Jump to content

Building the Foundation for Next-Gen Biologics R&D

October 12, 2026
Alison Waterfield

3D visualization of an antibody molecule suspended in a microscopic biological environment

Biopharmaceutical organizations are under pressure to accelerate discovery and development while managing growing volumes of data, increasingly complex therapeutic modalities, and rising demands for traceability and automation. As AI becomes part of biologics R&D, a fundamental question is emerging: how can organizations build the connected digital foundation needed to turn scientific data into better decisions?

That question brought scientists and technology leaders together at two Genedata Open Forums, hosted by AbbVie in Cambridge, MA, on September 29 and Pfizer in San Diego on October 1. Across customer case studies, Genedata presentations, posters, and demonstrations, participants explored practical approaches to connecting scientific workflows, increasing automation, and creating data foundations for analytics and AI. 

Across both Open Forums, a consistent message emerged: realizing the potential of AI and digitalization requires connected data, scientific context, and workflows that can scale with increasingly complex R&D. 

Four themes stood out:

  • AI-ready R&D starts with connected, contextualized scientific data
  • Connected ecosystems preserve context across scientific workflows
  • AI delivers more value when embedded directly into scientific workflows
  • Digital foundations must accommodate complexity across discovery and CMC 

AI-Ready R&D Starts with Connected, Contextualized Scientific Data 

AI may be changing what is possible in biologics R&D, but its value depends on the data behind it. Experimental results need to remain connected to the scientific relationships, lineage, and decisions that give them context, while governance and quality determine whether that data can be trusted and reused. 

Those requirements surfaced in different ways across the two Open Forums. Debra Touw, Joanne Murphy, and Mary Abdalla of AbbVie showed how a unified data infrastructure built around FAIR principles supports discovery as research expands across locations, therapeutic modalities, and increasingly complex molecules. With Genedata Biologics serving as a hub within the company’s informatics ecosystem, data can be centralized along the path to candidate nomination while remaining connected with external systems and analytical technologies. Custom integrations and automation further support standardized data capture while reducing manual work. 

But a connected infrastructure alone does not make data AI-ready. Drew Lynch of Sanofi emphasized the role of clear data requirements, ownership, quality processes, and monitoring in turning connected research data into trusted, reusable data products Brittany Ciesluk and Drew Nager of Pfizer added another challenge: modernization needs to preserve valuable scientific context. Their migration from a legacy registration environment following the integration of Seagen’s ADC capabilities required translating scientific relationships and lineage, preserving active production workflows, and expanding Genedata’s role within Pfizer’s wider digital ecosystem.

Digital visualization of an antibody connected to a network of scientific data and technology icons, representing integrated research data and drug discovery workflows.

Together, these examples illustrate a broader shift: AI readiness is not simply about making more data available. It depends on making scientific data connected, contextualized, governed, and reusable. 

Connected Ecosystems Preserve Context Across Scientific Workflows 

Digitalizing individual workflows solves only part of the problem. If data loses context as it moves between instruments, workflow applications, registration systems, inventory, and scientific records, researchers can still be left reconstructing the bigger picture. 

Bill Shirley and Epic Ding of Gilead illustrated a different model, with Genedata Biologics operating as part of a connected research ecosystem spanning the ELN, registration, inventory, shared data, and focused user views. The goal is not to make one application perform every task, but to connect specialized systems while preserving scientific context as information moves between them. Jorge Teran of Takeda approached the same challenge at the experimental workflow level, connecting instrument data, scientific context, analysis, QC and selection tools, automated registration, and results rather than relying on disconnected spreadsheets and manual processes. This keeps results linked to project, sample, and experimental context while making more data available for downstream scientific use. 

The value of connectivity therefore lies not in integration itself, but in allowing scientific context to travel with the data so that information remains usable wherever decisions are made. 

Bringing AI into the Scientific Workflow 

Better models alone will not transform biologics R&D. The next challenge is closing the distance between computational prediction and experimental science so that each can inform the other within day-to-day scientific workflows. 

Novartis provided complementary examples of what this looks like in practice. Barbara Brannetti described a closed design-make-test-analyze loop in which AI models learn from successive experimental rounds while scientists remain in control of decisions. The approach combines dry-lab modeling, high-throughput wet-lab experimentation, data automation, and people to support multi-parameter optimization and identify liabilities earlier. 

