Selecting the right AI strategy is critical to delivering accurate, timely, and impactful insights across drug discovery. With rapidly advancing technologies in foundation models, generative AI, and lab-in-the-loop systems, establishing robust workflows and best practices is more important than ever.
This upcoming webinar series, Advances in AI-Driven Drug Discovery ONLINE, brings together leaders from pharma and biotech to share practical insights into how they are implementing AI across the pipeline, optimizing workflows, and accelerating the journey from concept to candidate.
Please note, by registering for one webinar in the series, you will automatically gain access to the subsequent webinars.
Sponsored by Twist Bioscience
Webinar 2: How is AI Helping Us Co-Develop the Drugs of Tomorrow?
Tuesday 21st April 2026 at 3pm BST | 4pm CET | 10am EST
AI is an integral partner across the drug development pipeline. From accelerating discovery and protein design to enabling conversational data access and improving safety through AI-driven toxicology, researchers are combining human expertise with AI to make faster, more informed decisions. This evolving, collaborative approach is helping to streamline workflows, reduce risk, and ultimately accelerate the journey from concept to candidate.
- How is applied AI reshaping each stage of the drug development pipeline, from discovery to safety assessment?
- What does it take to make complex biomedical data truly usable for AI-driven decision-making (e.g. conversational AI and interoperable systems)?
- Where can AI deliver the greatest impact on cost, speed, and risk reduction, particularly in high-failure areas like toxicity?
Talk 1: Applied AI in early Drug Discovery – streamlining workflows and accelerating science
- Integration of AI tools into early-stage drug discovery - general purpose and more specialised tools
- Applied AI in protein design for targeted drug discovery
Lukas Westlake, Senior Research Scientist, AstraZeneca
Talk 2: Conversational AI in Drug Development: Promise, Reality, and What It Takes
- Clinical biomarker and omics data ecosystems have matured to support program-level decision-making; enabling AI-driven, cross-study insight requires evolving these foundations toward greater standardization and interoperability.
- This talk will outline what it takes to enable conversational access to these data: defining success criteria, evolving data and metadata to be AI-ready, and designing scalable, agent-based analytical workflows.
Ben Decato, Associate Director, Computational Biology, Amgen
Talk 3: AI-Based Toxicology Assessment in Tissue
- Toxicity is a major driver of attrition and cost in drug discovery, making it a high‑leverage target for AI.
- I will present research on an AI system for histopathology anomaly detection: trained on abundant healthy tissue and frequent pathologies, while safely flagging rare or unseen toxicities via OOD detection.
- Long‑term vision: AI‑assisted pathologist to improve efficiency in exploratory preclinical tox studies.
Fabian Heinemann, Head of Central Data Science, Boehringer Ingelheim
Webinar 3: Discussion – Beyond the Pilot: Making AI Work at Scale in Drug Discovery
Tuesday 28th April 2026 at 3pm BST | 4pm CET | 10am EST
After years of pilots and proof-of-concept projects, pharma companies are now under pressure to scale embedding AI across R&D to deliver real, measurable impact. But moving from experimentation to full integration brings new challenges in infrastructure, governance, and cross-functional alignment.
This interactive panel will explore what it truly takes to operationalise AI across the drug discovery pipeline and where the biggest opportunities and roadblocks lie. We’ll tackle three critical questions:
- How can organisations successfully scale AI from isolated pilots to business-wide deployment?
- What foundations data, infrastructure, and governance are essential for sustainable implementation?
- Where is AI delivering the most tangible value across discovery, development, and decision-making today?
Panellists:
-
Richard Hughes, Vice President, Head of Computational & Structural Sciences, Bicycle Therapeutics
-
Andy Nuzzo, Associate Director, Computational Biology, GSK
-
Elif Ozkirimli, Head of Computational Science Products, Roche
-
Shivaprasad Patil, Associate Director, Bioinformatics & Predictive AI, AstraZeneca
Webinar 1: How Close Are We to an AI-Driven Revolution in Drug Target Discovery?
NOW AVAILABLE ON-DEMAND
Sponsored by Twist Bioscience
AI is transforming how we understand disease biology and identify new drug targets. From foundation models to generative AI predicting targets, discovery is being reimagined through lab-in-the-loop workflows, high-throughput validation, and scalable AI-driven experimentation. This webinar addresses:
- How is the “lab-in-the-loop” model reshaping drug discovery and experimental workflows?
- How can AI, automation, and high-throughput systems combine to dramatically increase speed and scale in R&D?
- What foundations are required to successfully translate AI: a deeper look into data, validation, and collaboration.
Talk 1: Enabling Lab-in-a-loop: Implementing MLOps in Roche Pharma Research
- Accelerating Drug Discovery through Automation: MLOps enables "Lab-in-a-loop" by creating automated pipelines that ingest lab data, retrain models, and suggest subsequent experiments. This process dramatically shortens research cycles and helps identify better, safer medicines faster and more cost-effectively.
- Ensuring Reliability and Scalability: The framework provides strict version control and traceability for data and code, ensuring that predictive models are reproducible and meet regulatory standards. It also manages the necessary computational infrastructure to scale as experimental data grows.
- Fostering Collaborative Innovation: Implementing MLOps requires strong collaboration between lab scientists and computational experts to co-define data collection processes and prediction interpretations
Le (Muller) Mu, MLOps Lead, Computational Sciences Center of Excellence, gRED, Roche
Talk 2: A High-Throughput Ecosystem to Power AI-Driven Antibody Discovery
- “Lab-in-the-Loop” Validation: Rapidly move from in silico design to wet-lab testing with high-throughput antibody production and characterization to validate AI outputs in weeks, not months.
- Explore High Diversity Sequence Space: Build a custom variant library or use one of Twist's fully human, validated antibody libraries to explore antibody sequence space
Oren Beske, Head of Biologics, Twist Bioscience
Talk 3: Functional Genomics in the Era of Agentic AI
- AI is a key to unlock greater exploitation of AZ’s unique set of rapidly growing unbiased multimodal data assets
- The Lab in the loop with predictive biology in control. AI tools can be applied to any of the steps required to build a functional genomics workflow but require identification of predictive biology to maximise return on investment
- I will discuss a few examples of how we are using AI applied to extracting greater value from functional genomics wet lab capabilities
Davide Gianni, Senior Director, AstraZeneca
