Cambridge Healthtech Instituteの第5回年次会議
Artificial Intelligence for Early Drug Discovery - Part 2
（早期創薬向け人工知能（AI） - パート2）
2023年4月12 - 13日
Registration Open12:00 pm
Dessert Break in the Exhibit Hall with Poster Viewing12:45 pm
Welcome Remarks1:30 pm
AI FOR SCREENING & DEGRADATION
Differentiable Design: Dynamic Ternary Complex Structure Prediction with Multiscale Generative Diffusion Models
Designing bioactive molecules that serve a purpose - such as binding specifically to a protein - is central to medicinal chemistry and a common practice in drug discovery. Although automating design has tremendous promise, general-purpose methods do not yet exist. Here we explore a simple, fast, and robust approach to inverse design which combines learned forward simulators based on graph neural networks with gradient-based design optimization. Our approach solves high-dimensional problems with complex physical dynamics, including designing molecules that induce proximity between proteins forming a ternary complex.
In silico Screening for PROTAC Linkers
With more PROTACs being developed, the importance of linker design and optimization has been broadly recognized. Here we report an approach to virtually screen the linker designs by proactively using the available PROTAC ternary structures. Combined with expansive chemical building blocks, this approach helps us rapidly triage design ideas and expedite the discovery process.
Hits identified from recent AIDD campaigns are often lacking structural novelty, particularly synthesizability, with existing chemical spaces. In PharmaBlock, taking advantage of our 160k novel building block inventory and proved synthesis route collections, we established a megasize (1014) interactive virtual chemical space. Supported by our proprietary searching algorithm, we are able to generate structurally diverse hits, which are ready to be synthesized with our building blocks in stock.
Refreshment Break in the Exhibit Hall with Poster Viewing3:10 pm
Machine Learning-based Identification of Targets in Cancer for Protein Degradation
While Targeted Protein Degradation (TPD) can expand druggable targets, how protein degradation is dysregulated in cancer and how TPD drugs counteract this effect is incompletely understood. First, we developed a deep-learning model (deepDegron) to identify mutations that result in loss of protein degradation signals. Second, we developed a machine learning model (MAPD) to predict which protein targets are likely degradable by TPD compounds from unbiased proteomic experiments of the kinome.
Enhanced Active Learning by Combining Machine Learning and Structure-Based Methods
We will discuss an approach that applies AI and ML to design, test, and optimize lead molecules rapidly in silico and to suggest what compounds to synthesize and screen next in an ‘active learning’ process. A comprehensive platform approach offers tight integration between the virtual and real cycles (V+R). 3D modeling and simulation methods enhance the accuracy of predictions for drug potency, efficacy, and selectivity, while also addressing multi-target effects.
In-Person Group Discussions5:00 pm
Emerging Applications of AI/ML Tools in Drug Discovery
- Designing molecules that induce proximity between proteins
- Integrating fragment-based screening with molecular dynamics
- Virtually screening linker designs using available PROTAC ternary structures
- Machine learning to predict which protein targets are likely degradable
- AI and ML models to design, test, and optimize lead molecules rapidly in silico
Close of Day5:45 pm
Registration Open7:15 am
Diversity in Chemistry Breakfast Discussion (Sponsorship Opportunity Available)7:45 am
Diversity in Chemistry beyond Molecules: Gender and More
We encourage all to attend this moderated, audience-interactive discussion session. When it comes to increasing diversity among scientists, there continues to be a drop-off as one moves higher in leadership. Where do systemic challenges remain, what is your experience, and how can we continue to equalize the system?
Topics may include below, but will be guided by audience input:
- Where does the 'drop-off' of women in the chemistry career progression pipeline occur and why?
- How did the pandemic and other sea changes in the past three years bring us closer to or further from equality?
- What issues arose that you thought were solved?
- Diversity in life paths should include us all - how are men and nonbinary scientists being included?
- Intersectionality and equality - what is the experience of women of color, first-generation women scientists, and others?
PLENARY KEYNOTE SESSION
Plenary Keynote Introduction (Sponsorship Opportunity Available)8:35 am
Reflections on a Career as a Medicinal Chemist in Drug Discovery
A successful drug candidate depends on many factors: creativity of scientists involved, effective collaboration and commitment by the team, and the quality of the compound advanced. I reflect on a 40-year career pursuing the discovery of drug candidates designed to address unmet medical need in the cardiovascular, CNS, and viral diseases therapeutic areas and share undervalued strategies and other synthetic chemistry approaches for overcoming specific medicinal chemistry challenges.
Coffee Break in the Exhibit Hall with Poster Viewing9:30 am
EMERGING AI APPLICATIONS
Improving Machine Learning Predictions: Focus Is All You Need
The mainstream view in improving model predictions is identifying better modeling algorithms and assembling larger training sets curated to get more reliable data. In addition to these approaches we use other measures that are very effective: (1) eliminating data points that confuse the model algorithm, (2) biasing property distributions in training set with the goal to maximize performance metrics. Examples will be given with demonstrated material improvements.
