Building confidence early through intelligent lead optimization
The Discovery UGM at Certainty will focus on how modern cheminformatics is evolving to support confident decision-making at the very start of drug discovery. Focusing on intelligent lead optimization, the sessions will highlight how the combined capabilities of Chemaxon cheminformatics solutions and D360 enable scientists to evaluate, prioritize, and optimize compounds as new drug modalities reshape the discovery paradigm. The track will explore what it takes for informatics solutions to keep pace with increasing molecular complexity and data diversity – while helping teams build confidence in compound quality from day one.
Designed for discovery scientists and R&D informatics professionals, the Discovery UGM will appeal to attendees seeking to understand the latest advancements in cheminformatics and how they can be applied in real-world discovery workflows.
Through thought leadership presentations and hands-on tutorials, this track will provide:
- Insight into how emerging drug modalities are changing discovery strategies and informatics needs
- Practical examples of how Chemaxon and D360 can be used to support intelligent lead optimization
- A clearer understanding of how modern cheminformatics tools help de-risk early discovery and accelerate confident decision-making
Tuesday, April 14, 2026
Agenda
7:15 – 8:15
Breakfast & badge pick up
8:15 – 8:25
Sheila Rocchio
Welcome to Certainty 2026
Kick off Certainty 2026 with a look at the state of the industry, an overview of what to expect during the event and the key insights to take back with you.
Speaker: Sheila Rocchio, CMO, Certara
8:25 – 8:50
Amin Rostami, Piet van der Graaf, Mirjam Trame
Evolution of evidence generation with PBPK and QSP champions
Join the leaders spearheading PBPK and QSP science & strategy for an insightful discussion that will explore:
- Critical regulatory shifts and their impact on MIDD
- The evolution of MIDD to MIE (Model-Informed Evidence) and how it’s being used
- Recentering the patient experience by leveraging findings from virtual trials
- How AI can be most effective and where it won’t permeate scientific processes
Speakers:
Amin Rostami, PhD, FCP, FAAPS, FJSSX, FBPS, Senior Vice President of R&D and Chief Scientific Officer (CSO), Certara
Piet van der Graaf, PharmD, PhD, Senior Vice President, Quantitative Systems Pharmacology, Certara
8:50 – 9:15
Matt Rizk, Leif Pedersen
Fireside chat: Moving from models to decisions earlier
This insightful conversation will showcase Merck’s approach to integrating Certara solutions & scientific expertise to improve decision-making at every phase of drug development. The discussion will also examine how AI is solving fragmented workflows and how this leads to exponential growth in productivity.
Speakers:
Matt Rizk, PhD, Associate Vice President, Translational Sciences & Outsourcing Department of Pharmacokinetics, Dynamics, Metabolism, and Bioanalytics, Merck
Leif Pedersen, President, Chief Commercial Officer, Certara
9:15 – 9:35
Josh Apgar
Achieving faster time to insights with AI
It’s clear that AI used in tandem with sophisticated modeling is shifting the drug development timeline, from generating insights to making confident decisions at a rapid pace.
This presentation will demonstrate how:
- Real power comes from combining AI regulatory-grade modeling software AND scientists: not replacing everything that exists
- Certara is incorporating AI to support decision-making as a core tech provider, and by empowering scientists-in-the-loop
- The future is more connected as AI architectures bring data, models, and expertise together for achieve quality decisions faster.
Speaker:
Joshua Apgar, PhD, VP, Head of QSP Software, Certara
9:35 – 9:55
Amin Rostami, Piet van der Graaf, Matt Rizk, Chris Bouton,Leif Pedersen
Panel Discussion: From vision to action
This panel will examine the challenges being faced by our industry today, and how to operationalize innovative technology & science for faster, better decision-making. Panelists will share where they expect strategy and processes to evolve in the next year.
Panelists:
Amin Rostami, PhD, FCP, FAAPS, FJSSX, FBPS, Senior Vice President of R&D and Chief Scientific Officer (CSO), Certara
Piet van der Graaf, PharmD, PhD, Senior Vice President, Quantitative Systems Pharmacology, Certara
Matt Rizk, PhD, Associate Vice President, Translational Sciences & Outsourcing Department of Pharmacokinetics, Dynamics, Metabolism, and Bioanalytics, Merck
Chris Bouton, PhD, Chief Technology Officer, Certara
Leif Pedersen, President, Chief Commercial Officer, Certara
10:00 – 10:30
Morning break - demo bar
10:30 – 11:00
Pat Walters
OpenADMET: Raising the bar on ADMET prediction with open science
Potency optimization is rarely the primary bottleneck in drug discovery. Instead, the majority of medicinal chemistry resources are consumed by the complex task of balancing pharmacokinetics with the mitigation of off-target interactions. Challenges such as cytochrome P450-mediated drug-drug interactions and hERG-induced cardiotoxicity remain persistent hurdles that can derail a program in its final stages.
