Biology subjects

Reveiz, M.

Publications and source records attributed to Reveiz, M..

4 recordsLinked to original sources

bioRxiv2025

Boltz-2: Towards Accurate and Efficient Binding Affinity Prediction

Accurately modeling biomolecular interactions is a central challenge in modern biology. While recent advances, such as AlphaFold3 and Boltz-1, have substantially improved our ability to predict biomolecular complex structures, these models still fall short in predicting binding affinity, a critical property underlying molecular function and therapeutic efficacy. Here, we present Boltz-2, a new structural biology foundation model that exhibits strong performance for both structure and affinity prediction. Boltz-2 introduces controllability features including experimental method conditioning, distance constraints, and multi-chain template integration for structure prediction, and is, to our knowledge, the first AI model to approach the performance of free-energy perturbation (FEP) methods in estimating small molecule-protein binding affinity. Crucially, it achieves strong correlation with experimental readouts on many benchmarks, while being at least 1000x more computationally efficient than FEP. By coupling Boltz-2 with a generative model for small molecules, we demonstrate an effective workflow to find diverse, synthesizable, high-affinity binders, as estimated by absolute FEP simulations on the TYK2 target. To foster broad adoption and further innovation at the intersection of machine learning and biology, we are releasing Boltz-2 weights, inference, and training code 1 under a permissive open license, providing a robust and extensible foundation for both academic and industrial research.

molecular biology↗

bioRxiv2025

GLYCO-2: a Tool to Quantify Glycan Shielding of Glycosylated Proteins with Improved Data Processing and Computation Speed

MotivationGlycans bound to glycoproteins mediate immune response, including antibody recognition and immune evasion. Previously, we developed an in silico tool GLYCO (GLYcan COverage) to quantify the glycan shielding of a protein surface, applying it to various studies. However, GLYCO lacks sufficient computational efficiency when analyzing larger datasets. ResultsHere we introduce GLYCO-2 which improves the computational speed by [~]4- fold compared to GLYCO by adopting a new analytical cylinder method with k-d trees. GLYCO-2 can calculate glycan shielding from a single coordinate file or from multiple frames derived from molecular dynamics simulations accounting for the inherent flexibility of oligosaccharides. We applied GLYCO-2 to quantify glycan shielding of influenza hemagglutinin (HA) proteins across diverse subtypes that infect humans, revealing an increasing trend in glycan shielding over time within each subtype, likely contributing to immune evasion. Overall, the enhanced computational efficiency of GLYCO-2 allows for faster and easier quantification of glycans, which contributes to the understand of glycan shielding effects in fields such as immunology and vaccine design. Availability and implementationGLYCO-2 is freely available at https://github.com/meteosR/GLYCO-2/ Contactmyungjin.lee@nih.gov or reda.rawi@nih.gov Supplementary informationSupplementary data are available at Bioinformatics online.

bioinformatics↗

bioRxiv2024

Boltz-1: Democratizing Biomolecular Interaction Modeling

Understanding biomolecular interactions is fundamental to advancing fields like drug discovery and protein design. In this paper, we introduce BO_SCPLOWOLTZC_SCPLOW-1, an open-source deep learning model incorporating innovations in model architecture, speed optimization, and data processing achieving AO_SCPLOWLPHAC_SCPLOWFO_SCPLOWOLDC_SCPLOW3-level accuracy in predicting the 3D structures of biomolecular complexes. BO_SCPLOWOLTZC_SCPLOW-1 demonstrates a performance on-par with state-of-the-art commercial models on a range of diverse benchmarks, setting a new benchmark for commercially accessible tools in structural biology. Further, we push the boundary of capabilities of these models with BO_SCPLOWOLTZC_SCPLOWO_SCPCAP-C_SCPCAPO_SCPLOWSTEERINGC_SCPLOW, a new inference time steering technique that is able to fix hallucinations and non-physical predictions from the models. By releasing the training and inference code, model weights, datasets, and benchmarks under the MIT open license, we aim to foster global collaboration, accelerate discoveries, and provide a robust platform for advancing biomolecular modeling.

biophysics↗

bioRxiv2024

Vaccine-elicited and naturally elicited antibodies differ in their recognition of the HIV-1 fusion peptide

Broadly neutralizing antibodies have been proposed as templates for HIV-vaccine design, but it has been unclear how similar vaccine-elicited antibodies are to their naturally elicited templates. To provide insight, here we compare the recognition of naturally elicited and vaccine-elicited antibodies targeting the HIV-1-fusion peptide, which comprises envelope (Env) residues 512-526, with the most common sequence being AVGIGAVFLGFLGAA. Naturally elicited antibodies bound peptides with negative-charge substitutions around residues 517-520 substantially better than the most common sequence, despite these substitutions rarely appearing in HIV; by contrast, vaccine-elicited antibodies were less tolerant of sequence variation, with no substitution of residues 512-516 showing increased binding. Molecular dynamics analysis and cryo-EM structure of the naturally elicited ACS202 antibody in complex with HIV-Env trimer with A517E suggested enhanced binding to result from electrostatic interactions with positively charged antibody residues. Overall, vaccine-elicited antibodies appeared to be more fully optimized to bind the most common fusion peptide sequence. HIGHLIGHTSPeptide substitution scan reveals naturally elicited antibodies against fusion peptide (FP) can bind select non-canonical FP sequences with high affinity. Peptide substitution scan data for FP antibodies correlates significantly with their differential selection indicating variation in binding relates to neutralization tolerance. Structure and energetic analysis of naturally elicited ACS202 with HIV-Env trimer reveals basis for improved recognition of A517E mutant. Atomic level interactions from MD simulation analysis corroborate trends observed with peptide substitutions. verall, peptide substitution scans reveal vaccine-elicited antibodies against FP to be less permissive to FP-sequence variability than naturally elicited antibodies.

immunology↗