PeptiVerse: A Unified Platform for Therapeutic Peptide Property Prediction

Zenodo (CERN European Organization for Nuclear Research) · 2026-05-02

rsync -av --ignore-existing training_classifiers/ /path/to/zenodo_extracted/training_classifiers/ Try to merge with the Huggingface repo at https://huggingface.co/ChatterjeeLab/PeptiVerse.

Minimal-Action Discrete Schrödinger Bridge Matching for Peptide Sequence Design

Open MIND · 2026-01-29

Generative modeling of peptide sequences requires navigating a discrete and highly constrained space in which many intermediate states are chemically implausible or unstable. Existing discrete diffusion and flow-based methods rely on reversing fixed corruption processes or following prescribed probability paths, which can force generation through low-likelihood regions and require countless sampling steps. We introduce Minimal-action discrete Schrödinger Bridge Matching (MadSBM), a rate-based generative framework for peptide design that formulates generation as a controlled continuous-time Markov process on the amino-acid edit graph. To yield probability trajectories that remain near high-likelihood sequence neighborhoods throughout generation, MadSBM 1) defines generation relative to a biologically informed reference process derived from pre-trained protein language model logits and 2) learns a time-dependent control field that biases transition rates to produce low-action transport paths from a masked prior to the data distribution. We finally introduce guidance to the MadSBM sampling procedure towards a specific functional objective, expanding the design space of therapeutic peptides; to our knowledge, this represents the first-ever application of discrete classifier guidance to Schrödinger bridge-based generative models.

PeptiVerse: A Unified Platform for Therapeutic Peptide Property Prediction

Zenodo (CERN European Organization for Nuclear Research) · 2026-05-02

rsync -av --ignore-existing training_classifiers/ /path/to/zenodo_extracted/training_classifiers/ Try to merge with the Huggingface repo at https://huggingface.co/ChatterjeeLab/PeptiVerse.

mRNAutilus Training Data (Pretraining + Classifiers)

Zenodo (CERN European Organization for Nuclear Research) · 2026-04-16

Data used for training models described in Multi-Objective-Guided Generative Design of mRNA with Therapeutic Properties. Directory ensembl-raw.tar.gz ~ Raw FASTA files for pretraining mRNAutilus, separated by species Saluki_hl.xlsx ~ Excel spreadsheet describing sequences and values used to train half-life XGBoost regressor Ribo-NN-human.xlsx ~ Excel spreadsheet describing sequences and values used to train translation efficiency XGBoost regressor protein_abundance.xlsx ~ Excel spreadsheet describing sequences and values used to train protein abundance XGBoost regressor

Branched Schrödinger Bridge Matching.

PubMed · 2026-03-02

, a novel framework that learns branched Schrödinger bridges. BranchSBM parameterizes multiple time-dependent velocity fields and growth processes, enabling the representation of population-level divergence into multiple terminal distributions. We show that BranchSBM is not only more expressive but also essential for tasks involving multi-path surface navigation, modeling cell fate bifurcations from homogeneous progenitor states, and simulating diverging cellular responses to perturbations.

TD3B: Transition-Directed Discrete Diffusion for Allosteric Binder Generation

ArXiv.org · 2026-05-10

Protein function is often controlled by ligands that bias the direction of state transitions, such as agonists and antagonists, rather than stabilizing a single conformation. This is especially important for clinically relevant G protein-coupled receptors (GPCRs), where therapeutic efficacy depends on functional directionality. Structure-based design methods optimize binding to static conformations and cannot represent non-reversible, directional effects or systematically distinguish agonist from antagonist behavior. To address this gap, we introduce Transition-Directed Discrete Diffusion for Allosteric Binder Design (TD3B), a sequence-based generative framework that designs binders with specified agonist or antagonist behavior via a directional transition control objective. TD3B combines a target-aware Direction Oracle, a soft binding-affinity gate, and amortized fine-tuning of a pre-trained discrete diffusion model, enabling targeted agonist and antagonist generation decoupled from binding affinity and unattainable by equilibrium-based or inference-only guidance baselines. The code and checkpoints are available at https://huggingface.co/ChatterjeeLab/TD3B.

PeptiVerse: A Unified Platform for Therapeutic Peptide Property Prediction

bioRxiv (Cold Spring Harbor Laboratory) · 2026-01-03

Abstract Therapeutic peptides combine the advantages of small molecules and antibodies, offering target flexibility and low immunogenicity, yet their successful translation requires careful evaluation of multiple developability properties beyond binding alone. As chemically modified peptides become increasingly common in drug design, no unified platform currently supports systematic property assessment across both canonical sequences and SMILES-based representations. Leveraging the generalizability of large foundational models trained on protein and chemical data, we introduce PeptiVerse , a universal therapeutic peptide property prediction platform. PeptiVerse accepts either amino acid sequences or chemically modified peptide SMILES, delivers state-of-the-art performance across diverse property prediction tasks, and provides both a web interface and open-source implementation for rapid, accessible, and scalable pep-tide developability analysis. By unifying property prediction across representations, PeptiVerse directly supports early-stage peptide therapeutic development campaigns and property-aware generative design workflows.

AptaBLE: A Deep Learning Platform for Aptamer Generation and Analysis

bioRxiv (Cold Spring Harbor Laboratory) · 2026-01-07 · 1 citations

Abstract Aptamers are single-stranded oligonucleotides that bind molecular targets with high affinity and specificity. However, their discovery remains time-consuming, expensive, and susceptible to experimental biases. Here we present AptaBLE, a deep learning framework for predicting aptamer-protein binding. Additionally, we demonstrate two de novo generation methods that produce novel aptamers with desired specificity profiles and K d ’s as low as 31 nM to-date. AptaBLE represents a significant advance towards therapeutic and diagnostic aptamer development.

AI-Designed Peptides as Tools for Biochemistry

Biochemistry · 2026-04-10

Peptides occupy a unique niche as biochemical tools: they are small, modular reagents capable of perturbing protein function with a precision that is often inaccessible to small molecules or antibodies. Historically, their broader use in biochemical research has been constrained by slow discovery workflows, limited control over specificity, and poor physicochemical properties. Recent advances in artificial intelligence have begun to change this landscape by enabling the rational, data-driven design of peptides tailored for specific experimental tasks. In this review, we focus on AI-designed peptides as practical tools for biochemistry. We survey sequence-based and structure-based design paradigms, highlighting how each supports distinct classes of peptide tools, including isoform- and motif-specific binders, multi-objective assay-ready reagents, and functional peptides that enable degradation, stabilization, or biophysical interrogation of proteins. By emphasizing experimental utility, design constraints, and appropriate use cases, we aim to provide a framework for selecting and deploying AI-designed peptides as on-demand reagents in modern biochemical research.

mRNAutilus Training Data (Pretraining + Classifiers)

Zenodo (CERN European Organization for Nuclear Research) · 2026-04-16

Data used for training models described in Multi-Objective-Guided Generative Design of mRNA with Therapeutic Properties. Directory ensembl-raw.tar.gz ~ Raw FASTA files for pretraining mRNAutilus, separated by species Saluki_hl.xlsx ~ Excel spreadsheet describing sequences and values used to train half-life XGBoost regressor Ribo-NN-human.xlsx ~ Excel spreadsheet describing sequences and values used to train translation efficiency XGBoost regressor protein_abundance.xlsx ~ Excel spreadsheet describing sequences and values used to train protein abundance XGBoost regressor