Fergus Boyles

The antibody repertoires of individuals and groups have been used to explore disease states, understand vaccine responses, and drive therapeutic development. The arrival of B-cell receptor repertoire sequencing has enabled researchers to get a snapshot of these antibody repertoires, and as more data are generated, increasingly in-depth studies are possible. However, most publicly available data only exist as raw FASTQ files, making the data hard to access, process, and compare. The Observed Antibody Space (OAS) database was created in 2018 to offer clean, annotated, and translated repertoire data. In this paper, we describe an update to OAS that has been driven by the increasing volume of data and the appearance of paired (VH/VL) sequence data. OAS is now accessible via a new web server, with standardized search parameters and a new sequence-based search option. The new database provides both nucleotides and amino acids for every sequence, with additional sequence annotations to make the data Minimal Information about Adaptive Immune Receptor Repertoire compliant, and comments on potential problems with the sequence. OAS now contains 25 new studies, including severe acute respiratory syndrome coronavirus 2 data and paired sequencing data. The new database is accessible at http://opig.stats.ox.ac.uk/webapps/oas/, and all data are freely available for download.

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 2.81Å, a 0.09Å 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 2.89Å, a 0.55Å 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://doi.org/10.5281/zenodo.7258553 ).

MOTIVATION: Machine learning scoring functions for protein-ligand binding affinity prediction have been found to consistently outperform classical scoring functions. Structure-based scoring functions for universal affinity prediction typically use features describing interactions derived from the protein-ligand complex, with limited information about the chemical or topological properties of the ligand itself. RESULTS: We demonstrate that the performance of machine learning scoring functions are consistently improved by the inclusion of diverse ligand-based features. For example, a Random Forest (RF) combining the features of RF-Score v3 with RDKit molecular descriptors achieved Pearson correlation coefficients of up to 0.836, 0.780 and 0.821 on the PDBbind 2007, 2013 and 2016 core sets, respectively, compared to 0.790, 0.746 and 0.814 when using the features of RF-Score v3 alone. Excluding proteins and/or ligands that are similar to those in the test sets from the training set has a significant effect on scoring function performance, but does not remove the predictive power of ligand-based features. Furthermore a RF using only ligand-based features is predictive at a level similar to classical scoring functions and it appears to be predicting the mean binding affinity of a ligand for its protein targets. AVAILABILITY AND IMPLEMENTATION: Data and code to reproduce all the results are freely available at http://opig.stats.ox.ac.uk/resources. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract 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 2.81Å, a 0.09Å 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 2.89Å, a 0.55Å 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 ).

Abstract Graph Neural Networks (GNNs) have recently gained in popularity, challenging molecular fingerprints or SMILES-based representations as the predominant way to represent molecules for binding affinity prediction. Although simple ligand-based graphs alone are already useful for affinity prediction, better performance on multi-target datasets has been achieved with models that incorporate 3D structural information. Most recent advances utilize complex GNN architectures to capture 3D protein-ligand information by incorporating ligand-interacting protein atoms as additional nodes in the graphs; or by building a second protein-based graph in parallel. This expands the graph considerably while obfuscating the shape of the underlying ligand, diminishing the advantage that GNNs have when encoding molecular structures. There is therefore a need for a simple and elegant molecular graph representation that retains the topology of the ligand while simultaneously encoding 3D protein-ligand interactions. We present Protein-Ligand Interaction Graphs (PLIGs): a simple way of representing atom-atom contacts of 3D protein-ligand complexes as node features for GNNs. PLIGs featurize an atom node in the molecular graph by describing each atom’s properties as well as all atom-atom contacts made with protein atoms within a distance threshold. The edges of the graph are therefore identical to ligand-based graphs, but the nodes encode the 3D protein-ligand contacts. Since PLIGs are applicable to any GNN architecture, we have benchmarked their performance with six different GNN architectures, and compared them to conventional ligand-based graphs and fingerprint-based multi-layer perceptron (MLP) models using the CASF-2016 benchmark set where we found PLIG-based Graph Attention Networks (GATNet) to be the best performing model ( ρ =0.84, RMSE=1.22 pK). In summary, we created a novel graph-based representation that incorporates 3D structural information into the node features of ligand-shaped molecular graphs. The PLIG representation is simple, elegant, flexible and easily customizable, opening up many possibilities of incorporating other 2D and 3D properties into the graph. Access The code and implementation for PLIGs and all models can be found at github.com/MarcMoesser/Protein-Ligand-Interaction-Graphs .