Biology subjects

Abanades, B.

Publications and source records attributed to Abanades, B..

9 recordsLinked to original sources

bioRxiv2026

Can Folding Models Tell Binders from Bluffers? Evidence from POISK: The Patent-Derived Antibody Dataset

Structure prediction models are increasingly deployed as zero-shot digital screens in antibody design, operating under the assumption that high folding confidence implies true binding. However, rigorous validation of this premise has been limited by small test sets, narrow target diversity, and the risk of data leakage from training-adjacent structures. Here we introduce POISK: Patent-extracted Organization of Immunoglobulin Sequence Knowledge, a large-scale dataset derived from the US patent literature, comprising 89074 antibody-antigen records extracted from 10110 patents using an LLM-based pipeline. From this resource, we curate a balanced benchmark set of 22494 complexes: 11247 patent-validated binders without a known resolved structure paired with putative non-binders constructed from sequence-similar human repertoire antibodies. Using this dataset, we benchmark six leading structure prediction models (AlphaFold-Multimer, Boltz-2, IntelliFold v2, OpenFold3, Protenix-v1, and Protenix-v2) on their ability to discriminate binders from decoys. The best-performing model, Protenix-v2, achieves 72% pairwise accuracy over all pairs and 87% on the most confident half of predictions, and correctly identifies the true binder from a pool of 50 candidates in 29% of trials; this represents a substantial enrichment over random selection, yet remains far from reliable single-candidate identification. Structural consensus across models also discriminates binders from decoys, with pairwise accuracy reaching 73.1% when three or more models converge on a pose (DockQ>0.23, covering around 54.2% of pairs). Epitope localization remains unreliable: the best model contacts annotated epitope residues in only 62.8% of cases, indicating that binding discrimination does not establish accurate epitope localization. Our results suggest that current folding models provide useful but limited signal for antibody screening and should be complemented by orthogonal methods. We release POISK as a community resource for future benchmarking and model training.

bioinformatics↗

bioRxiv2026

MIMOSA: Guiding De Novo Antibody Design with Natural Interaction Fingerprints

De novo antibody design promises to transform therapeutic antibody and nanobody discovery. Yet, success remains inconsistent across targets, with generative models often yielding no experimentally validated binders. Here we introduce MIMOSA (MIMic-Oriented Structural Antibody generation), a model-agnostic, inference-time framework that constrains diffusion-based antibody design pipelines to reproduce the interaction geometry and chemistry of a target's known cognate binder interface. Rather than searching for a productive interface from scratch, MIMOSA constrains generation around residue identities and geometries already known to support binding, while allowing the generative model to complete the surrounding antibody interface. We test MIMOSA on different interface topologies: contiguous motifs that fit within a single CDR loop and spatially dispersed hotspots distributed across the paratope. Applied without retraining to two architecturally distinct generators (RFAntibody and BoltzGen), this framework delivers experimentally validated binders-achieving sub-micromolar to single-digit nanomolar affinities-at 20-26% hit rates on three targets on which unconstrained methods produce zero or near-zero binders: KEAP1, uPA, and IL-8. By turning any target with a structurally characterised cognate binder into an accessible de novo design problem, this framework broadens the practical reach of generative antibody design to targets where unconstrained approaches currently fail.

bioinformatics↗

bioRxiv2024

T-cell receptor structures and predictive models reveal comparable alpha and beta chain structural diversity despite differing genetic complexity

T-cell receptor (TCR) structures are currently under-utilised in early-stage drug discovery and repertoire-scale informatics. Here, we leverage a large dataset of solved TCR structures from Immunocore to evaluate the current state-of-the-art for TCR structure prediction, and identify which regions of the TCR remain challenging to model. Through clustering analyses and the training of a TCR-specific model capable of large-scale structure prediction, we find that the alpha chain VJ-recombined loop (CDRA3) is as structurally diverse and correspondingly difficult to predict as the beta chain VDJ-recombined loop (CDRB3). This differentiates TCR variable domain loops from the genetically analogous antibody loops and supports the conjecture that both TCR alpha and beta chains are deterministic of antigen specificity. We hypothesise that the larger number of alpha chain joining genes compared to beta chain joining genes compensates for the lack of a diversity gene segment. Overall, our study demonstrates that valuable structure-function relationships can lie in alpha chains despite their simpler junctions. We also provide over 1.5M predicted TCR structures to enable repertoire structural analysis and elucidate strategies towards improving the accuracy of future TCR structure predictors.

