DMARDy

DMARDy is an open‑source computational framework developed in 2020 by a consortium of pharmaceutical researchers. It is designed to support the research and development of disease‑modifying antirheumatic drugs (DMARDs) by providing integrated in silico tools for target identification, virtual screening, pharmacokinetic prediction and safety assessment. The platform combines a curated database of approved and investigational DMARDs with machine‑learning models that predict binding affinity, plasma protein binding, and drug–drug interaction potential. Users can submit chemical structures via a web interface or programmatic API, and the system returns a ranked list of candidate molecules along with predicted ADMET properties.

Key features of DMARDy include a modular architecture that allows addition of new predictive algorithms, a graphical workflow builder for custom pipeline creation, and interoperability with popular cheminformatics libraries such as RDKit and OpenMM. The database contains over 2,000 compounds, indexed by target pathway (for example, JAK, TNF, IL‑6) and clinical phase. DMARDy’s models have been benchmarked against published clinical outcomes, achieving an average AUC of 0.84 for efficacy predictions and 0.79 for adverse‑event risk scoring.

The project is distributed under the MIT license and actively maintained on GitHub. Contributors from academia and industry collaborate through pull requests and issue discussions. DMARDy is widely used in academic projects that aim to accelerate the discovery of next‑generation therapies for rheumatoid arthritis, systemic lupus erythematosus and other autoimmune diseases.