Cambridge Healthtech Instituteの第5回年次会議
Artificial Intelligence for Early Drug Discovery - Part 1
（早期創薬向け人工知能（AI） - パート1）
2023年4月11 - 12日
*Premium Pricing or separate registration required. See Short Courses page for details.
Registration and Morning Coffee7:00 am
Welcome Remarks8:00 am
AI FOR DRUG DESIGN
AI and Informatics Navigation of Huge Chemical Spaces
Computational drug discovery often makes use of screening methods to prioritize virtual libraries, selecting from amongst all available ligands only those most likely to have good efficacy and ADMET-related properties. In recent years, the size of these virtual libraries has greatly expanded by including not just already-available ligands but also ligands expected to be readily synthesizable from already-available building blocks. Within Janssen, we are expanding this space even further by identifying readily synthesizable building blocks as starting points for virtual libraries. This presentation will discuss early work in this area, recent successful implementations, and future-looking efforts.
3D Pride without 2D Prejudice: Bias-Controlled Generative Models for Structure-Based Design
Generative models for structure-based molecular design hold significant promise for drug discovery, however, data sparsity and bias are two main roadblocks to the development of 3D-aware models. Here we present a first-in-kind protocol for bias control and data efficiency, combining large 2D medicinal chemistry data resources with datasets of 3D ligand-protein complexes. We will illustrate how this framework is used to deliver augmented interactive drug design to modelers and medicinal chemists at Astex.
Applying AI to Precision Engineer Medicines
Exscientia's motivation to produce better drugs faster has led to the pioneering approach of patient-first, AI-driven drug discovery, design, and development. This presentation will demonstrate how automation and multiple data types are incorporated in every step of our end-to-end pipeline and will showcase our variety of proprietary toolkits and how they are integrated into a design and optimisation workflow.
Makya is a generative design platform that makes use of reinforcement learning to optimize virtual molecules. We have used Makya in virtual hit-finding efforts to optimize not only predicted activity but also predicted ADME properties. Makya can now be used to find optimal compounds from within ultra-large libraries and scaffold hop with chemistry-focused algorithms. Makya also designs novel compounds with a focus on synthesizability thanks to our retrosynthesis tool Spaya.
Networking Coffee Break10:10 am
FEATURED SESSION: SEPARATING HOPE AND HYPE
Can Humans Learn from Machine Learning in Drug Discovery?
Acquiring high-quality data is critical to the success of ML model deployment. Building such a knowledge graph based on genes, diseases and drugs informed an ML model which led to the identification of five novel genes associated with Alzheimer's disease. In addition to shortening the time from idea to cure, ML models can evaluate synthetic drug complexity with focus on sustainability. These lessons from ML model development will be discussed.
Repurposing and Machine Learning for COVID Drug Discovery
The small molecules approved for treating COVID are limited. Our machine learning efforts to repurpose molecules for Ebola led to novel molecules for COVID. We additionally explored developing machine learning models with quantum computing. We have progressed several molecules through in vitro and in vivo testing and identified several inhibitors of SARS-CoV-2. We also identified additional antiviral activities. The challenges and opportunities for this strategy will be presented.
End-to-End Drug Discovery Using AI and Robotics
Recent advances in AI technologies allowed for the development of end-to-end drug discovery platforms spanning multi-model target discovery, multiparameter molecular optimization, synthetic route planning, prediction of clinical trial outcomes, and many other steps. In this talk, we will show several case studies of end-to-end approaches and acceleration of target discovery using laboratory robotics.
Transition to Lunch12:05 pm
Luncheon Presentation (Sponsorship Opportunity Available) or Enjoy Lunch on Your Own12:15 pm
Session Break12:45 pm
AI FOR TARGETING RNA
Discovery of Novel Degraders of RNA-Binding Proteins by Integrating Molecular Dynamics with Fragment Screening
RNA-binding proteins (RBPs) are paramount effectors of gene expression, and their malfunction underlies the origin of many diseases. However, therapeutically targeting RBPs with small molecules has proven challenging due to highly polar orthosteric ribonucleic interactions and a lack of lipophilic cavities indicative of druggability. We present an integrated screening campaign integrating fragment-based differentiable design with molecular dynamics to discover a cryptic site enabling the discovery of a novel non-functional RBP binder. We demonstrate the physics-driven optimization of this hit molecule to induce the proximity of an E3 ligase, facilitating the proteosome-specific degradation of an RBP and restoring a tumor-suppressive miRNA.
