CONFERENCE DAY ONE | Wednesday, October 23, 2024
For full session details, download the event guide.
8:00 am Registration & Networking
8:55 am Chairs Opening Remarks
Computational Approaches for De Novo Biologics Design: Innovations & Applications
9:00 am Harnessing AI for De Novo Antibody Design
9:30 am AI-guided multiparameter antibody optimization enabled by lead-specific data
10:00 am AI-Driven Design of Novel Mini Proteins & Multispecific Compounds for Oncology
10:30 am Morning Break & Speed Networking
Synopsis
As the CDD for Biologics community is reunited, this valuable session will ensure you get the chance to reconnect with peers and make brand-new connections too! This structured networking opportunity will pair you with fellow attendees for several 3-minute introductions, ensuring you have the opportunity to meet and network with your academic and industry colleagues!
11:30 am From Enzymes to Antibodies – the Surprising Insights of ML-driven Optimization
Synopsis
In this presentation, we will discuss unexpected findings from our work on ML-guided, lab-in-the-loop protein design targeting affinity, expression, and stability for antibodies. Notably, we demonstrate that it is possible to achieve improved binding affinity without altering the complementarity-determining regions (CDRs). On the other hand, we show that even substantial modifications to CDRs, including the highly variable CDR3, can be applied while retaining strong binding properties, provided the changes are guided by a custom-conditioned protein language model (PLM). This opens an exciting path for active-learning-driven antibody optimization starting from an immunization campaign or even a single lead.
Cutting Edge Insights into Antibody Discovery to Unlock Therapeutic Potentials
12:00 pm Innovative Computational Strategies in Protein-Small Molecule Binding Design
Synopsis
- Advanced computational methods for protein-small molecule binding design;
- Challenges, methodologies, and successful outcomes of the sensor design project
- Exploration of enzyme engineering using small molecule-protein complex design pipeline
12:30 pm Exploring the Application of Generative AI to Antibody Discovery
1:00 pm Harnessing Advanced Imaging & Organoid Technology in AI-driven Drug Discovery
1:30 pm Lunch Break
2:30 pm Levitate Bio: Automated Rosetta & AI Software
Machine Learning Approaches for Enhanced Biologic Structure & Stability
3:00 pm The Role of Dynamics in Antibody Structure Prediction & Design: Current Implementations & Best Practices
Synopsis
- Examination of the functionalities and features offered by each tool, including molecular modelling, sequence analysis, and immunogenicity prediction
- Discussion on the strengths and limitations of different tools in addressing specific challenges in biologics discovery and development
- Application of molecular dynamics and structure prediction tools to design/engineer antibodies
3:00 pm Afternoon Break & Networking
3:30 pm Machine Learning Predictions of Biologic Stability in Early-Stage Antibody Discovery
4:30 pm Affinity-Aware In Silico Humanization Using a Domain-Specific Deep Learning Affinity Oracle
Synopsis
- Developed and validated an affinity oracle for antibody-antigen binding by integrating context-specific protein-protein interaction (PPI) data with ESM-2 and transfer learning techniques
- Leveraged the affinity oracle to assess mutations that revert antibodies to their germline state for the purpose of humanization
- Conducted experimental validation of model predictions through biophysical characterization