Biology subjects

Deane, C. M.

Publications and source records attributed to Deane, C. M..

At least 19 recordsLinked to original sources

bioRxiv2026

The first OpenBind release: An open experimental structure-affinity dataset and benchmark for structure-based AI

High-quality experimental datasets that link protein-ligand structures with binding affinity data are essential for developing and evaluating structure-based machine learning methods. To help address this need, we established OpenBind as an open-science initiative to generate large-scale experimental datasets for structure-based AI and molecular discovery. Here, we describe the first public OpenBind release, which, to the best of our knowledge, is the largest public single-target experimental structure-affinity dataset. The dataset focuses on enteroviral 2A protease, comprising 925 crystallographic binding events from 699 compounds and associated affinity measurements for 601 compounds. It combines structures from an initial fragment screen and follow-on molecules, together with affinity data, linking experimentally determined protein-ligand binding modes to biophysical measurements within a coherent antiviral discovery campaign. We used this dataset to evaluate protein-ligand structure prediction, binding-affinity prediction, and virtual screening using representative structure-based methods, including docking and cofolding. This exposed several challenges that are central to practical structure-based modelling: docking performance depends strongly on binding-pocket conformation, poses are difficult to rank, and structure-based affinity prediction remains challenging. Fine-tuning OpenFold3-p2 on the fragment-screen structures substantially improved pose prediction and virtual screening for related follow-on compounds, demonstrating how early-stage experimental structures can support target-specific model adaptation.

bioinformatics↗

bioRxiv2026

Fast retrieval of structurally similar antibodies from large sequence databases with AbSLang

The first steps in antibody therapeutic discovery involve identification of sequences with desirable binding properties. A way of finding these lead molecules is through the search of large sequence databases. Current methods, due to the size of databases, rely on germline or complementarity-determining-region (CDR) sequence identities, overlooking structurally similar antibodies with divergent sequences which can have identical binding properties . To address this, we introduce AbSLang, a model trained for pairwise CDR RMSD prediction using a contrastive learning approach. We demonstrate that AbSLang has comparable accuracy to exact RMSD calculation after explicit structure prediction with state-of-the-art models. Building on this model, we implemented AbSLang-search, a pipeline for retrieval of structurally similar antibodies from large sequence databases. AbSLang-search is highly compute efficient and allows to search datasets with 10 million sequences in less than 2 seconds.

bioinformatics↗

bioRxiv2026

SAbDab2: The structural antibody database in the age of machine learning

The Structural Antibody Database (SAbDab) is a publicly available repository of experimentally determined antibody structures, first released in 2013. Explicit support for single-domain antibodies was added in 2021, with SAbDab-nano. Recently, increasing interest in antibodies has led to a proliferation of novel antibody formats, while simultaneous advances in machine learning have increased demand for standardised, high-quality structure data. Here, we present SAbDab2, re-engineered for the machine-learning age. It introduces support for a variety of new formats, and makes it easy to retrieve and compare all known structures of a given antibody. In addition, SAbDab2 provides ready access to ML-grade structures of antibody and antibody-antigen-complexes, with standardised, versioned train/test splits. These will be updated every six months going forward, and are available at https://zenodo.org/records/20083995. SAbDab2 itself is updated weekly and is freely available at https://sabdab2.opig.stats.ox.ac.uk.

bioinformatics↗

bioRxiv2026

Modelling antibody structures at the speed of language

Protein structure prediction is currently substantially slower than obtaining sequence representations of proteins. This leads to most property prediction methods relying solely on trivial or learned sequence embeddings. However, contemporary structure prediction and sequence models are both based on Transformers, and structure prediction models often have fewer parameters, suggesting that there might be domains where accurate structure prediction adds no practical overhead to sequence-only modelling. Here, we demonstrate this can be achieved for adaptive immune proteins by introducing FlashABB, which predicts highly accurate antibody structures, and does so faster than even modestly-sized language models can embed sequences. As a component of FlashABB, we develop Flashpoint Attention, a fast and linear memory analog of Invariant Point Attention. To our knowledge, FlashABB is the first example of a model that accurately predicts protein structure faster than protein language models can generate embeddings, enabling efficient access to 3D information without the need for precomputed structures. Using FlashABB, we develop methods for predicting antibody stability and developability which can be scaled to repertoires of millions of sequences. Our results show how the computational bottleneck of protein structure prediction can be removed in some real-world cases. The code and model weights for FlashABB are available on GitHub: https://github.com/oxpig/FlashABB

