Augmented Intelligence in Drug Discovery Xchange
East Coast, BOSTON
mAY 19, 2022
Welcome to hubXchange’s East Coast Augmented Intelligence (AI) in Drug Discovery Xchange 2022, bringing together executives from pharma and biotech to address and find solutions to the key issues faced in AI-led drug discovery.
Discussion topics will cover Data Quality, Target Identification, Lead Generation, Lead Optimization and Drug Response Prediction.
Take advantage of this unique highly interactive meeting format designed for maximum engagement, collaboration and networking with your peers.
Please note this is an In-Person meeting with a hybrid option to join virtually.
VENUE DETAILS: Hilton Boston Woburn Hotel, 2 Forbes Road, Woburn MA 01801
Data Quality
Opening Address & Keynote Presentation
X-Chem’s ArtemisAI Platform: A Flexible Approach to AI for Drug Discovery
- Intro to X-Chem
- ArtemisAI: X-Chem’s AI platform
- Overview of ArtemisAI case studies
Senior Vice President & General Manager, X-Chem
Noor is a serial biotech entrepreneur with a track record of achievements in AI having held an Assistant Professorship from Aalborg University.
Noor has published numerous papers in the field of artificial intelligence and is an inventor on a handful of patents. She is passionate about data and AI and on a mission to cure disease with the power of human and machine learning.
She is a recognised healthcare leader, MIT innovator under 35 and in BBC 100 women.
Vice President, Scientific Computing and Data Science, X-Chem
Marie-Aude has dedicated her career to the application of informatics in the field of drug discovery. At X-Chem, she has led a collaborative approach to generating high-quality data management solutions and a world-leading integrated suite of software tools to support X-Chem’s DNA-encoded library (DEL) platform. Having been part of X-Chem’s groundbreaking forays into machine learning on DEL, she now plays a key role in the continual expansion of X-Chem’s AI platform.
The challenges of inefficient data integration – what methods can be used to ensure the integration of more data?
- How to get good quality of data?
- Sources of raw data?
- Impact of too much or too little data?
- How to process the data for integration?
- Best approaches to design the data integration framework?
Senior Scientist, Novartis
Rohit is an initiative-taking, creative, and enthusiastic scientist/data scientist with profound understanding of medicinal/organic chemistry to invent innovative medicines and advance the frontiers of drug discovery. Rohit has over 18+ years experience in discovery and optimization of small molecule drug candidates in both biotech and pharmaceutical industry. Rohit has a strong track record of success as a member of various cross functional project teams, contributing to the discovery and identification of multiple pre-clinical compounds and is listed as a co-inventor on over 12 patents. Rohit is a life learner and passionate about improving and extending people’s life by solving complex medical challenges through drug discovery.
Applications of machine learning for antibody discovery and optimization
- How is machine learning useful in the:
– antibody discovery process?
– antibody discovery process?
– antibody developability assessment? - What does the future hold for machine learning in the antibody space?
Chief Scientific Officer, Twist Biopharma
Aaron is CSO of Twist Bioscience and leads our Twist Biopharma team. Prior to Twist, he served as Chief Scientific Officer of LakePharma, leading the California Antibody
Center, which discovers novel antibody therapeutics for its clients. He also oversaw all discovery research functions both as Vice President of Protein Sciences at Surrozen, and previously, as Vice President of Research at Sutro Biopharma, Inc. He also held director level positions at both Oncomed and Dyax Corp.
Networking Lunch
Data curation for property prediction in Drug Discovery
- Strategies for combining public and private data repositories of ADME properties
- Leveraging NLP & text mining for curating datasets from patents
- Best practice to follow for standardizing ADME property data
- Strategies for operating in low-data regime
Associate Director, Computational Chemistry, Nimbus Therapeutics
Leela Dodda is Associate Director, Computational Chemistry at Nimbus Therapeutics. At Nimbus, Leela is working on developing novel GNN models for predicting ADMET properties. In particular, he is interested in leveraging GNN’s ability to do multi-task learning and Transfer learning from Quantum Mechanical data for creating models that are not data-hungry. He got his Ph.D. in Computational Chemistry from Yale University, working on novel methodologies for structure-based drug discovery.
