Biology subjects

Visani, G. M.

Publications and source records attributed to Visani, G. M..

6 recordsLinked to original sources

bioRxiv2026

Additive baselines furnish no evidence for epistasis learning by MULTI-evolve

Recent work from Tran et al. (Science, 2026) introduced MULTI-evolve, a framework for protein engineering that combines single-mutant nomination via a protein language model (PLM) or a deep mutational scan (DMS), experimental single- and double-mutant characterization, and neural networks to engineer hyperactive multimutant proteins. The authors attribute the frameworks performance to "epistasis-aware modeling" and claim that their neural networks "learn the epistatic landscape" and "identify synergistic interactions" from limited double-mutant training data. Additive models, by definition, cannot represent epistasis, making them a natural null baseline for such claims. Here we show that MULTI-evolves multimutant predictions are almost perfectly correlated with an additive models across all three engineering applications (APEX, dCasRx, and HuABC2), such that the engineering of multimutants reduces to combining beneficial mutations with the largest additive effects--a standard protein engineering strategy for over four decades. We also find that MULTI-evolves neural networks do not outperform an additive model on held-out test set predictions, and do not even represent epistasis in their training data. Finally, we revisit a DMS benchmark finding presented as evidence of epistasis learning and show that the same pattern is expected even under a null additive model, due to an elementary statistical phenomenon; when we fit an additive model to the benchmark data, it reproduces the reported pattern. More broadly, our findings underscore the need to benchmark models for machine learning-guided directed evolution against additive null baselines before attributing performance to learned epistasis.

bioinformatics↗

bioRxiv2026

IMMREP25: Unseen Peptides

T cell receptors (TCRs) can bind to peptides presented by MHC molecules (pMHC) as a first step to trigger a T cell response. Reliable approaches to predict TCR:pMHC binding would have broad applications in clinical diagnostics, therapeutics, and the fundamental understanding of molecular interactions. IMMREP is a community organized series of prediction contests that asks participants to predict TCR:pMHC binding on unpublished datasets. Previous iterations in 2022 and 2023 showed multiple approaches can predict TCR-pMHC binding with significant accuracy (median AUC_0.1[≥]0.7) for peptides where experimental data is available ("seen" peptides). In contrast, models did not outperform random guessing for peptides that have no such data available ("unseen" peptides). Here we report on the results of IMMREP25, which focused solely on unseen peptides in order to evaluate the cutting edge of the field. We received 126 named submissions predicting the specificity of 1,000 TCRs against twenty unseen peptides restricted by one of two MHC molecules (HLA-A*02:01 and HLA-B*40:01). The best performing methods showed a macro-AUC_0.1 of 0.60, significantly better than random, demonstrating significant advances in the field. The top performing methods incorporated structural modeling into their approach, indicating that especially for unseen peptides, a structural understanding aids in the prediction of TCR:pMHC interactions. The results from this benchmark highlight the significant challenges remaining for TCR:pMHC predictions and will inform future method development.

bioinformatics↗

bioRxiv2025

T-cell receptor specificity landscape revealed through de novo peptide design

T-cells play a key role in adaptive immunity by mounting specific responses against diverse pathogens. An effective binding between T-cell receptors (TCRs) and pathogen-derived peptides presented on Major Histocompatibility Complexes (MHCs) mediate an immune response. However, predicting these interactions remains challenging due to limited functional data on T-cell reactivities. Here, we introduce a computational approach to predict TCR interactions with peptides presented on MHC class I alleles, and to design novel immunogenic peptides for specified TCR-MHC complexes. Our method leverages HERMES, a structure-based, physics-guided machine learning model trained on the protein universe to predict amino acid preferences based on local structural environments. Despite no direct training on TCR-pMHC data, the implicit physical reasoning in HERMES enables us to make accurate predictions of both TCR-pMHC binding affinities and T-cell activities across diverse viral epitopes and cancer neoantigens, achieving up to 0.72 correlation with experimental data. Leveraging our TCR recognition model, we develop a computational protocol for de novo design of immunogenic peptides. Through experimental validation in three TCR-MHC systems targeting viral and cancer peptides, we demonstrate that our designs--with up to five substitutions from the native sequence--activate T-cells at success rates of up to 50%. Lastly, we use our generative framework to quantify the diversity of the peptide recognition landscape for various TCR-MHC complexes, offering key insights into T-cell specificity in both humans and mice. Our approach provides a platform for immunogenic peptide and neoantigen design, as well as for evaluating TCR specificity, offering a computational framework to inform design of engineered T-cell therapies and vaccines.

