Unlocking High-Throughput Biology in Drug Discovery – View Speaker Presentations
Industry and academic scientists presented new paradigms in discovery, applications, and workflows, including HT-SPR.
The speakers included scientists from:
You can view slides from the presentations below.
Jinqiu Wang, PhD, Senior Scientist, Prellis Biologics, Inc
Rapid Fully Human Antibody Discovery via High-throughput SPR Characterization of Targets with Diverse Formats
Abstract: The EXIS™ platform at Prellis Biologics accelerates antibody discovery by integrating natural immune processes with advanced AI to navigate the complexity of the human immune repertoire. Using two-photon laser bioprinting, we generate 3D organoids that recapitulate the complexity of immune responses in vitro. Antibodies identified through this platform are characterized using Carterra LSAXT, supporting a portfolio of fully human therapeutic candidates across multiple disease target areas. Discovery targets span diverse structural formats, including peptides, monomers, multimers, nanodiscs, and virus-like particles. We present optimized screening and analysis strategies that enable efficient characterization and pilot lead selection across these multiple formats.
Rui Yin, AI Scientist & Group Lead, Absci
Origin-1: A Generative AI Platform for De Novo Antibody Design Against Novel Epitopes
Abstract: Generative artificial intelligence has advanced antibody discovery, yet de novo design of therapeutic antibodies against targets with “zero-prior” epitopes remains a fundamental challenge. We define “zero-prior” epitopes as target sites lacking structural data from any reported antibody-antigen or protein-protein complex involving the target. Here we present Origin-1, a generative AI platform that overcomes this by integrating epitope-conditioned all-atom structure generation, paired complementarity determining region sequence design, and a specialized co-folding-based scoring protocol to select antibody designs predicted to be high-confidence, specific binders with favorable developability. We evaluated Origin-1 on a panel of ten targets selected to have no available protein–protein complex structures and minimal homology (≤60% sequence identity) to proteins with known complexes, creating stringent design conditions. In fewer than one hundred design attempts per target, we identified developable, specific antibodies, validated across multiple biophysical and developability assays, for four targets: COL6A3, AZGP1, CHI3L2, and IL36RA, with functional inhibition demonstrated for IL36RA. Cryogenic electron microscopy confirmed the atomic accuracy of our designs, revealing complexes that closely matched the computational models with high structural fidelity (3.0-3.1 Å resolution; 0.73-0.83 DockQ). Furthermore, we employed AI-guided affinity maturation to optimize a de novo antibody against IL36RA into a functional antagonist with 104 nM potency. This design was further improved to 590 pM affinity through a second round of computational affinity maturation. These results demonstrate a framework for targeting epitopes without structural precedent, expanding the programmable therapeutic antibody landscape.
Jon Barricelli, Senior Scientist, AstraZeneca
A Few Degrees of Separation: Making Sense of SPR Kinetics Across Temperatures
Abstract:The kinetic values measured for antibody-antigen interactions are inherently sensitive to temperature and are driven by underlying thermodynamic forces encompassed by the epitope-paratope interface. These forces are not easily predicted in large molecule-antigen complexes and, therefore, require empirical confirmation. Here, we report a detailed comparison of binding kinetics measured at both 25°C and 37°C over multiple screening campaigns and identify large changes (both increases and decreases) in association and dissociation rates. We also underscore how these changes have important consequences for candidate selection, in vivo translatability, and can impact and sometimes fundamentally alter decision-making, and project trajectories.
Hetal Marble, PhD, Chief Business Officer, Immuto Scientific
Multimodal Fusion of Empirical Data and Deep Learning for Improved Modeling of Antibody–Antigen Complexes and Accelerated Hit-to-Lead and Optimization of Biologics
Abstract: The best antibodies are those that exhibit specificity to disease or targets, that limit liabilities for production and therapeutic delivery, and that demonstrate the best function – but these are difficult to predict when relying on the typical antibody characterization approaches. By leveraging high-throughput, high-resolution structure determination and combining it with bioanalytical and functional assessments, we systematically selected for the antibodies exhibiting the most desirable attributes against a number of mechanistic targets. Additionally, using Immuto’s multimodal AI framework, fusing radical footprinting, bioanalytical, and functional data with deep learning to more accurately model antibody–antigen complexes, these experimental constraints enabled advanced, actionable models for antibody engineering, affinity maturation, and specificity design.
