Speaker: Natalia Pinto Estéban.

Abstract: The intersection of artificial intelligence and protein engineering represents an innovative frontier in scientific exploration. In this presentation, titled ‘Large Language Models in Protein Engineering,’ we delve into the field of advanced language models, focusing on Meta’s ESM-1v and ESM-1b, as well as the intriguing baseline FLIP (Fitness Landscape Inference for Proteins). The replication of results from the FLIP paper sets the stage for the final degree project development.
Beyond established models, the discussion expands to introduce a revolutionary contribution: ESM2, another innovative creation from Meta. Experiments conducted with ESM2 are shared.
To sum up with this presentation and substain this research, it is provided diverse experiments using experimental protein data provided by the CSIC (Spanish National Research Council).
As we navigate the complexities of protein engineering, the integration of AI models emerges as a powerful tool. This presentation serves as a testament to reflect the broad development and research lines where language models proves to be of great value for the progress in computational biology.