Pre-Conference Workshop Day | October 24 2023

Pre-Conference Workshop Day

Tuesday 24th October 2023

Workshop A 10.00 – 12.30

In silico Strategies for Engineering Peptides and Proteins with Enhanced Properties

Synopsis

Functional enhancement of existing properties and introduction of new functionalities are important objectives in the development of therapeutic peptides, antibodies and industrial enzymes. These goals are commonly achieved through multiple time-consuming rounds of directed evolution. Incorporation of computational structure-based methods grounded in physics can accelerate this process by enabling reliable prediction of the location and identities of potentially beneficial mutations. The end result is a handful of productive variants prioritized for expression or a small enriched library that simplifies variant screening.

Learn From Schrödinger’s experts who will discuss strategies for:

  • Refinement and preparation of experimental structures to make them suitable for modeling, or generation of 3D models when experimental structures are not available
  • Computational alanine scanning to identify mutagenic hotspots
  • Exhaustive in silico mutagenesis to predict mutations that enhance stability, potency and selectivity
  • Prioritization and selection of variants for gene synthesis and expression

Workshop B 1.30 – 3.30

Utilizing Artificial Intelligence and Machine Learning to Design and Predict Biologic Candidates’ Safety

  • Andy Vo Principal Scientist - Computational Toxicology Research, Abbvie
  • Christian Grant Scientific Associate Director, Preclinical Safety, Abbvie

Synopsis

With the biologics space developing more candidates every year, the importance of understanding toxicity and addressing these safety aspects is ever-growing. Just as in biologics discovery, prediction tools can help reduce time and costs, and work to supercharge biologics developability.

Join this interactive workshop to learn how to:

  • Assess best-practices for incorporating integrated databases for predictive models for from discovery to development
  • Overcome data quantity & quality challenges to train models
  • Tackle common challenges for end-to-end integration of AI