Biology subjects

Sperandio, O.

Publications and source records attributed to Sperandio, O..

5 recordsLinked to original sources

bioRxiv2026

GTA-5: A Unified Graph Transformer Framework for Ligands and Protein Binding Sites - Part I: Constructing the PDB Pocket and Ligand Space

Structural recognition between a protein target and a ligand underpins therapeutic innovation, yet computational representations of protein binding sites and small molecules remain largely disjoint. Here we introduce GTA-5, a unified graph transformer auto-encoder framework designed to capture the geometric structure and chemical composition of ligands and protein binding pockets, embedding them into multidimensional latent spaces where proximity reflects functional compatibility. Ligands and pockets are represented as three-dimensional point clouds annotated with Tripos atom type labels, omitting explicit bond connectivity to enable structural reasoning based on spatial context rather than predefined connectivity graphs. By not enforcing bond topology, GTA-5 maintains representational flexibility across molecular modalities while preserving chemically meaningful local environments. The model was trained on a curated dataset from the Protein Data Bank comprising 64,124 liganded pockets and 23,133 unique ligands spanning 2,257 protein families. We find that functional protein families cluster coherently in both pocket and ligand latent spaces while retaining biologically meaningful heterogeneity. The model captures physicochemical pocket properties such as volume, exposure, and hydrophobicity directly from raw structural data, while ligands with distinct scaffolds co-localise when occupying similar binding environments. This provides a basis for several downstream applications including scaffold hopping in ligand-based virtual screening, QSAR/QSPR modelling using embedding-derived descriptors, and drug repurposing via pocket similarity. More broadly, the GTA-5 framework establishes a foundation for structural reasoning across molecular modalities in drug discovery.

bioinformatics↗

bioRxiv2026

Structure-guided generative design of peptides targeting the FtsQBL divisome complex inhibit Escherichia coli cell division.

The discovery of antibiotics targeting Gram-negative bacteria remains limited in part by the difficulty of pharmacologically modulating protein-protein interactions essential for bacterial physiology. The divisome complex formed by FtsQ, FtsB and FtsL represents an attractive but challenging target, as its assembly relies on {beta}-strand-mediated interface interactions within the bacterial periplasm. Here, we combined interpretable interface mapping using InDeep with hotspot-constrained RFdiffusion design to generate peptides targeting the FtsB-binding site of Escherichia coli FtsQ. The designed peptides mimic the native {beta}-augmentation interaction and selectively engage the FtsQ interface both in bacterial cells and in vitro. X-ray crystallography of one of these peptides in complex with FtsQ reveals that it accurately adopts the native binding geometry while introducing additional stabilizing interactions within a hydrophobic pocket. Complementary NMR analyses further show that optimized peptides adopt pre-organized {beta}-hairpin conformations in solution consistent with the bound state. Several of these peptides disrupt bacterial cell division and inhibit growth in an E. coli strain exhibiting increased outer membrane permeability. Together, these results establish a structurally validated framework in which predictive interface analysis and generative design can be combined to target cooperative protein-protein interfaces in bacteria and provide a foundation for the development of divisome-targeting antibacterial strategies.

microbiology↗

bioRxiv2026

A pocket-centric framework for selective targeting of amyloid fibril polymorphs

The rapid expansion of high-resolution cryo-EM structures of amyloid fibrils has transformed our understanding of fibril polymorphism, yet it has not been matched by comparable progress in the rational development of protein-selective or polymorph-specific amyloid ligands. One possible explanation is that ligand selectivity is governed not only by global fibril folds, but also by local surface pockets accessible to small molecules. Here, we present a systematic analysis of 400 cryo-EM structures of amyloid-{beta}, tau, and -synuclein fibrils. Using a unified pocket similarity index and minimum spanning tree representations, we construct global and protein-specific graph representations of the amyloid binding pocket space, and examine how surface cavities are distributed across proteins, polymorphs, and structural contexts. We find that many detectable pockets are shared across multiple fibrillar folds and, in several cases, across distinct amyloid-forming proteins, suggesting that pocket-level convergence may contribute to the limited selectivity of amyloid-directed ligands. Conversely, only a restricted subset of pockets occupies isolated regions of pocket similarity space, defining rare structural opportunities for protein-selective or polymorph-restricted targeting. Analysis of structures of extracted fibrils further shows that disease-derived fibril pockets do not form a completely isolated pocketome subset, but can resemble pockets observed in selected in vitro polymorphs. Together, these results reframe amyloid ligand development as a problem of pocket-level discriminability within a constrained fibril landscape, and provide a structural framework for identifying promising binding sites while avoiding intrinsically non-discriminatory pockets. Significance StatementDespite major advances in cryo-EM structure determination of amyloid fibrils, the development of selective ligands for amyloid assemblies remains challenging. By systematically comparing surface binding pockets across 400 amyloid-{beta}, tau, and -synuclein fibrillar structures, we show that many ligand-accessible cavities exhibit similar geometric and physicochemical properties across fibrillar polymorphs made of distinct proteins. This pocket-level convergence provides a structural basis for understanding why many amyloid ligands exhibit broad binding profiles, while also identifying rare pockets that are sufficiently isolated to support more selective targeting strategies. Our work establishes a pocket-centric framework for interpreting amyloid ligand selectivity and for prioritizing fibril binding sites in imaging and therapeutic ligand development.