biophysics↗

bioRxiv2024

Design of high specificity binders for peptide-MHC-I complexes

Class I MHC molecules present peptides derived from intracellular antigens on the cell surface for immune surveillance, and specific targeting of these peptide-MHC (pMHC) complexes could have considerable utility for treating diseases. Such targeting is challenging as it requires readout of the few outward facing peptide antigen residues and the avoidance of extensive contacts with the MHC carrier which is present on almost all cells. Here we describe the use of deep learning-based protein design tools to de novo design small proteins that arc above the peptide binding groove of pMHC complexes and make extensive contacts with the peptide. We identify specific binders for ten target pMHCs which when displayed on yeast bind the on-target pMHC tetramer but not closely related peptides. For five targets, incorporation of designs into chimeric antigen receptors leads to T-cell activation by the cognate pMHC complexes well above the background from complexes with peptides derived from proteome. Our approach can generate high specificity binders starting from either experimental or predicted structures of the target pMHC complexes, and should be widely useful for both protein and cell based pMHC targeting.

immunology↗

bioRxiv2024

HERMES: Holographic Equivariant neuRal network model for Mutational Effect and Stability prediction

Accurately predicting how amino acid substitutions alter protein function is a central challenge in biology, with applications from interpreting disease variants to designing vaccines and therapeutics. We introduce HERMES, a family of fast, structure-based models that predict mutational effects from the local atomic environment around each residue. Pre-trained on masked amino acid prediction, HERMES shows strong zero-shot performance for predicting changes in thermodynamic stability and protein-protein binding affinity. Analyzing its predictions, we uncover a pre-training bias toward size-conserving substitutions, which we reduce through an amortized fine-tuning strategy that incorporates packing flexibility. When fine-tuned on experimental data, HERMES matches state-of-the-art stability predictors without costly data augmentation. HERMES also identifies antigen-stabilizing mutations across multiple viral envelope proteins, enabling an efficient pipeline for designing mutation libraries for vaccine development. Together, this work establishes HERMES as a fast and practical structure-based framework for mutation screening and offers insight into the mechanisms underlying its predictions.

bioinformatics↗

bioRxiv2022

Holographic-(V)AE: an end-to-end SO(3)-Equivariant (Variational) Autoencoder in Fourier Space

Group-equivariant neural networks have emerged as a data-efficient approach to solve classification and regression tasks, while respecting the relevant symmetries of the data. However, little work has been done to extend this paradigm to the unsupervised and generative domains. Here, we present Holographic-(Variational) Auto Encoder (H-(V)AE), a fully end-to-end SO(3)-equivariant (variational) autoencoder in Fourier space, suitable for unsupervised learning and generation of data distributed around a specified origin in 3D. H-(V)AE is trained to reconstruct the spherical Fourier encoding of data, learning in the process a low-dimensional representation of the data (i.e., a latent space) with a maximally informative rotationally invariant embedding alongside an equivariant frame describing the orientation of the data. We extensively test the performance of H-(V)AE on diverse datasets. We show that the learned latent space efficiently encodes the categorical features of spherical images. Moreover, H-(V)AEs latent space can be used to extract compact embeddings for protein structure microenvironments, and when paired with a Random Forest Regressor, it enables state-of-the-art predictions of protein-ligand binding affinity.

bioinformatics↗