Keynote Speaker: Nicola Richmond, Ph.D.

Nicola Richmond, Ph.D.
Vice President of AI
BenevolentAI

Richmond is the vice president of AI at BenevolentAI, where she holds responsibility for the company's AI strategy, ensuring its continued leadership in the AI-enabled drug discovery industry. Fundamentally, Richmond is driven by the application of AI/ML technology to address challenges in drug discovery to positively impact patients' lives.

Nicola has a Ph.D. in pure mathematics and has worked at the intersection of AI and drug discovery for 22 years. During her post-doc, she developed the algorithm at the heart of a commercial product (GALAHAD), which has been used throughout the pharmaceutical industry to elucidate 3D pharmacophores for virtual screening. In 2004, Richmond joined GlaxoSmithKline (GSK), where she made several key contributions. Her statistical approaches for actioning high-throughput screening data are in continual use across GSK’s early drug discovery portfolio, and her work on predicting the stable expression of therapeutic antibodies has reduced manufacturing cell line development timelines by 85%. Richmond also built and led the GSK.ai Fellowship Programme that continues to educate and inspire the next generation of brilliant minds who want to apply AI to drug discovery and impressively achieved a 50:50 ratio of women to men.

Presenting

Knowing Why: Target Prediction with Explainable Large Language Models

Understanding the "why" in science often holds equal or greater significance than comprehending the "what." For bench scientists, this insight is invaluable. A medicinal chemist may want to understand which characteristics of a molecule are driving target affinity so they can incorporate that thinking into subsequent design cycles. Similarly, a protein expression scientist may want to understand why some engineered cell lines are better than others at stably expressing recombinant protein, again so they can incorporate that thinking into their engineering process.

At BenevolentAI, the company believes in augmenting their scientists with AI-enabled tools that allow them to make informed, data-driven decisions in a timely fashion. Like all scientists, we share the innate curiosity to delve into the "why" behind our results. The large language model is within our toolbox – a complex and notoriously black box. Unraveling its predictions poses a formidable challenge.

In this presentation, Richmond will guide you through BenevlonetAI's journey, showcasing how they've harnessed the power of large language models to identify potential therapeutic targets for a specific disease, and how they've ensured the scientist knows why.