Open Source Machine Learning for

|

Ersilia is a tech nonprofit organisation. We equip universities, hospitals, and laboratories in low-resourced countries with data science tools for infectious and neglected disease research.

We are on a mission to make science accessible to all

Data science opens an affordable path to scientific discoveries. All countries must benefit from it.

The research ecosystem is imbalanced

5%

Research output

Low income countries produce less than 5% of the world's scientific publications.

60%

Infectious diseases

Six out of the top 10 causes of death in these countries are due to infections.

10%

Antimicrobial drugs

However, only 10% of the drugs in development target infectious diseases (WHO, 2021).

Our approach

We use data science and machine learning to support researchers working in low-resourced settings. Together, we develop tools to speed up experiments and reduce the cost to develop new drugs.

Open science

We remove barriers to knowledge. Our code is free and open source, and we focus on making our tools accessible to all scientists.

In-country research

We put emphasis on the scientific leadership of local institutions, and support in-situ implementation of our tools.

Sustainable collaborations

We work together with local champions to ensure knowledge transfer and project continuity.

Our solution

We have created an open source platform to unlock the AI/ML expertise for all researchers. With our platform, wet-lab scientists and clinicians can browse a collection of models and run predictions, without having to write a single line of code or pay expensive software licenses.

Increase the visibility of existing methods

We collect and deploy AI/ML models currently scattered in scientific papers, code repositories and supplementary materials.

Exploit publicly available data

We search the scientific literature for relevant experimental data and use them to train our own AI/ML models.

Build AI/ML models in collaboration

We partner with researchers from the Global South and develop AI/ML assets adapted to their needs.

Prof. Kelly Chibale (H3D Centre) speaks about Ersilia.

Sustainable Development Goals

We are working towards the UN's Sustainable Development Goals 3, 4 and 10.

Building the Ersilia Model Hub

We are assembling the largest collection of ready-to-use AI/ML models for infectious and neglected disease research

Data driven

Ersilia technology achieves state-of-the-art performance thanks to the integration of chemical, genomic and biomedical text data.

Bioactivity signatures

We build upon bioactivity signatures, a compressed data representation technique that captures all sorts of information of small molecules, from molecular to the clinics.

Production-ready

We do all the necessary work to ensure our models are easy to adopt Models are served in a unified framework and ready for deployment.

Open source

All our assets are released under permissive open source linceses. We work collaboratively and adhere to the principles of open science.

Get involved

Join our efforts to make science more equitable.

And stay up to date with Ersilia's monthly newsletter! Sign Up.