molecular biology↗

bioRxiv2026

Reverse engineering the fatally cross-reactive A3A TCR to decouple potency and specificity

T cell receptor (TCR) affinity enhancement can introduce off-target cross-reactivity with life-threatening consequences, as illustrated by the MAGE-A3-specific A3A TCR, which caused fatal cardiotoxicity through recognition of a Titin-derived peptide. Here, we reconstructed the cross-reactivity landscape by reverse-engineering A3A toward its wild-type precursor, generating intermediate variants in which engineered CDR2 residues are systematically reverted to the wild-type sequence. Reverting just two engineered residues yields a receptor, v9, that retains MAGE-A3 cytotoxicity comparable to A3A while eliminating Titin and other acquired cross-reactivities. Structurally, these substitutions reduce CDR2-MHC contacts and disrupt an intra-TCR CDR2-CDR3{beta} interaction, propagating conformational changes across CDR3 loops that reshape peptide engagement without altering docking geometry. These results demonstrate that mutations outside the peptide-contacting CDR3 loops can allosterically reconfigure antigen specificity and establish simple stepwise reverse engineering to wild-type as a strategy for correcting TCR cross-reactivity.

immunology↗

bioRxiv2026

Sampling antibody conformational ensembles withABodyBuilder4-STEROIDS

Conformational flexibility is fundamental to the function of many proteins and in the case of antibodies can impact key properties such as affinity and specificity. While it is possible to predict single, static protein structures with high accuracy, predicting conformational ensemble remains challenging. Molecular dynamics simulations suffer from high computational costs, while deep learning methods are yet to achieve the same level of accuracy. Here, we introduce ABB4-STEROIDS a generative structure prediction model that samples conformational ensembles of antibodies. We trained our model on 4.2 million structural frames derived from [~]136,000 coarse-grained and a set of 83 new all-atom antibody MD simulations. We benchmarked our model on reproducing MD ensembles and evaluated the diversity of sampled structures and the covered conformational space against experimental evidence. ABB4-STEROIDS achieves state-of-the-art accuracy, particularly within the experimental benchmarks. The model is openly available and provides a robust resource for large-scale investigations of antibody conformational ensembles.

bioinformatics↗

bioRxiv2026

Teaching Diffusion Models Physics: Reinforcement Learning for Physically Valid Diffusion-Based Docking

Molecular docking aims to predict the binding conformation of a small molecule to its protein target. Recent work has proposed diffusion models for this task, from rigid-body docking that diffuses over ligand degrees of freedom to co-folding approaches that jointly generate protein structure and ligand pose. However, diffusion-based docking models have been shown to frequently produce physically implausible poses and fail to consistently recover key protein-ligand interactions. To address this, we introduce a reinforcement learning framework for training diffusion-based docking models directly on non-differentiable objectives. Fine-tuning DiffDock-Pocket for physical validity with our approach substantially increases the number of generated poses that are physically valid and interaction-preserving, with no increase in inference-time compute. Importantly, this comes without sacrificing structural accuracy; in fact, our approach increases the proportion of structures with near-native poses. These effects are most pronounced for protein targets that are dissimilar to the training data. Our fine-tuned DiffDock-Pocket model outperforms both classical docking algorithms and machine learning-based approaches on the PoseBusters set. Our results demonstrate that reinforcement learning can teach diffusion-based docking models to better respect physical constraints and recover key interactions, without the requirement to rely on inference-time corrections.

bioinformatics↗

bioRxiv2025

Structural basis for synergistic antibody protection against the essential malaria invasion complex protein RIPR

