Back to All Events

D4- Data Driven Drug Development w/ Alix Lacoste

Meet Alix at D4 Data-driven Drug Development in New York to discuss machine learning for holistic drug target discovery.

The identification of therapeutic targets for a given disease is a key early bottleneck in the drug discovery pipeline. In addition to selecting potentially efficacious targets linked to disease mechanisms, a variety of other factors, such as druggability and novelty, drive a target’s success downstream.

At BenevolentAI, we take a holistic approach to target identification using machine learning and data science throughout the entire process. Relation extraction models on the scientific literature, combined with structured data from dozens of biomedical databases, form the basis of our knowledge graph. Next, relational inference algorithms, including matrix factorization and graph convolutional models are trained on the biological knowledge graph to predict novel targets for a wide variety of diseases and disease mechanisms.

These predictions are aggregated together with machine learning models built on transcriptomics and genomics data. We present ranked targets together with rich metadata mined from text and databases to our drug discovery scientists for an informed and nuanced evaluation. Promising targets from this pipeline are tested in the lab and results feedback to our inference models. This pipeline has led to early success in diverse disease areas such as ALS and glioblastoma.

Alix’s biography

Alix Lacoste BenevolentAI

Alix is VP Data Science at BenevolentAI. She has significant experience using data science and machine learning to advance biomedical discoveries. She holds a PhD in Molecular and Cellular Biology from Harvard University. Previously at IBM Watson Health, Alix led computational research projects in target identification and drug repurposing, most notably for Parkinson’s disease and amyotrophic lateral sclerosis, in collaboration with academic and pharma partners. At BenevolentAI, Alix connects AI and Drug Discovery groups to continuously improve the hypothesis generation pipeline.



AI in Healthcare: Towards a purposeful future

Read Alix’s latest blog where she explains how AI and machine learning represent great opportunities in healthcare and how improving best practices for the use of data in AI can lead to better and fairer outcomes.