Kannan Sankar showed how predictive approaches can assess biophysical properties and post-translational modification liabilities earlier in discovery, helping scientists rank promising candidates and deselect those with unfavorable properties. Crucially, the models do not have to remain isolated in computational notebooks: predictions can be generated within Genedata Biologics, retained in scientific context, and evaluated alongside wet-lab results. This reflects the broader platform approach described by Genedata’s Chris Morgan, in which molecule identity, relationships, lineage, and decision history provide the context connecting in silico predictions with experimental results.  

The emerging principle is clear: AI becomes more useful when its predictions are not separated from experimental science, but become part of the same connected workflow in which scientists test hypotheses, evaluate evidence, and make decisions.

Building for Complexity Across Discovery, Development, and CMC

Complexity is increasing in two directions: in the therapeutic modalities scientists are designing and in the development processes needed to advance them. An underlying digital foundation needs to accommodate both. 

For complex antibody formats, that means finding structured ways to represent molecular designs that can involve multiple chains, domains, and combinations. Pfizer’s Joel Bard demonstrated a modular approach based on reusable antibody modules, with automation supporting format and molecule-template creation, naming, visualization, and consistency checks. Standardizing how these designs are represented can make increasingly complex biotherapeutics more searchable and interpretable across discovery workflows. 

As candidates move into development, the nature of the complexity changes but the need for connected context remains. Process development and CMC bring together materials, process parameters, analytical results, characterization, quality data, and instrumentation. Genedata’s Jolyon Terragni and Jessy Sheng demonstrated across the two Open Forums how these data can be connected across upstream and downstream process development, formulation, analytical, and quality activities. Integrated instrument data, automated analysis and reporting, and data transfer can reduce manual handling while maintaining the context required for analysis and decision-making. 

Scientific data management platform consolidating process, characterization, and quality data from multiple laboratory systems.

Together, these examples reinforce that a scalable digital foundation needs to accommodate complexity across the R&D lifecycle, connecting scientific context from molecule design and discovery through process development and CMC.

Beyond the Presentations: Peer Exchange in Practice

The exchange of ideas continued beyond the formal presentations. Across both Open Forums, attendees connected with peers and Genedata experts to explore specific scientific and digitalization challenges and discuss how emerging approaches could be applied within their own R&D environments. Poster and demo sessions complemented these conversations with practical applications spanning biologics discovery, process development, workflow automation, data management, and AI/ML.

Attendees at Bio Open Forum

Posters

The poster sessions provided a closer look at practical applications across biologics discovery and development, spanning AI-supported developability assessment, next-generation conjugates, integrated sample management, protein characterization, and data-driven process development and CMC. Explore the posters presented across the two Open Forums to learn more about these approaches. 

AI-Supported Antibody Developability


View poster

Next-Gen ADC and Antigen-Targeting Conjugates


View poster

Integrated Sample Management From Bench to Freezer


View poster

Structured Knowledge for Bioprocess Development

View poster

AI-Ready Bioprocess Data Foundation



View poster

Digitalizing CMC Processes at Novartis


View poster

Driving Consistency and Data Quality in Protein Mapping


View poster

Automated Stability Analysis For Antibody Developability

View poster

Demo Sessions

Interactive demo sessions translated many of the themes discussed on stage into practical scientific workflows. Across the two Open Forums, demonstrations covered ADC and LNP discovery, in silico molecule assembly, lab data capture, request and workflow management, AI/ML-ready data structures, mass spectrometry data analysis, custom tools, and upstream process development. Together, the use cases showed how connected data and automation can support day-to-day scientific work across biologics discovery, development, and CMC.  

Interested in exploring these workflows in more detail?  

Request a demo to see how Genedata solutions can support your next-generation biologics R&D.

From Digital Foundation to Scientific Impact 

Across the two Open Forums, one message was consistent: building the foundation for next-generation biologics R&D is not simply about generating more data or deploying more digital tools. The greater opportunity lies in preserving scientific context as data moves across workflows, connecting computational and experimental science, and creating infrastructure that can accommodate increasing complexity across the R&D lifecycle.  

Technology will continue to evolve, as will therapeutic modalities, experimental approaches, and the volumes and types of scientific data they generate. A connected digital foundation provides the continuity needed to adapt to that change and turn data, analytics, and AI into meaningful scientific impact.