Plasmodium falciparum RH5-interacting protein (RIPR) is central to the essential PTRAMP-CSS-RIPR-CyRPA-RH5 (PCRCR)-complex; a leading target of blood-stage malaria vaccines. However, mechanisms whereby anti-RIPR antibodies inhibit parasite invasion are poorly understood. Here, we characterise 83 human IgG mAbs from RIPR-vaccinated Kymouse platform mice. Single mAbs have minimal neutralising activity, however, high-level synergistic inhibition is observed with pools of mAbs targeting the RIPR-Tail region. Structural characterisation and molecular dynamics simulations of RIPR-Tail show that mAbs targeting EGF-like domains 6-8 (RIPREGF (6-8)), but not EGF-like domains 9-10 or the C-terminal domain (RIPREGF (9-10)-CTD), synergise to constrain the RIPR-Tail conformation. The same antibodies dissociate PTRAMP-CSS from RIPR, thereby enabling anti-RIPREGF (9-10)-CTD mAbs or anti-CSS sdAbs to bind and potentiate anti-RIPREGF (6-8) IgG. Addition of these mAbs to IgG from humans immunised with the R78C (RIPREGF (7-8)-CyRPA) candidate vaccine enhances malaria growth inhibition. These data provide a framework to guide next-generation blood-stage malaria vaccine design.

immunology↗

bioRxiv2025

Uncovering the flexibility of CDR loops in antibodies and TCRs through large-scale molecular dynamics

Antibody structures are composed of framework regions that adopt a conserved fold and complementarity determining regions (CDR) loops which are far more variable. Flexibility of CDR loops has been linked to key properties such as affinity and specificity. However, owing to the scarcity of available data it has not been possible to study the functional implications of their dynamics in detail. To overcome these data limitations, we introduce CALVADOS 3-Fv, a customised set-up of the residue-based CALVADOS 3 model, utilizing restraints and parameterization tailored for immune receptor simulations. CALVADOS 3-Fv reproduces ensemble metrics in all atom simulations and experimental data with high accuracy. Having validated our protocol, we created FlAbDab and FTCRDab, two databases containing simulations of more than 150,000 antibodies and T-cell receptors. The databases are released open source to enable the study of CDR dynamics and as a large data source for training machine learning models.

biophysics↗

bioRxiv2025

Pox-AbDab: the Orthopoxvirus Antibody Database

In August 2024, the World Health Organisation declared the mpox orthopoxvirus to be a Public Health Emergency of International Concern for the second time in three years, emphasising the need for continued studies into its microbiology and potential therapeutic interventions. Here, we present the Orthopoxvirus Antibody Database (Pox-AbDab), a repository of data on antibodies known to bind or neutralise viruses from the same genus as mpox (https://opig.stats.ox.ac.uk/webapps/poxabdab). Beyond standardising and centralising the data, we highlight challenges in translating knowledge across orthopoxviruses, such as the absence of a function-based nomenclature for virion surface antigens. We also performed an exploratory analysis of the known orthopoxvirus-binding antibody landscape, highlighting their aggregate molecular properties, cross-binding/cross-neutralisation profiles, evidence for immunodominance or immune escape from their epitopes, and gaps in coverage to help orient future research.

immunology↗

bioRxiv2025

The Therapeutic Nanobody Profiler: characterising and predicting nanobody developability to improve therapeutic design

Developability optimisation is an important step for successful biotherapeutic design. For monoclonal antibodies, developability is relatively well characterised. However, progress for novel biotherapeutics such as nanobodies is more limited. Differences in structural features between antibodies and nanobodies render current antibody computational methods unsuitable for direct application to nanobodies. Following the principles of the Therapeutic Antibody Profiler (TAP), we have built the Therapeutic Nanobody Profiler (TNP), an open-source computational tool for predicting nanobody developability. Tailored specifically for nanobodies, it accounts for their unique properties compared to conventional antibodies for more efficient development of this novel therapeutic format. We calibrate TNP metrics using the 36 currently available clinical-stage nanobody sequences. We also collected experimental developability data for 108 nanobodies and examine how these results are related to the TNP guidelines. TNP is available as a web application at opig.stats.ox.ac.uk/webapps/tnp.