Before Nimbus, Leela was at VantAI, the machine learning-focused subsidiary of Roivant Sciences. At VantAI/Roivant, he helped set up the de novo degrader design engine and led the company’s degrader efforts in collaboration with Proteovant. Prior to this, Leela was part of the Computational Design Group at Silicon Therapeutics, supporting early discovery projects and developing the company’s De Novo Design workflows.
Poster Session: AI for drug discovery at X-Chem
- X-Chem’s AI services
- The application of ML to internal core activities
Research Scientist, Discovery Chemistry, XChem
Ryan Walsh is a research scientist on the discovery chemistry team at X-Chem. His background encompasses synthetic organic chemistry, cheminformatics, and machine learning, and he is passionate about leveraging the richness of DNA-encoded library (DEL) data to navigate therapeutically relevant
chemical space. He has designed, developed, and synthesized numerous libraries at X-Chem, and
coauthored a review article describing recent advances in the field of DEL reaction methodology
development. Ryan received his B.A. in chemistry and mathematics from the College of the Holy Cross, and his M.S. in chemistry from Northeastern University.
Target Identification
Opening Address & Keynote Presentation
X-Chem’s ArtemisAI Platform: A Flexible Approach to AI for Drug Discovery
- Intro to X-Chem
- ArtemisAI: X-Chem’s AI platform
- Overview of ArtemisAI case studies
Senior Vice President & General Manager, X-Chem
Noor is a serial biotech entrepreneur with a track record of achievements in AI having held an Assistant Professorship from Aalborg University.
Noor has published numerous papers in the field of artificial intelligence and is an inventor on a handful of patents. She is passionate about data and AI and on a mission to cure disease with the power of human and machine learning.
She is a recognised healthcare leader, MIT innovator under 35 and in BBC 100 women.
Vice President, Scientific Computing and Data Science, X-Chem
Marie-Aude has dedicated her career to the application of informatics in the field of drug discovery. At X-Chem, she has led a collaborative approach to generating high-quality data management solutions and a world-leading integrated suite of software tools to support X-Chem’s DNA-encoded library (DEL) platform. Having been part of X-Chem’s groundbreaking forays into machine learning on DEL, she now plays a key role in the continual expansion of X-Chem’s AI platform.
- Best use cases for mining literature
- Primary sources of data
Abstracts vs. Full text articles - Favorite commercial or publicly available tools for extracting data
- How to preserve extracted data—knowledge graphs vs. databases
Machine Learning & AI
Technical Lead, Pfizer
Peter Henstock is the Machine Learning & AI Lead at Pfizer and based in greater Boston. His work has focused on the intersection of AI, visualization, statistics and software engineering applied mostly to drug
discovery but more recently to clinical trials. Peter holds a PhD in Artificial Intelligence from Purdue University along with 6 Master’s degrees. He was recognized as being among the top 12 leaders in AI and Pharma globally by the Deep Knowledge Analytics group. He also currently teaches graduate AI,
and Software Engineering courses at Harvard.
- Siloed nature of OMICs data
- Limited clinical data availability
- Approaches for credentialing targets
Associate Vice President, Bioinformatics, Excelra
Chandra Sekhar Pedamallu comes with extensive experience in Bioinformatics and Computational Biology specializing in cancer genomics. Before joining Excelra, Chandra was associated with Sanofi Inc., identifying novel drug targets in Oncology (Molecular and Immuno-oncology). Prior to Sanofi, he was at Dr. Meyerson Lab, Dana-Farber Cancer Institute, and a visiting scientist at The Broad Institute of MIT and Harvard. During his 6-year tenure at Harvard, he led microbial analysis, pathogen discovery projects in cancer and other diseases. He has
contributed substantially to over a dozen projects under the auspices of the Cancer Genome Atlas (TCGA).
Chandra holds a Ph.D. in Systems Engineering (area of research : Global Optimization) from School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore. He has co-authored over 80
manuscripts, published in highly reputed journals including Nature, Nature Genetics, Science, Cell, PNAS, etc., and has been inducted as a Fellow of the Royal Society of Biology (FRSB).