Application of Artificial Intelligence to Discover RNA-Targeting Small Molecules
High-throughput screening of RNA-targeting small molecules remains a challenge due to the lack of RNA-focused small molecule libraries. We applied AI-assisted technology to select a suitable small molecule library and carried out compound screening using ALIS. The increased hit rate was significantly higher compared to our previous campaigns, which demonstrates that AI-driven compound selection strategies can accelerate RNA-targeted small molecule drug discovery.
Unlocking the Druggable Universe of 3D RNA Structures with Artificial Intelligence
Atomic AI has developed PARSE, the Platform for AI-driven RNA Structure Exploration, which can locate 3D structures at unprecedented speed and accuracy in disease-relevant RNA targets. PARSE builds on our work featured on the cover of Science, and involves a tight integration of high-throughput wet-lab experiments and cutting-edge artificial intelligence capabilities. Through this data-driven approach, Atomic AI is enabling and pursuing drug discovery against undruggable targets.
Refreshment Break in the Exhibit Hall with Poster Viewing3:20 pm
PLENARY KEYNOTE SESSION
Plenary Keynote Introduction (Sponsorship Opportunity Available)4:35 pm
Targeting Nodes and Edges in Protein Networks
Protein interaction networks consist of protein nodes and interaction edges. We aim to inhibit or stabilize specific protein-protein interactions to dissect these complex networks for chemical biology and therapeutics discovery. Through covalent fragment-based approaches, we discovered compounds that selectively stabilized the chaperone 14-3-3 bound to diverse client proteins and altered their function. Additionally, function-selective inhibitors for the multifunctional enzyme VCP/p97 are providing new tools and drug leads for cancer.
Welcome Reception in the Exhibit Hall with Poster Viewing5:30 pm
Close of Day6:30 pm
Registration Open7:00 am
In-Person Group Discussions with Continental Breakfast7:45 am
Challenges with AI Adoption and Implementation for Drug Discovery
- Trends in investing in and effectively using AI for preclinical drug discovery
- The challenge of continuous evolution of models in response to big data, data types, and computational platforms
- Understanding the caveats of AI-driven predictions for drug and target screening
- Machine learning models for predicting ADME-Tox properties of small molecules
AI FOR SCREENING & OPTIMIZATION
Giga-Screening for Preclinical Candidates with a Defined Multi-Target Profile
Rapid identification of a first-in-class or best-in-class preclinical drug candidate against a new target, new mutant, new functional state was always an important objective for computational drug discovery. However, even more challenging is screening for candidates with particular fine-grained specificity, target and property profile which may include multiple desired and undesired activities and properties. To add insult to injury, the screening of billions of compounds for candidates with those complex profiles present an extra barrier. Methods, models, implementations and applications will be discussed.
Chemistry-Aware Machine Learning on DNA-Encoded Libraries
DNA-encoded libraries (DELs) are used for rapid large-scale screening of small molecules against a protein target. At Anagenex, we employ machine learning techniques that incorporate nuances of chemical data to create improved deep learning these models. These models can be extended to screen external molecules and predict their binding affinity to a target of interest.
Coffee Break in the Exhibit Hall with Poster Awards Announced9:35 am
Poster Award (Sponsorship Opportunity Available)
AI Approach to Predict Inhibitors of Drug Metabolizing Enzymes and Transporters for the Design of Safer Drugs
Drug-drug interactions (DDI) are key for safety treatments. Drug metabolizing enzymes (DME) and drug transporters strongly influence the drug disposition and can be involved in DDI of a large number of drugs. We will present an AI approach integrating structural bioinformatics and machine learning methodologies to predict interactions of drugs with DME and drug transporters. We focus on the DME: Phase I (cytochrome P450), Phase II conjugate enzymes, sulfotransferase and UDP-glucuronosyltransferase, and on the BCRP transporter. Such AI approaches can improve the prediction of DDI in clinical practice and drug development pipelines.
Integration of Physics and Machine Learning for Challenging Drug Discovery Targets
Drug discovery is challenging for many reasons. We developed a flexible toolkit combining best-in-class capabilities in physics-based simulations and machine learning to enable our discovery projects. First, we describe the QUAISAR platform, which combines quantum mechanics, molecular dynamics, and machine learning to predict structure-activity relationships. We then show an example of how molecular simulations on massive GPU resources predict complex biological motions and reveal strategies for small molecule allosteric modulation.
In silico ADME: Application and Impact of QSAR Models in Drug Discovery
ADME properties are important considerations for drug discovery. Machine learning models were developed for predicting the ADME properties of small molecules. These models existed for more than a decade and were used on various occasions in lead optimizations. We will discuss the development of these models and the use cases in Genentech.
Close of Artificial Intelligence for Early Drug Discovery - Part 1 Conference12:00 pm