bioinformatics↗

bioRxiv2025

LICHEN: Light-chain Immunoglobulin sequence generation Conditioned on the Heavy chain and Experimental Needs

In developing therapeutic antibodies, the heavy chain is often prioritised due to its higher variability and its central role in antigen binding. An appropriate pairing of the light sequence is however important for antibody function. Here we present LICHEN, a heavy chain conditioned light sequence generation tool that enables collaborative light sequence design by leveraging computational capabilities alongside experimental expertise. LICHEN generates light sequences which are valid (antibodylike), diverse in sequence and structure, and conditioned on a specific heavy chain. LICHEN can also condition on germline and CDRs and automatically filter generated sequences for required properties. This allows LICHEN to be used across multiple antibody development use cases. We carry out experimental validation of the method conditioning only on the heavy sequence and on the heavy sequence and binding information. Our in vitro results show that sequences created by LICHEN have effective expression yields and can retain antigen-binding.

bioinformatics↗

bioRxiv2025

STCRpy: a software suite for T cell receptor structure parsing, interaction profiling and machine learning dataset preparation

SummaryComputational methods to guide early-stage TCR drug discovery and TCR repertoire informatics currently under-utilise solved and predicted structure data. Here, we streamline use of these data through an open-source python package for high-throughput TCR structure handling and analysis (STCRpy), facilitating analyses such as TCR:peptide-MHC complex orientation calculation/scoring, root-mean-square-distance evaluation, interaction profiling, and machine learning dataset curation. Availability and implementationFreely available as a Python package at https://github.com/npqst/STCRpy. Contactdeane@stats.ox.ac.uk, raybould@stats.ox.ac.uk

immunology↗

bioRxiv2025

ANARCII: A Generalised Language Model for Antigen Receptor Numbering

Antigen receptor numbering allows the rapid delineation of the antigen-binding regions of antibody and T cell receptor (TCR) sequences, from sequence alone. It also allows the comparison of the vast diversity of antigen receptors in a consistent frame of reference. Numbering of antigen receptors is currently achieved by aligning sequences to a reference set. This approach may result in different numbering, depending on the reference set used or may fail to number query sequences derived from new species or rare sequence types. To address this problem, we have built a new numbering method (ANARCII) which requires no alignment step and is based on a Seq2Seq language model. Our results show that ANARCII can deal with the complexity that arises in experimentally collected sequencing data and generalise to sequences which are highly dissimilar to those in training. In test sets designed to contain challenging and ambiguous sequence patterns ANARCII numbering was identical to existing methods for over 99.99% of conserved residues and over 99.94% for complete CDR regions. The lightweight architecture allows numbering of over 90,000 sequences per minute on a single A100 GPU. Furthermore, the ANARCII package can be conditioned to fit rare sequence types and provide new training data for fine-tuning. We demonstrate that fine-tuned versions of ANARCII can correctly number other immunoglobulin domains such as TCRs and VNARs. Our model is freely available as a web tool (https://github.com/oxpig/ANARCII), as well as a package for high throughput numbering of next generation sequencing data (https://opig.stats.ox.ac.uk/webapps/sabdab-sabpred/sabpred/anarcii/).

bioinformatics↗

bioRxiv2025

Predicting the conformational flexibility of antibody and T-cell receptor CDRs

AbstractMany proteins are highly flexible and their ability to adapt their shape can be fundamental to their functional properties. We can now computationally predict a single, static protein structure with high accuracy. However, we are not yet able to reliably predict structural flexibility. A major factor limiting such predictions is the scarcity of suitable training data. Here, we focus on predicting the structural flexibility of the functionally important antibody and T-cell receptor CDR3 loops. We extracted a dataset of CDR3 like loop motifs from the PDB to create ALL-conformations, a dataset containing 1.2 million structures and more than 100,000 unique sequences. Using this dataset, we develop ITsFlexible a method classifying CDR3 flexibility, which outperforms all alternative approaches on our crystal structure datasets and successfully generalises to MD simulations. We also used ITsFlexible to predict the flexibility of three completely novel CDRH3 loops and experimentally determined their conformations using cryo-EM.

bioinformatics↗

bioRxiv2025

Expanding the Synthome: Generating and integrating novel reactions into retrosynthesis tools.