3:45 – 4:15pm
Poster Session: Leveraging AI for target discovery and indication selection
- Integrating gene-disease annotations and biological networks from multiple sources
- Predicting novel associations from known genetic associations
- Contextualising results with public knowledge in the Euretos platform
Chief Scientific Officer, Euretos
Kees has a MSc degree in Physics and a PhD in the discipline of Physics, Mathematics and Computer Science from the Radboud University of Nijmegen on the topic of algorithms for statistical analysis of genetic data. As a postdoc at the University of Cambridge and the Wellcome Sanger Institute he developed algorithms and pipelines for large-scale sequence analysis and contributed to the identification of the genes responsible for a number of hematological disorders. After moving back to the Netherlands he became group leader in the departments of Human Genetics and Molecular Developmental Biology at Radboud University Medical Center. At Euretos he oversees the scientific activities and leads the development of the machine learning methods and data science capabilities powering the Euretos platform.
4:20 – 5:20pm
Machine learning for accelerating target identification: Costs and benefits
- The changing landscape of the drug discovery process and the role of target identification
- The landscape of opportunities for the application of machine learning methods for target identification
- Challenges towards modernizing the process of target identification
- The cost vs. benefit analysis of introducing this paradigm shift of using machine learning models for target identification
Director of Data Science, GlaxoSmithKline
Gurpreet Singh is an accomplished researcher in the use of machine learning to solve challenges in
healthcare. He has authored several publications in top-tier medical journals and is a proponent of the
development and implementation of machine learning-based assistants in daily patient care to improve efficiency and efficacy. He completed his doctoral thesis at the National University of Singapore’s
Department of Chemical and Biomolecular Engineering, where he focused on the use of machine
learning-based systems to improve early identification of Parkinson’s and Alzheimer’s disease utilizing imaging biomarkers. His team is now working at GSK to reimagine and modernize the drug discovery
pipeline.
Lead Generation
Opening Address & Keynote Presentation
X-Chem’s ArtemisAI Platform: A Flexible Approach to AI for Drug Discovery
- Intro to X-Chem
- ArtemisAI: X-Chem’s AI platform
- Overview of ArtemisAI case studies
Senior Vice President & General Manager, X-Chem
Noor is a serial biotech entrepreneur with a track record of achievements in AI having held an Assistant Professorship from Aalborg University.
Noor has published numerous papers in the field of artificial intelligence and is an inventor on a handful of patents. She is passionate about data and AI and on a mission to cure disease with the power of human and machine learning.
She is a recognised healthcare leader, MIT innovator under 35 and in BBC 100 women.
Vice President, Scientific Computing and Data Science, X-Chem
Marie-Aude has dedicated her career to the application of informatics in the field of drug discovery. At X-Chem, she has led a collaborative approach to generating high-quality data management solutions and a world-leading integrated suite of software tools to support X-Chem’s DNA-encoded library (DEL) platform. Having been part of X-Chem’s groundbreaking forays into machine learning on DEL, she now plays a key role in the continual expansion of X-Chem’s AI platform.
How can artificial intelligence accelerate drug discovery and development?
- Leverage all relevant data and technologies to understand your target and molecule.
- Leverage AI to navigate large chemical spaces and drug the challenge targets
- Leverage AI to rapidly generate testable hypothesis and reduce cycle times
- Challenges to leverage AI in projects
Director, Computational
Biology, Agios Pharmaceuticals
Huijun Wang is the Director of Computational Drug Design at Agios Pharmaceuticals. She is leading AI in Chemistry there and is very much interested in new technologies that could identify the right targets and design better molecules fast. Before joining Agios, Huijun had her career at Pfizer and Merck working on chemical biology, virtual screening, lead optimization, competitive intelligence, computational toxicology, etc.
High-throughput assays as tool to use AI in lead generation
- Solutions for high-throughput & automation in screenings for gene and protein biomarkers
- How do you set up lead generation in order to leverage AI in drug discovery and development?
- AI in RNA therapeutics development
Global Product Manager, Thermo Fisher Scientific
Stefan Jellbauer is a Global Product Manager at Thermo Fisher Scientific
supporting the high-throughput QuantiGene gene expression platform and Luminex
instruments. After 10 years in academic research, Stefan joined
Affymetrix/eBioscience, where he worked in the roles of Field Application Scientist and Technical Specialist with Affymetrix and Thermo Fisher Scientific. He continued his work in cancer diagnostic research with Farcast Biosciences as Technical liaison
supporting BioPharma business development for an ex-vivo human tumor platform. Stefan re-joined the QuantiGene team at Thermo Fisher Scientific as a Product Manager.