Weijie Zhou, PhD, Senior Scientist, Merck
Implementing High-Throughput Epitope Binning with Carterra for Early-Stage Biologics Hit Triage
Abstract: We describe our implementation of Carterra’s high-throughput SPR platform, emphasizing epitope binning to accelerate early-stage biologics hit triage across multiple therapeutic areas. By integrating binning data into our discovery workflow, we rapidly classify hit diversity, prioritize complementary binding profiles, and anticipate functional behavior observed in downstream assays. Case examples show how epitope-resolved clustering helps explain divergent assay outcomes and guides selection of candidates with distinct mechanisms of action. This approach reduces downstream attrition, informs lead selection strategies, and enhances cross-program efficiency by providing actionable structural and functional insight at an early decision point.
Katie Smith, Associate Scientist, GSK
Accelerating Early Antibody Discovery at GSK Through High Throughput Screening with the Carterra LSAXT
Abstract: Early antibody discovery requires rapid, data‑driven decision‑making to efficiently progress high‑quality candidates through pre-clinical development. At GSK, we are integrating high‑throughput experimental platforms with robust data analysis workflows to enable timely and informed lead selection.
This presentation will provide an overview of GSK’s early antibody discovery process, with a focus on the screening and characterisation stages that generate key decision-making data for program progression. We will describe how the Carterra LSAXT is embedded within our current workflow to enable high‑throughput kinetic and epitope binning analyses, significantly increasing screening capacity while reducing experimental turnaround times.
In addition, we will demonstrate how experimental outputs from the Carterra platform can be integrated into our data workflows, supporting efficient comparison of candidates and streamlining decision‑making across project teams.
Ilse Roodink, PhD, Chief Scientific Officer, AVS Bio
Data-rich Discovery and Optimization to Expedite Lead Antibody Generation
Abstract: The importance of diversity during the discovery phase of therapeutic antibody campaigns is widely recognized. However, to maximize success, such diversity should be accompanied by meaningful, data-rich outputs to guide hit triaging and lead candidate optimization approaches. The presented case study highlights an integrated multi-parametric platform leveraging high-throughput, industry-leading wet lab characterizations – including SPR-based profiling -, and highly scalable, in silico-driven developability assessment and optimization via AVS Bio’s AbRefineTM to expedite lead antibody generation. The seamless integration of AVS Bio’s computational AbRefineTM toolbox in the lead antibody process empowered diversity-driven antibody discovery and optimization, and advanced the delivery of optimized antibodies for further clinical development.
Michelle Vandeloo, MS, Research Associate and Lab Automation Engineer, Cradle
Emma Beadle, Research Associate, Cradle
Accelerating Antibody Engineering with Cradle’s Machine Learning Models Supported by Automated HT-SPR
Abstract: Cradle provides machine learning models that enable labs to accelerate protein engineering. We continually iterate on and improve our models using data from our automated, in-house wet lab, which generates high-quality, high-throughput data in as little as one week. The HT-SPR platform from Carterra facilitates parallel kinetic characterization of hundreds of antibody fragment variants against multiple targets, providing essential antibody engineering datasets for our models. We demonstrate this capability through the multi-round optimization of an anti-SARS-CoV-2 VHH, and the hit identification and optimization of a polyspecific snake antivenom from panning data. Furthermore, the Cradle lab’s modular approach to assay design enables us to characterize more than just binding. We present some of our additional SPR assays including studying pH-dependent binding, membrane protein targets, and epitope binning.
Carterra Application Scientists
Rapid Efficient Workflows for Nearly all Sample Types from Antibodies to Fragments—Introducing the 48-Channel Carterra Vega™ SPR Platform
Abstract: Surface plasmon resonance (SPR) has long been the gold standard for characterizing molecular interactions in drug discovery, but throughput limitations have historically constrained its use in early-stage screening campaigns where large numbers of compounds or biologics candidates require rapid triage. With high sensitivity, high throughput biosensors, Carterra’s biosensors enable SPR screening and characterization at a new scale. The Ultra is the newest in the line of one on many biosensors, building on the one-on-many array-based architecture of Carterra’s LSA platforms which has been broadly adopted in biologics discovery, now with the sensitivity and sample handling features to enable small molecule and fragment assays. Carterra Vega™, Carterra’s newest platform, moves from the one-on-many format to a new architecture with 48 needles and 48 parallel flow cells each with two ligands and a reference. Paired with an optional plate handling robot, Carterra Vega can process up to 20,000 analyte compounds in a day against two targets. A 384-well-plate of analytes can be injected in as little as 35 minutes. This enables new approaches like dose responses in primary fragment screens and kinetic characterization of thousands of compounds a day. This talk will highlight the range and scale of assays now possible on these platforms.