the order that kills a bad one soonest.
Designed candidates taken from structure to compound through CRO partners, with the analogue series that turns a single hit into a tractable series.
Biophysical confirmation that the intended complex forms, then degradation and functional readouts in cells. This is the gate every programme has to clear.
Primary-tumour and patient-derived models, so a compound meets real tumour heterogeneity before it meets a development decision rather than after.
Translational models and disciplined chemistry are what turn a ranked shortlist into a decision. We would rather build those relationships before we need them.
Patient-derived tumour organoid models, including primary-tumour and circulating-tumour-cell-derived lines, for cell-based evaluation of designed compounds.
Patient-derived models and compound-testing capacity, bringing programme candidates into clinically relevant material earlier than a conventional discovery timeline allows.
What has changed recently, and what it means for the programmes.
TriaX Therapeutics has entered a strategic partnership with Diagnostics Research Laboratory (DRL) covering patient-derived models and compound testing. The collaboration puts designed candidates in front of clinically relevant material at the point where binding data arrives, rather than waiting for a later translational stage.
TriaX Therapeutics has entered a strategic partnership with Cellentia for patient-derived tumour organoid models. Primary-tumour and circulating-tumour-cell-derived lines give the programmes a cell-based readout that reflects tumour heterogeneity rather than a single immortalised line.
Provisional patent applications covering the designed compounds have been filed, with a further provisional covering a four-body architecture and its pipeline drafted and filing shortly.
For partnering, licensing, CRO work or clinical collaboration. Tell us the programme stage and what you are looking to do, and we will come back with next steps.