Stefan received his Ph.D. in Biology (Medical Microbiology/Immunology) from Ludwig Maximilian’s University of Munich (Germany), focusing on tumor vaccination. He completed his post-doctoral work at the University of California, Irvine studying mucosal
immunology and host-pathogen interactions.
Modeling drug physiological and in-vivo effects with generative AI
- What are the promises of generative AI comparing traditional modeling approaches?
- Most useful data types for modeling physiological drug effects with generative AI
- Learning disentangled representations of drug and disease mechanisms
- Necessity, advantages, challenges of learning interpretable representations
Senior Director, Data Science, Candel Therapeutics
Vladimir Morozov is Senior Director of Data Science in Candel Therapeutics where he designs a novel oncovirus platform. After receiving PhD in Chemical Enzymology from Moscow State University in 1997 he has been developing Data Science solutions in the biotech/pharma industry. His work has focused on OMICS data analysis and machine-learning
Lead Optimization
Opening Address & Keynote Presentation
X-Chem’s ArtemisAI Platform: A Flexible Approach to AI for Drug Discovery
- Intro to X-Chem
- ArtemisAI: X-Chem’s AI platform
- Overview of ArtemisAI case studies
Senior Vice President & General Manager, X-Chem
Noor is a serial biotech entrepreneur with a track record of achievements in AI having held an Assistant Professorship from Aalborg University.
Noor has published numerous papers in the field of artificial intelligence and is an inventor on a handful of patents. She is passionate about data and AI and on a mission to cure disease with the power of human and machine learning.
She is a recognised healthcare leader, MIT innovator under 35 and in BBC 100 women.
Vice President, Scientific Computing and Data Science, X-Chem
Marie-Aude has dedicated her career to the application of informatics in the field of drug discovery. At X-Chem, she has led a collaborative approach to generating high-quality data management solutions and a world-leading integrated suite of software tools to support X-Chem’s DNA-encoded library (DEL) platform. Having been part of X-Chem’s groundbreaking forays into machine learning on DEL, she now plays a key role in the continual expansion of X-Chem’s AI platform.
12:20 – 1:20pm
Network Lunch
1:25 – 2:25pm
- Challenges and opportunities in applying AI/ML in biologic engineering
- How can computational approaches help with antibody optimization?
- How will the successes of AlphaFold and other protein structure prediction approaches impact drug design and discovery?
Head, Bioinformatics and Data Sciences, Alexion Pharmaceuticals
Guillermo del Angel, PhD is Head of Bioinformatics and Data Science at Alexion, AstraZeneca Rare Disease in Boston, MA. Since joining Alexion in 2015 he has been leading research focused on applying different machine learning and data science approaches to improve rare disease target discovery, drug development and patient identification. He previously held positions at McKinsey & Co. in Mexico City and Boston, and as a computational biologist at the Broad Institute of Harvard and MIT in Cambridge, MA. A native of Mexico City, he originally trained as an electrical engineer and applied mathematician, earning a BS degree from ITESM in Mexico, a MS from Boston University and a Ph.D. from Cornell University.
2:30 – 3:00pm
3:30 – 3:45pm
4:20 – 5:20pm
Best practice in multiple parameter optimization by combining generative AI with physics/non-physics based approaches to optimize binding affinity, ADMET, etc that leads to a drug candidate
- The advantage and limitations of using AI to design molecules
- How can we do better by combining AI with other methods to prioritize ideas in lead
optimization, including binding affinity, ADMET, etc.? - What’s your best practice of using AI with other approaches in lead optimization?
Associate Principal Scientist, Computational Chemistry
Modeling & Informatics, Discovery, Alkermes
Yuan Hu is a computational chemist at Alkermes. He is modeling lead and chemistry lead for several small molecule and biologics projects. He has a wide experience on applying various physics-based and AI-based approaches in drug discovery to tackle challenging problems. He works very closely with medicinal chemists, biologists, DMPK experts and other members in the team. Prior to that, he was a postdoc at Merck and collaborated with Rutgers University, developed GPU TI in Amber software suites. He received his Ph.D. in Computational Chemistry from University of Delaware. Additionally, he has a M.S. degree in Organic Chemistry with hands-on experience in synthesis.