Synthesising compounds is a major component of the time and cost of drug discovery. Retrosynthesis approaches have grown in prominence in efficiently predicting synthetic routes. For such tools to be truly impactful they should be able to effectively incorporate rarely seen chemical reactions and synthesise unknown molecules. We show that whilst existing tools can often effectively replicate known routes in familiar chemical spaces, they tend to fail in less explored regions. To overcome this challenge, we present a novel approach that integrates under-utilised reactions, resulting in improved performance in nearly 70% of tested cases. We introduce a tool that identifies, prioritises, and incorporates entirely new chemical transformations, demonstrating an average reduction in predicted synthesis costs of [~] 20%. Importantly, our tool can highlight new chemical reactions that should be prioritised for development to improve synthesis efficiency. Our findings lay the groundwork for next generation retrosynthesis tools, which will be capable of driving the discovery of novel chemical transformations and more efficient exploration of chemical space.

bioinformatics↗

bioRxiv2024

Computational design of developable therapeutic antibodies: efficient traversal of binder landscapes and rescue of escape mutations

Developing therapeutic antibodies is a challenging endeavour, often requiring large-scale screening to produce initial binders, that still often require optimisation for developability. We present a computational pipeline for the discovery and design of therapeutic antibody candidates, which incorporates physics- and AI-based methods for the generation, assessment, and validation of developable candidate antibodies against diverse epitopes, via efficient few-shot experimental screens. We demonstrate that these orthogonal methods can lead to promising designs. We evaluated our approach by experimentally testing a small number of candidates against multiple SARS-CoV-2 variants in three different tasks: (i) traversing sequence landscapes of binders, we identify highly sequence dissimilar antibodies that retain binding to the Wuhan strain, (ii) rescuing binding from escape mutations, we show up to 54% of designs gain binding affinity to a new subvariant and (iii) improving developability characteristics of antibodies while retaining binding properties. These results together demonstrate an end-to-end antibody design pipeline with applicability across a wide range of antibody design tasks. We experimentally characterised binding against different antigen targets, developability profiles, and cryo-EM structures of designed antibodies. Our work demonstrates how combined AI and physics computational methods improve productivity and viability of antibody designs.

bioinformatics↗

bioRxiv2024

Improving Structural Plausibility in 3D Molecule Generation via Property-Conditioned Training with Distorted Molecules

Traditional drug design methods are costly and time-consuming due to their reliance on trial-and-error processes. As a result, computational methods, including diffusion models, designed for molecule generation tasks have gained significant traction. Despite their potential, they have faced criticism for producing physically implausible outputs. We alleviate this problem by conditionally training a diffusion model capable of generating molecules of varying and controllable levels of structural plausibility. This is achieved by adding distorted molecules to training datasets, and then annotating each molecule with a label representing the extent of its distortion, and hence its quality. By training the model to distinguish between favourable and unfavourable molecular conformations alongside the standard molecule generation training process, we can selectively sample molecules from the high-quality region of learned space, resulting in improvements in the validity of generated molecules. In addition to the standard two datasets used by molecule generation methods (QM9 and GEOM), we also test our method on a druglike dataset derived from ZINC. We use our conditional method with EDM, the first E(3) equivariant diffusion model for molecule generation, as well as two further models--a more recent diffusion model and a flow matching model--which were built off EDM. We demonstrate improvements in validity as assessed by RD-Kit parsability and the PoseBusters test suite; more broadly, though, our findings highlight the effectiveness of conditioning methods on low-quality data to improve the sampling of high-quality data.

bioinformatics↗