8:50 am Chair’s Opening Remarks

Utilising Current and Novel Predictive Models for Enhanced Forecasting of Developability and Manufacturability

9:00 am Computational Tools to Aid in Developability and Early Risk Assessment of Biologics


  • Overview of the in silico tools developed to aid in assessing physical and chemical liabilities of biologics such as aggregation, oxidation, deamidation etc
  • Demonstrate the use of these tools both in Discovery and Development for developability assessment and in the identification of key critical quality attributes (CQAs) during manufacturing
  • Case studies will be discussed to showcase the use of computational tools to aid in mechanistic understanding of degradation as well as higher-order structural characterization

9:30 am Modeling Protein Properties using pH-dependent Conformational Sampling

  • John Gunn Senior Research Scientist, Chemical Computing Group


  • Sampling protein conformational and charge states using grand-canonical dynamics
  • Calculating ensemble-averaged properties and descriptors in a variable-pH ensemble
  • Developing models for biologics developability assessment and liability prediction

10:00 am Building a Computational Pipeline for Early Developability Assessment and Optimization


  • Overview of the developability funnel at Denali and impact of structural analysis on predicting aggregation propensity
  • Introducing an ML tool for experimentalists to quickly glean insight from their sequence-“activity” data and rationally design next round of variants with improved developability characteristics
  • Examples from integrating a design-predict-make-test cycle for antibody engineering and hit triage

10:30 am Morning Refreshments and Networking

Implementing the Capabilities of AI and Machine Learning into the Process of Engineering Complex Biologics

11:30 am Modifying Discovery Workflow to Utilize Machine Learning Algorithms for Better and Faster Optimization of Biologics


  •  Merck’s biologics developability discovery platform
  •  High throughput analytical characterization data management
  •  Example of analytical characterization data overview; looking for patterns and correlations in analytical characterization data
  • Case study 1: Hydrophobicity and developability profiles
  • Case study 2: Thermal stability and developability profiles

12:00 pm AI & Machine Learning for High Quality Targets and Disease Indications for Both Biologics and Small Molecules


  • Practical and strategic bioinformatics to enable drug discovery
  • Published examples of how to use AI and ML to predict clinical successes
  • How to find the first and additional disease indications for your assets leveraging the computational solutions

12:30 pm Mastermind Session


This is an interactive session designed to help broaden your understanding of the challenges with the field and potential opportunities. Engage with different industry stakeholders and develop a deeper appreciation for how the field can move forward.

1:00 pm Lunch and Networking

Computational Tools for Accelerating Antibody Discovery & Lead Selection

2:00 pm Computational Design of Single Domain Antibodies


  • Challenges in the computational design of single domain antibodies
  • Selection of scaffolds for single domain antibody design
  • Utilization of machine learning and molecular dynamic simulations for single domain antibody design

2:30 pm Antibody Dynamics from the Nanosecond to Millisecond Timescale – Implications for Affinity Maturation and Humanization


  • Linking loop flexibility and affinity maturation
  • Relative VH-VL interdomain dynamics
  •  Kinetics vs. thermodynamics

3:00 pm CDR-H3 Loop Dynamics – Implications for Antibody Design and Structure predictions


  •  Conformational ensembles of CDR-H3 in solution vs. crystal structures
  • Induced fit and Conformational selection as paradigms for binding
  •  Timescales of conformational transitions of loops

3:30 pm Chair’s Closing Remarks