Drug Response Prediction
Opening Address & Keynote Presentation
X-Chem’s ArtemisAI Platform: A Flexible Approach to AI for Drug Discovery
- Intro to X-Chem
- ArtemisAI: X-Chem’s AI platform
- Overview of ArtemisAI case studies
Senior Vice President & General Manager, X-Chem
Noor is a serial biotech entrepreneur with a track record of achievements in AI having held an Assistant Professorship from Aalborg University.
Noor has published numerous papers in the field of artificial intelligence and is an inventor on a handful of patents. She is passionate about data and AI and on a mission to cure disease with the power of human and machine learning.
She is a recognised healthcare leader, MIT innovator under 35 and in BBC 100 women.
Vice President, Scientific Computing and Data Science, X-Chem
Marie-Aude has dedicated her career to the application of informatics in the field of drug discovery. At X-Chem, she has led a collaborative approach to generating high-quality data management solutions and a world-leading integrated suite of software tools to support X-Chem’s DNA-encoded library (DEL) platform. Having been part of X-Chem’s groundbreaking forays into machine learning on DEL, she now plays a key role in the continual expansion of X-Chem’s AI platform.
- Selecting and curating relevant biomarker data
- Defining the disease and intended patient populations
- Setting expectations for model performance and insights
- Managing growing algorithmic complexity
Chief Executive Officer, President and Board Member,
Lantern Pharma
Panna Sharma is the President, CEO, and Board Member of Lantern Pharma Inc., a clinical-stage
oncology biotech using artificial intelligence (AI) and genomics to innovate the rescue, revitalization, and development of precision cancer therapeutics. Lantern is focused on improving patient outcomes by
using its proprietary AI platform – Response Algorithm for Drug Repositioning & Rescue (RADR®) – to
rescue, revitalize and develop abandoned or failed cancer drugs, and to accelerate their development through precision trials that help identify patient groups more likely to respond to its pipeline of targeted cancer therapies.
1:25 – 2:25pm
Applications of AI to patient segmentation and drug response prediction
- Patient segmentation and drug response – from cox regression to deep learning – how do we leverage new methods to develop better predictors?
- xAI (explainability in AI) – is that the ultimate path to gain trust and adopt AI in drug discovery?
- Knowledge graphs and network analysis as the new normal to link together biological insights, prior knowledge and clinical outcomes
- High dimensional data in small clinical cohorts – how to make best use of omics and clinical data for drug response prediction
Director, Head of Data Science and AI, Early Oncology, AstraZeneca
Etai is an R&D director with more than 15 years of experience in developing and applying computational biology and ML/AI in drug discovery and early clinical development. His group at AZ specializes in ML/AI for ‘omics and clinical bioinformatics in translational medicine and discovery. Etai holds a PhD in Computational Biology from Bar-Ilan University and the Weizmann Institute of Science (in collaboration), MSc in Computational Biology and BA in Computer Science.
4:20 – 5:20pm
Enhancing drug response prediction in cancer patients
- What are some productive pairings of disease models, data modalities, and ML approaches for predicting response to combination therapies?
- Preclinical models: cancer cell lines, PDX, ex vivo tumor models, others
- Data modalities: bulk vs single-cell, SNP/CNV, chromatin accessibility, RNA expression, proteomics, perturbations, spatial, IHC, others
- ML approaches: integrating multiomics data, deep learning (limited N’s, dimension reduction, out-of-sample predictions), encoding biological networks, predicting response to single drug or drug combinations, others
Group Leader, Computational Biology, HiFiBio Therapeutics
Dean joined HiFiBiO Therapeutics in 2019. He has since expanded the company’s computational capabilities in using single-cell NGS data for target identification, target validation, patient stratification, and biomarker discovery in immuno-oncology and autoimmune disease drug development pipelines. His team developed HiFiBiO’s ML/deep learning approaches to integrate disparate data, automatically call cell types/states, and identify novel pathological cell populations. His team is expanding this capability to include more data modalities. He holds a Master’s in Neuroscience from Harvard University and a Master’s in Applied Statistics from Penn State.