Unlike potency optimization, where computational tools like Free Energy Perturbation (FEP) provide a rigorous framework for decision-making, ADMET optimization has historically relied on empirical heuristics and trial-and-error. To formalize this challenge, Mark Murcko and James Fraser recently coined the term “Avoidome”, a collective map of the targets and biological pathways that drug candidates must navigate around to ensure safety and efficacy.
OpenADMET addresses the Avoidome through a systematic, integrated approach that synthesizes three critical pillars: targeted data generation, high-resolution structural insights from X-ray crystallography and cryo-EM, and predictive machine learning. This framework moves beyond the frustrating, reactive “whack-a-mole” cycle by providing a deeper understanding of the molecular determinants of off-target binding. By shifting from empirical trial-and-error to a proactive strategy grounded in structural and data science, OpenADMET enables the development of reusable design principles to navigate off-target space and accelerate the path to the clinic.
11:00 – 11:30
Adrian Stevens
Increasing clinical drug success rates: Applying AI-driven mechanistic biosimulation decision-making into drug discovery
This session will highlight how an integrated, AI-enabled modeling and simulation platform can deliver deeper predictive insights and workflows from across the drug discovery process, with the aim of improving clinical success rates for compounds in discovery.
Speaker: Adrian Stevens, PhD, Chief Product Officer, Certara Chemaxon
11:30 – 12:00
Adrian Stevens, David Lowis
Aligning the early-stage discovery strategies across Chemaxon and Certara
Certara’s vision is to deliver an integrated biosimulation experience that accelerates Model-Informed Drug Discovery and Development. Our strategy centers on connecting applications across our portfolio to create seamless workflows. In this session, we’ll share recent progress in drug discovery software development and outline plans to unify key products and services. By building an informatics ecosystem that links chemistry, biology, and data science, we aim to enable smarter candidate selection and improve clinical trial outcomes. We’ll highlight enhancements in small molecule and sequence-based informatics—boosting performance, expanding entity support, and strengthening interoperability.
Speakers:
Adrian Stevens, PhD, Chief Product Officer, Certara Chemaxon
David Lowis, DPhil, Executive Director, Scientific Informatics, Certara D360
12:00 – 12:30
Justin Bryans
Supercharging discovery with intelligent deployment of AI and predictive models
Intelligent deployment of AI, predictive models, and automation is transforming early drug discovery. CRL’s full‑stack, end‑to‑end discovery platform that integrates human expertise with generative AI, physics‑based design, ML‑driven ADME prediction, structural biology, and rapid 5‑day DMTA cycles can deliver higher‑quality clinical candidates in roughly half the industry-standard timelines and with double the success rate. Case studies demonstrate the platform’s ability to rapidly generate novel chemical matter, solve complex design challenges, rescue failing programs, and progress assets to clinical and commercial milestones. The approach leverages human expertise augmented by advanced computational tools to drive confident decision‑making, accelerate hit‑to‑candidate workflows, and deliver differentiated, first‑in‑class therapies.
Speaker: Justin Bryans, Chief Scientific Officer (Discovery), Charles Rivers Laboratories
12:30 – 1:00
Rajarshi Guha
Developing predictive models by sharing predictions
The development of high‑quality predictive models in drug discovery is frequently limited by the cost and scarcity of experimental data, especially for key pharmacokinetic endpoints. We describe the use of the Student–Teacher modeling framework, that enables multiple pharmaceutical companies to collaboratively build predictive models without sharing proprietary data or internal methodologies. In this approach, participating organizations train internal teacher models and generate predictions for a shared public compound set. A neutral honest broker aggregates and anonymizes these predictions, which are then used to train a single student model that captures the collective knowledge of all contributors.
Using volume of distribution at steady state (Vdss) as a case study we show that the resulting student model consistently outperforms individual teacher models and demonstrates strong performance on an external benchmark. Model performance improves as additional contributors participate, without reliance on any single model.
This simple, low‑cost, and model‑agnostic workflow illustrates how structured precompetitive collaboration can unlock greater value from existing data, accelerate model development, and expand the effective use of predictive modeling earlier in the drug design cycle.