Fingerprinting Drug Effects and Disease Phenotypes of the CNS Using Deep Functional Readouts of Human iPSC-Derived Neurons
We built a map of the electrophysiology of human excitatory neurons, with applications to high-throughput screening, drug repurposing, and target deconvolution. First, we measured hundreds of parameters from millions of iPSC-derived neurons perturbed by tool compounds, yielding a database of activity profiles. Then, we used representation learning to distill these data down to concise fingerprints, where similar compounds form neighborhoods and several important tasks (e.g. drug repurposing, target deconvolution) reduce to path-matching problems.
Medicinal chemistry synthesis for drug discovery is an expensive and laborious aspect of drug development, often requiring years of effort by skilled chemists. XtalPi is developing a custom robotics and human chemist platform. The goal-deliver productivity and reduce downtime in the development cycle. Real-world complex molecule applications will be presented to demonstrate the progress toward this goal.
Protein Flexibility and Binding Affinity: What Can AI Predict?
Accurately predicted binding affinities can be a powerful tool for virtual screening of ligands. Protein flexibility plays an important role in ligand binding. Hence, using parameters that determine protein flexibility will lead to better prediction of binding affinities in ML algorithms. This work will evaluate the use of such independent variables to determine binding affinities.
ASPIRE: Lowering the Barrier to Drug Development by Applying Automation, Data Analytics, and AI/Machine Learning to Chemistry and Biology
The gap between drug discovery and information science continues to close, hence there has never been a better time to leverage the power of AI/ML techniques to advance our understanding of the relationships between chemical and biological space. NCATS has identified, through the input of the greater scientific community, focus areas that need to be addressed in order to transform the design-synthesize-test cycle to transition to be more data-driven. The ASPIRE Program was created to support the development of AI/ML tools to process captured data to inform the next iteration of the process.
Transition to Lunch12:40 pm
Luncheon Presentation (Sponsorship Opportunity Available) or Enjoy Lunch on Your Own12:50 pm
Refreshment Break in the Exhibit Hall with Poster Awards Announced1:20 pm
Poster Award (Sponsorship Opportunity Available)
AI FOR HIT IDENTIFICATION
Manifold Embedding of Molecular Surface: A New Dimension in Chemical Deep Learning
To effectively differentiate huge quantities of molecules in data-driven drug research, we have developed a low-dimensional representation of molecules and utilize it in deep learning for predicting molecular properties and designing new drugs. The concept centers on transforming a molecule's 3D electronic attributes of local hardness and softness to lower-dimensional manifold embeddings. The representation carries the inherent information of intermolecular interaction strength and specificity. Both predictive and generative deep learning models have been developed out of the concept of data-driven drug research.
The CACHE Computational Hit-Finding Competition: Lessons Learned So Far
CACHE is a benchmarking exercise modeled after CASP where every four months, computational chemistry and AI experts predict up to 100 compounds for a predefined protein target. Hit candidates are then procured, tested experimentally at CACHE, and all data and a generic description of the methods are released publicly. The emerging landscape of the most successful computational hit-finding approaches so far will be outlined.
A Unified System for Molecular Property Predictions
There is no unified API for molecular property predictors (MPPs), which makes it difficult to share, distribute, version, retrain and manage predictors. We present Oloren ChemEngine (OCE), an open-source Python library with a unified, reproducible, and easy-to-integrate API for MPPs. Using OCE, we create models with the best leaderboard performances on 19 ADME/Tox benchmarks with MPP ensembling strategies. Using such API, we integrate model-agnostic uncertainty quantification and interpretability methods.
Networking Refreshment Break3:35 pm
AI FOR PROTEIN THERAPEUTICS
FEATURED PRESENTATION: Protein Design Using Deep Learning
Proteins mediate the critical processes of life and beautifully solve the challenges faced during the evolution of modern organisms. Our goal is to design a new generation of proteins that address current-day problems not faced during evolution. In this talk, I will describe recent advances in protein design using both traditional physics-based approaches as well as deep learning methods to design sequences predicted to fold into desired structures.
Designing Therapeutic Antibodies with Synthetic Biology and Machine Learning
BigHat Biosciences is designing safer, more effective antibody therapies for patients using machine learning and synthetic biology. Machine learning guides the search for better molecules by directing and learning from each cycle of our high-speed, automated wet lab that synthesizes and characterizes hundreds of antibodies each week. We’ll highlight key features of our platform and share several case studies of protein engineering using this novel platform.
Peptide Discovery and Optimization Using Artificial Intelligence Approaches
Successful peptide drug discovery programs today require attainment of multiple performance metrics to progress a compound to clinical stage. To aid decision-making, Koliber has developed an AI peptide platform based on state-of-the-art machine learning methods to analyze peptide properties, profile positions, and predict new variants. The capabilities and wet-lab validation of the AI platform will be demonstrated with examples from immunology and antimicrobial peptide discovery and optimization.
Close of Conference5:25 pm