Ersilia is a nonprofit enterprise that aims to provide the community with a set of computational tools related to global health drug discovery.
Our goal is to strengthen the research capacity against infectious and neglected tropical diseases in low and middle income countries (LMICs). We do so by democratising access to machine learning (ML) models for drug discovery, facilitating the incorporation of data-driven research into experimental pipelines.
Computational methods hold the promise to revolutionise the drug discovery field, capitalising on the vast amount of experimental data accumulated over the years and the latest advances in machine learning.
Unfortunately, due to the lack of a unified framework for effective dissemination, ML models remain inaccessible to the broad scientific community, and their use is not yet integrated in day-to-day biomedical research.
We are creating Ersilia, a platform hosting hundreds of pre-trained models to help non-expert researchers identify drug candidates for orphan diseases, design molecules de novo, understand mechanisms of action or anticipate adverse side-effects.
The ultimate goal of Ersilia is to effectively lower the barrier to drug discovery, encouraging academic groups and enterprises to pursue the development of new medicines following the principles of Open Source Drug Discovery.
Our journey started in September 2020. We will launch a first version of the platform containing 10 models by the end of April 2021. A version with 100 models will be available by September 2021.
Our first task will be to curate peer-reviewed ML models existing in the literature and gather them in a common framework.
Next, we will train a battery of in-house models focused on diseases that are currently neglected by the pharmaceutical industry due to estimated low returns. Ersilia core technology for model building is the Chemical Checker, an existing platform that embeds most of the available drug biomedical data into a unified vectorial format, providing an excellent basis for efficient modelling. Finally, Ersilia will be released as an intuitive, user-friendly hub of ready-to-use Open Source Drug Discovery (OSDD) models, ultimately unlocking the full potential of ML to transform biomedicine and global health.
We are just getting started and, at the moment, our model deployment is only available via CLI. We keep all code open, so you can follow our progress! A web-based app will come soon.
Ersilia is distributed under an MIT license.