Speaker: Rajarshi Guha, Senior Director, Data & Computational Sciences, Vertex Pharmaceuticals
1:00 – 2:00
Lunch & demos
2:00 – 2:30
Krista Gipson, Cen Gao
Data standardization and visualization in small molecule drug discovery for the laboratory of the future
The rapid evolution of artificial intelligence (AI) and automation technologies is reshaping the landscape of small molecule drug discovery. As research workflows become increasingly digitized and integrated, the need for robust data standards and effective visualization methods has never been greater. This talk will explore the critical role that data standardization plays in enabling seamless data integration, sharing, and interpretation across multidisciplinary teams and advanced laboratory environments. We will showcase visualization tools and techniques that empower researchers to extract actionable insights from complex, high-dimensional datasets generated in AI-driven labs.
Speakers:
Krista Gipson, PhD, Senior Scientist, Takeda Pharmaceuticals
Cen Gao, Head of Computational Chemistry, Takeda
2:30 – 3:00
Ania Hajdukiewicz
End‑to‑end control of chemical compound shipments: Integrating compliance into the scientific supply chain
Managing the movement of chemical compounds in a research environment requires continuous regulatory oversight across the entire lifecycle – from purchasing through final shipment. This presentation describes how Novartis Biomedical Research controls the end‑to‑end (E2E) lifecycle of chemical compounds by embedding compliance directly into its internal shipping workflow.
Using the Nomex internal shipping system, every shipment request is evaluated based on material type and destination, automatically triggering the appropriate compliance checks. For chemical compounds, this includes Controlled Substance Screening, Dangerous Goods Classification, and – where applicable – tariff classification, valuation, and permits and licenses for exports. Each compliance step must be completed and validated before a shipment can progress, ensuring regulatory requirements are met before release.
A key element of this process is the integration of Chemaxon Compliance Checker, which screens chemical structures using recognized identifiers such as SMILES, CAS numbers, or peptide sequences. The tool identifies regulatory controls across jurisdictions and flags potential issues for expert review, preventing non‑compliant materials from moving forward. Proprietary chemical information is deliberately excluded from shipping documents to protect intellectual property.
By embedding compliance screening into operational execution, Novartis ensures shipments are fully assessed, internally controlled, and audit‑ready – demonstrating how compliance can function as an integral part of the scientific supply chain rather than a downstream checkpoint.
Speaker: Ania Hajdukiewicz, Global Head of Trade Compliance, Novartis Biomedical Research
3:00 – 3:30
Nicholas Rioux
Tying it all together - Agentic strategies for disconnected research systems
Life science operations are inherently complex, researchers juggle procurement workflows, supplier negotiations, compliance requirements, and scientific data across a patchwork of disconnected platforms. Agentic AI changes the calculus here fundamentally. Rather than forcing researchers to manually bridge these gaps, agents can act autonomously across systems: pulling catalog data, cross-referencing regulatory constraints, and triggering purchase orders in a single coordinated workflow. The key insight is that agents don’t just automate individual tasks; they orchestrate across the seams between systems, which is precisely where the most friction and the most opportunity lives in research operations.
3:30 – 4:00
Afternoon break
4:00 – 5:00
David Fajgenbaum, MD
Keynote presentation
5:00 – 5:15
Close of day
Wednesday, April 15, 2026
7:30 – 8:30
Breakfast & badge pick up
8:30 – 9:00
Sean McGee
From protocol to submission: an optimized digital workflow
Like a decathalon, a clinical study is a series of challenges. Selecting the right endpoints. Drafting a protocol. Setting up data collection. All this needs to happen before the first patient visit. From there, it’s wave after wave of data validation and analysis, reporting and table-creation, and the million-and-one checkpoints of submission.
How much time, cost, and pain might be saved by relying on a suite of interconnected solutions, each purpose-built for its task? Find out in this fast-paced demonstration of Certara capability from protocol to submission. In 30 minutes, you’ll see how our increasingly AI-driven technology supports your organization right from start, when a study is little more than a justified hope. Sean McGee will present solutions for endpoint selection, optimal trial design, data standardization, analysis, reporting, and more. Live 5-plus years of the trial life in a half-hour, all with Certara by your side.
9:00 – 9:30
David Lowis
Supporting the Analysis and Design of Peptide Therapeutics in D360
Research into peptide therapeutics has accelerated over the last two years with interest growing more rapidly than other modalities. We will cover customer driven capability improvements in D360 that support the representation, analysis of and design of new peptide sequences that combine with D360’s data access capabilities and modality agnostic analytics to improve the efficiency and effectiveness peptide research.
Speaker: David Lowis, Executive Director, Scientific Informatics, Certara D360
9:30 – 10:00
Dóra Barna
The sketcher that stays out of your way: Marvin in the modern research environment
Every chemist knows the small annoyances that add up over a day: switching between tools, reformatting structures for a report, hunting for the right template, wondering whether that stereocenter is correctly encoded. Solving these inconveniences is the promise behind the new Marvin.
