A platform built by a scientist, for scientists.

Predictive modelling in drug discovery has become both more capable and more fragmented. New algorithms arrive faster than teams can integrate them. Each step in a typical study, from descriptor calculation through interpretation through report writing, often lives in a different tool. The seams between those tools are where reproducibility quietly breaks.

MergenKit's mission is to make rigorous predictive modelling routine for teams who do not have the budget of a top-twenty pharmaceutical company. The platform unifies the pipeline into a single guided workflow, so a structure-activity question can be modelled, interpreted, and reported without leaving one environment. Interpretation is the default, not an optional extra. Reporting structures match documentation frameworks used in regulated assessments, so the scientific record is ready for the discussions that follow it.

The platform is built for researchers in drug discovery, including computational scientists and pharmaceutical R&D teams, and for academic groups doing rigorous QSAR or QSPR work. No programming background or in-house infrastructure is required to operate the system, so the workflow is accessible to medicinal chemists and pharmacology teams as well as computational specialists.

MergenKit is at the proof-of-concept stage, with ongoing methodological validation on community benchmark datasets. Customer conversations begin with the founder, focused on the modelling question and the data, before any commercial discussion. The platform's roadmap reflects what early users find useful in their actual work, not what looks attractive in a feature comparison. Development priorities are set against real studies brought by early collaborators, so each addition earns its place by answering a question a researcher actually has.

If you have a discovery question that would benefit from a guided modelling environment with integrated interpretation and reporting, the next step is a demo conversation. A short call covers the scientific question, the data you would model, and what success would look like for your team. Nothing about that first conversation depends on a procurement process, and the aim is to establish whether the platform fits the science before either side commits further effort.

The platform takes a deliberately narrow view of its own scope. It is a scientific modelling environment, not a regulatory authority and not a decision-making system. Predictions, interpretations, and reports are inputs to a researcher's judgement, not substitutes for it. This boundary matters in regulated industries, where the line between supporting evidence and binding decision must stay clear, and it matters in academic work, where the published record needs to reflect the chemistry rather than the tooling.

You can read more about the methodological principles shaping the platform, explore the three analytical modules in detail, or to explore a research partnership, see the collaboration page.