bioinformatics↗

bioRxiv2021

InDeep : 3D fully convolutional neural networks to assist in silico drug design on protein-protein interactions

AO_SCPLOWBSTRACTC_SCPLOWO_ST_ABSMotivationC_ST_ABSProtein-protein interactions (PPIs) are key elements in numerous biological pathways and the subject of a growing number of drug discovery projects including against infectious diseases. Designing drugs on PPI targets remains a difficult task and requires extensive efforts to qualify a given interaction as an eligible target. To this end, besides the evident need to determine the role of PPIs in disease-associated pathways and their experimental characterization as therapeutics targets, prediction of their capacity to be bound by other protein partners or modulated by future drugs is of primary importance. ResultsWe present InDeep, a tool for predicting functional binding sites within proteins that could either host protein epitopes or future drugs. Leveraging deep learning on a curated data set of PPIs, this tool can proceed to enhanced functional binding site predictions either on experimental structures or along molecular dynamics trajectories. The benchmark of InDeep demonstrates that our tool outperforms state of the art ligandable binding sites predictors when assessing PPI targets but also conventional targets. This offers new opportunities to assist drug design projects on PPIs by identifying pertinent binding pockets at or in the vicinity of PPI interfaces. AvailabilityThe tool is available on GitHub3 along with a PyMol plugin for visualization. Predictions of InDeep can be consulted at iPPI-DB4

bioinformatics↗

bioRxiv2021

Characterization of the interaction domains between the phosphoprotein and the nucleocapsid of human Metapneumovirus

Human metapneumovirus (HMPV) causes severe respiratory diseases in young children. The HMPV RNA genome is encapsidated by the viral nucleoprotein, forming an RNA-N complex (NNuc), which serves as template for genome replication and mRNA transcription by the RNA-dependent RNA polymerase (RdRp). The RdRp is formed by the association of the large polymerase subunit (L), which has RNA polymerase, capping and methyltransferase activities, and the tetrameric phosphoprotein (P). P plays a central role in the RdRp complex by binding to NNuc and L, allowing the attachment of the L polymerase to the nucleocapsid template. During infection these proteins concentrate in cytoplasmic inclusion bodies (IBs) where viral RNA synthesis occurs. By analogy to the closely related pneumovirus respiratory syncytial virus (RSV), it is likely that the formation of IBs depends on the interaction between P and NNuc. However, the HMPV P-NNuc interaction still remains to characterize. Here, we finely characterized the binding domains involved in HMPV P and NNuc interaction by studying binding between recombinant proteins, combined with the use of a functional assay of the polymerase complex activity and the study of the recruitment of these proteins to IBs by immunofluorescence. We show that the last 6 C-terminal residues of HMPV P are necessary and sufficient for binding to NNuc, that P binds the N-terminal domain of N (NNTD), and identified conserved N residues critical for the interaction. Our results allowed to propose a structural model of the HMPV P-NNuc interaction. IMPORTANCELike respiratory syncytial virus (RSV), human metapneumovirus (HMPV) is a leading cause of severe respiratory infections in children but also affects human populations of all ages worldwide. Nowadays, no vaccine or efficient antiviral treatments are available for these two pneumoviruses. A better understanding of the molecular mechanisms involved in viral replication could help the design or discovery of specific antiviral compounds. In this work we have investigated the interaction between two major viral proteins involved in HMPV RNA synthesis, the N and P proteins. We finely characterized their domains of interaction, in particular a pocket on the surface of the N protein that could be used as a template for the design of compounds interfering with N-P complexes and viral replication.

microbiology↗