The best drawing tool is the one you stop thinking about. It works in your ELN, in your registration system, in your Word document, behaving consistently and getting out of the way so you can focus on the science.
In this session we explore how this future can be achieved. We look at migration experiences from earlier versions and the integration of Marvin into existing environments: what worked, and where the new capabilities made a genuine difference. We also look at what comes next: the push toward broader modality support means that the same tool chemists rely on today for small molecules is being extended to serve a wider scientific community.
10:00 – 10:30
Jonathan Buttrick
Improving data quality after registration system migrations
Registration system migrations inherently come with many difficult decisions regarding data quality and harmonization. These decisions often lead to pain points, both expected and unexpected, that require efforts to alleviate. This presentation will share a project aimed to resolve such pain points experienced following the migration of multiple regional registration systems from multiple vendors into one global deployment of Chemaxon Compound Registration at Takeda.
Speaker: Jonathan Buttrick, Principal Scientific Application Manager, Kalleid | Takeda
10:30 – 11:00
Morning break
11:00 – 11:30
Xin Zhang
The making of our cheminformatics platform: Building a connected foundation for data-driven discovery
High-quality chemical data and seamless workflows are essential for effective drug discovery, yet organizations often struggle with fragmented systems, inconsistent compound registration, and limited visibility across the design cycle.
At Cellarity, we developed an integrated discovery platform built on a cloud-based data foundation and powered by Chemaxon’s Compound Registration and Design Hub. By addressing challenges such as duplicate structure detection, inconsistent vendor submissions, and unstandardized data, we established a unified environment seamlessly integrated molecular design tracking, SAR analysis, and real-time predictive ADME modeling, enabling more efficient collaborative and data-driven decision-making.
In this presentation, we will introduce our D3S framework for enterprise solution development, share our multi-year journey, and highlight key lessons in implementation and change management, along with how embedding AI-driven predictive technologies to connect compound registration, design, and prediction accelerates discovery and drives continuous innovation.
Speaker: Xin Zhang, Senior Director of Cheminfo and Platform Integration, Cellarity
11:30 – 12:00
József Kozma
Compound Registration: What changed, what didn't, and why it matters
Drug discovery no longer fits neatly inside one chemistry. Pipelines are growing broader, and the systems that support them need to grow along with that reality. This talk is a checkpoint on that journey for Compound Registration: what we’ve built on since we last met, what we’re working toward, and what all of this means for the people who rely on this tool as a cornerstone of their discovery infrastructure. Some things have shifted considerably; others are exactly as they were, by design. We’ll explore both: where the product has grown into new shapes, and where its identity as a flexible, reliable, scientifically rigorous system remains the thread that runs through everything.
Speaker: József Kozma, Associate Product Manager Compound Registration, Certara
12:00 – 1:00
Jan Christopherson, Jeremy Malerich
Chemaxon portfolio: Recent updates and what's next
A rapid-fire overview of where the Chemaxon portfolio stands and where it’s heading. We’ll walk through recent updates and upcoming roadmap items across the core products including Marvin, Design Hub, Compound Registration, and Compliance Checker, touching on performance improvements, new capabilities, interoperability upgrades. A useful checkpoint for anyone working with or planning around these tools.
Speakers:
Jan Christopherson, Senior Solution Consultant
Jeremy Malerich, Solution Consultant, Certara
1:00 – 2:00
Lunch & demos
2:00 – 3:30
Jeremiah Malerich, Jan Christopherson
Introducing Chemaxon Python API
This tutorial will walk through two of our exciting technologies, the upcoming Python API and Design Hub. With options for hands-on participation or simple audience observation, we will familiarize you with the Chemaxon Python API which will soon come out of its Beta phase, culminating in the creation of a new predictive Machine Learning model. We’ll then incorporate that model into the hypothesis and compound design features of Design Hub, where you’ll also learn about the other capabilities that make it an ideal platform for collaboration in Med Chem teams.
Speakers:
Jeremiah Malerich, PhD, Solution Consultant, Certara
Jan Christopherson, Senior Solutions Consultant, Certara
While all forms of participation are welcome at our workshop, attendees who wish to follow along hands-on, you will need to bring your own laptop. All steps will be carried out in cloud environments, so there will not be a need to download or install any software.
Reserve your spot
Reserve your spot for the Certainty US 2026 customer event for exclusive access to the Certara community of experts, industry leaders, and peers as we collectively explore the latest innovations and opportunities to bring greater certainty to drug discovery and development.