immunology↗

bioRxiv2023

Investigating the ability of deep learning-based structure prediction to extrapolate and/or enrich the set of antibody CDR canonical forms

Deep learning models have been shown to accurately predict protein structure from sequence, allowing researchers to explore protein space from the structural viewpoint. In this paper we explore whether "novel" features, such as distinct loop conformations can arise from these predictions despite not being present in the training data. Here we have used ABodyBuilder2, a deep learning antibody structure predictor, to predict the structures of [~]1.5M paired antibody sequences. We examined the predicted structures of the canonical CDR loops and found that most of these predictions fall into the already described CDR canonical form structural space. We also found a small number of "new" canonical clusters composed of heterogeneous sequences united by a common sequence motif and loop conformation. Analysis of these novel clusters showed their origins to be either shapes seen in the training data at very low frequency or shapes seen at high frequency but at a shorter sequence length. To evaluate explicitly the ability of ABodyBuilder2 to extrapolate, we retrained several models whilst with-holding all antibody structures of a specific CDR loop length or canonical form. These "starved" models showed evidence of generalisation across CDRs of different lengths, but they did not extrapolate to loop conformations which were highly distinct from those present in the training data. However, the models were able to accurately predict a canonical form even if only a very small number of examples of that shape were in the training data. Our results suggest that deep learning protein structure prediction methods are unable to make completely out-of-domain predictions for CDR loops. However, in our analysis we also found that even minimal amounts of data of a structural shape allow the method to recover its original predictive abilities. We have made the [~]1.5 M predicted structures used in this study available to download at https://doi.org/10.5281/zenodo.10280181.

bioinformatics↗

bioRxiv2023

The Patent and Literature Antibody Database (PLAbDab): an evolving reference set of functionally diverse, literature-annotated antibody sequences and structures

Antibodies are key proteins of the adaptive immune system, and there exists a large body of academic literature and patents dedicated to their study and concomitant conversion into therapeutics, diagnostics, or reagents. These documents often contain extensive functional characterisations of the sets of antibodies the describe. However, leveraging these heterogeneous reports, for example to offer insights into the properties of query antibodies of interest, is currently challenging as there is no central repository through which this wide corpus can be mined by sequence or structure. Here, we present PLAbDab (the Patent and Literature Antibody Database), a self-updating repository containing over 150,000 paired antibody sequences and 3D structural models, of which over 65,000 are unique. Each entry in the database also contains the title and authors of its literature source. Here we describe the methods used to extract, filter, pair, and model the antibodies in PLAbDab, and showcase how PLAbDab can be searched by sequence, structure, or keyword. PLAbDab uses include annotating query antibodies with potential antigen information from similar entries, analysing structural models of existing antibodies to identify modifications that could improve their properties, and compiling bespoke datasets of antibody sequences/structures known to bind to a specific antigen. PLAbDab is freely available via Github (https://github.com/oxpig/PLAbDab) and as a searchable webserver (https://opig.stats.ox.ac.uk/webapps/plabdab/).

bioinformatics↗

bioRxiv2023

Improved computational epitope profiling using structural models identifies a broader diversity of antibodies that bind the same epitope

The function of an antibody is intrinsically linked to which epitope it engages. Clonal clustering methods, based on sequence identity, are commonly used to group antibodies that will bind the same epitope. However, such methods neglect the fact that antibodies with highly diverse sequences can exhibit similar binding site geometries and engage common epitopes. In a previous study we described SPACE1, a method that structurally clustered antibodies in order to predict their epitopes. This methodology was limited by the inaccuracies and incomplete coverage of template-based modelling. It was also only benchmarked at the level of domain-consistency on one virus class. Here, we present SPACE2, which uses the latest machine learning based structure prediction technology combined with a novel clustering protocol and benchmark it on binding data that has epitope level resolution. On six diverse sets of antigen specific antibodies we demonstrate that SPACE2 accurately clusters antibodies that engage common epitopes and achieves far higher data set coverage than clonal clustering and SPACE1. Furthermore, we show that the functionally consistent structural clusters identified by SPACE2 are even more diverse in sequence, genetic lineage, and species origin than those found by SPACE1. These results reiterate that structural data improves our ability to identify antibodies that bind the same epitope, adding information to sequence-based methods, especially in data sets of antibodies from diverse sources. SPACE2 is openly available on Github (https://github.com/oxpig/SPACE2).

bioinformatics↗

bioRxiv2022

ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins.

