Samenvatting

Wynand Alkema vertelt over het gebruik van ChatGPT in het ontwikkelen van nieuwe medicijnen. Een samenvatting van het praatje is hieronder weergegeven:The FAIR use of Large Language Models in the drug development pipeline.With the emergence of applications that are based on large language models (LLMs), such as ChatGPT and BARD, large amounts of unstructured scientific texts have become readily available for a wide public, including researchers in drug discovery and development. LLM based applications can now be used for a wide range of drug development tasks such as target finding and evaluation, lead optimisation and biomarker discovery. The stochastic nature of these models and the dependence of the result on correct prompt engineering poses a number of challenges when integrating the use LLMs into the drug discovery pipeline in a reproducible and interpretable way.In this talk we will give examples of how LLMs can be used for fast and highly efficient literature searches to discover novel drug-target combinations. We will also discuss the potential drawbacks of these technologies as well as the opportunities to link the results of LLMs to structured databases with chemical and biological data.