Immune receptor proteins play a key role in the immune system and have shown great promise as biotherapeutics. The structure of these proteins is critical for understanding their antigen binding properties. Here, we present ImmuneBuilder, a set of deep learning models trained to accurately predict the structure of antibodies (ABodyBuilder2), nanobodies (NanoBodyBuilder2) and T-Cell receptors (TCRBuilder2). We show that ImmuneBuilder generates structures with state of the art accuracy while being far faster than AlphaFold2. For example, on a benchmark of 34 recently solved antibodies, ABodyBuilder2 predicts CDR-H3 loops with an RMSD of 2.81[A], a 0.09[A] improvement over AlphaFold-Multimer, while being over a hundred times faster. Similar results are also achieved for nanobodies, (NanoBodyBuilder2 predicts CDR-H3 loops with an average RMSD of 2.89[A], a 0.55[A] improvement over AlphaFold2) and TCRs. By predicting an ensemble of structures, ImmuneBuilder also gives an error estimate for every residue in its final prediction. ImmuneBuilder is made freely available, both to download (https://github.com/oxpig/ImmuneBuilder) and to use via our webserver (http://opig.stats.ox.ac.uk/webapps/newsabdab/sabpred). We also make available structural models for ~150 thousand non-redundant paired antibody sequences (https://zenodo.org/record/7258553).

bioinformatics↗

bioRxiv2022

KA-Search: Rapid and exhaustive sequence identity search of known antibodies

Antibodies with similar amino acid sequences, especially across their complementary-determining regions, often share properties. Finding that an antibody of interest has a similar sequence to naturally expressed antibodies in healthy or diseased repertoires is a powerful approach for the prediction of antibody properties, such as immunogenicity or antigen specificity. However, as the number of available antibody sequences is now in the billions and continuing to grow, repertoire mining for similar sequences has become increasingly computationally expensive. Existing approaches are limited by either being low-throughput, non-exhaustive, not antibody specific, or only searching against entire chain sequences. Therefore, there is a need for a specialized tool, optimized for a rapid and exhaustive search of any antibody region against all known antibodies, to better utilize the full breadth of available repertoire sequences. We introduce Known Antibody Search (KA-Search), a tool that allows for the rapid search of billions of antibody sequences by sequence identity across either the whole chain, the complementarity-determining regions, or a user defined antibody region. We show KA-Search in operation on the [~]2.4 billion antibody sequences available in the OAS database. KA-Search can be used to find the most similar sequences from OAS within 30 minutes using 5 CPUs. We give examples of how KA-Search can be used to obtain new insights about an antibody of interest. KA-Search is freely available at https://github.com/oxpig/kasearch.

bioinformatics↗

bioRxiv2021

ABlooper: Fast accurate antibody CDR loop structure prediction with accuracy estimation

Antibodies are a key component of the immune system and have been extensively used as biotherapeutics. Accurate knowledge of their structure is central to understanding their antigen binding function. The key area for antigen binding and the main area of structural variation in antibodies is concentrated in the six complementarity determining regions (CDRs), with the most important for binding and most variable being the CDR-H3 loop. The sequence and structural variability of CDR-H3 make it particularly challenging to model. Recently deep learning methods have offered a step change in our ability to predict protein structures. In this work we present ABlooper, an end-to-end equivariant deep-learning based CDR loop structure prediction tool. ABlooper rapidly predicts the structure of CDR loops with high accuracy and provides a confidence estimate for each of its predictions. On the models of the Rosetta Antibody Benchmark, ABlooper makes predictions with an average CDR-H3 RMSD of 2.49[A], which drops to 2.05[A] when considering only its 76% most confident predictions.

bioinformatics↗