arXiv subjects

Cesar de la Fuente-Nunez

Publications and source records attributed to Cesar de la Fuente-Nunez.

14 recordsLinked to original sources

arXiv2026

OmegAMP: Targeted AMP Discovery via Biologically Informed Generation

Deep learning-based antimicrobial peptide (AMP) discovery faces critical challenges such as limited controllability, lack of representations that efficiently model antimicrobial properties, and low experimental hit rates. To address these challenges, we introduce OmegAMP, a framework designed for reliable AMP generation with increased controllability. Its diffusion-based generative model leverages a novel conditioning mechanism to achieve fine-grained control over desired physicochemical properties and to direct generation towards specific activity profiles, including species-specific effectiveness. This is further enhanced by a biologically informed encoding space that significantly improves overall generative performance. Complementing these generative capabilities, OmegAMP leverages a novel synthetic data augmentation strategy to train classifiers for AMP filtering, drastically reducing false positive rates and thereby increasing the likelihood of experimental success. Our in silico experiments demonstrate that OmegAMP delivers state-of-the-art performance across key stages of the AMP discovery pipeline, enabling us to achieve an unprecedented success rate in wet lab experiments. We tested 25 candidate peptides, 24 of them (96%) demonstrated antimicrobial activity, proving effective even against multi-drug resistant strains. Our findings underscore OmegAMP's potential to significantly advance computational frameworks in the fight against antimicrobial resistance.

cs.LG↗

arXiv2026

Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

Manually curated biomedical repositories -- spanning bioactivity, genomics, and chemistry -- are expensive to maintain, lag behind primary literature, and discard experimental context, obscuring nuances needed to assess data correctness and coverage. We show that PubMed itself can be autonomously and cost-effectively turned into structured datasets that are larger, more nuanced, and more accurate than the curated databases they replace. We present three coupled contributions: (1) an LLM-based entity-tagging pipeline, grounded in nine biomedical ontologies, that tags 4.5B entities across 19 categories in a 22.5M-paper, 2.5T-token PubMed corpus; (2) hybrid sparse-dense retrieval supporting entity-filtered semantic queries over the tagged corpus; and (3) Starling, a multi-agent deep research system that, given only a natural-language task description, designs precision- and recall-targeted retrieval filters, induces an extraction schema, and emits structured records with nuance-rich fields and supporting passages. Across six tasks -- blood-brain barrier permeability, oral bioavailability, acute toxicity (LD50), gene-disease associations, protein subcellular localization, and chemical reactions -- Starling produces ~6.3M records (91K-3M per task); several are, to our knowledge, the largest public datasets for their property. Frontier-model rejection of our extractions is 0.6-7.7% across tasks, far below error rates we measure on widely used curated counterparts (e.g., 16.5% on BBB_Martins, 7.3% on Bioavailability_Ma). Beyond scale and accuracy, the supporting passages carry nuance tabular databases discard -- e.g., oral bioavailability may depend on fed vs. fasted state. Together, the corpus, retrieval, and agent establish a foundation for AI-driven therapeutic design. Code and datasets: https://github.com/starling-labs/starling.

cs.LG↗

arXiv2026

Purely Agent-Driven Black-Box Optimization for Biological Design

Many key challenges in biological design -- such as small-molecule drug discovery, antimicrobial peptide development, and protein engineering -- can be framed as black-box optimization over vast, complex structured spaces. Existing methods rely mainly on raw structural data and struggle to exploit the rich scientific literature. While large language models (LLMs) have been added to these pipelines, they have been confined to narrow roles within structure-centered optimizers. We instead cast biological black-box optimization as an agent-driven, language-based reasoning process. We introduce Purely Agent-driven BLack-box Optimization (PABLO), a hierarchical agentic system that uses scientific LLMs pretrained on chemistry and biology literature to generate and iteratively refine biological candidates. On both the standard GuacaMol molecular design and antimicrobial peptide optimization tasks, PABLO achieves state-of-the-art performance, substantially improving sample efficiency and final objective values over established baselines. Compared to prior optimization methods that incorporate LLMs, PABLO achieves competitive token usage per run despite relying on LLMs throughout the optimization loop. Beyond raw performance, the agentic formulation offers key advantages for realistic design: it naturally incorporates semantic task descriptions, retrieval-augmented domain knowledge, and complex constraints. In follow-up in vitro validation, PABLO-optimized peptides showed strong activity against drug-resistant pathogens, underscoring the practical potential of PABLO for therapeutic discovery.

cs.LG↗

arXiv2026

PepCompass: Navigating peptide embedding spaces using Riemannian Geometry

Antimicrobial peptide discovery is challenged by the astronomical size of peptide space and the relative scarcity of active peptides. Generative models provide continuous latent "maps" of peptide space, but conventionally ignore decoder-induced geometry and rely on flat Euclidean metrics, rendering exploration and optimization distorted and inefficient. Prior manifold-based remedies assume fixed intrinsic dimensionality, which critically fails in practice for peptide data. Here, we introduce PepCompass, a geometry-aware framework for peptide exploration and optimization. At its core, we define a Union of $κ$-Stable Riemannian Manifolds $\mathbb{M}^κ$, a family of decoder-induced manifolds that captures local geometry while ensuring computational stability. We propose two local exploration methods: Second-Order Riemannian Brownian Efficient Sampling, which provides a convergent second-order approximation to Riemannian Brownian motion, and Mutation Enumeration in Tangent Space, which reinterprets tangent directions as discrete amino-acid substitutions. Combining these yields Local Enumeration Bayesian Optimization (LE-BO), an efficient algorithm for local activity optimization. Finally, we introduce Potential-minimizing Geodesic Search (PoGS), which interpolates between prototype embeddings along property-enriched geodesics, biasing discovery toward seeds, i.e. peptides with favorable activity. In-vitro validation confirms the effectiveness of PepCompass: PoGS yields four novel seeds, and subsequent optimization with LE-BO discovers 25 highly active peptides with broad-spectrum activity, including against resistant bacterial strains. These results demonstrate that geometry-informed exploration provides a powerful new paradigm for antimicrobial peptide design.

cs.LG↗

arXiv2025

ApexGen: Simultaneous design of peptide binder sequence and structure for target proteins

Peptide-based drugs can bind to protein interaction sites that small molecules often cannot, and are easier to produce than large protein drugs. However, designing effective peptide binders is difficult. A typical peptide has an enormous number of possible sequences, and only a few of these will fold into the right 3D shape to match a given protein target. Existing computational methods either generate many candidate sequences without considering how they will fold, or build peptide backbones and then find suitable sequences afterward. Here we introduce ApexGen, a new AI-based framework that simultaneously designs a peptide's amino-acid sequence and its three-dimensional structure to fit a given protein target. For each target, ApexGen produces a full all-atom peptide model in a small number of deterministic integration steps. In tests on hundreds of protein targets, the peptides designed by ApexGen fit tightly onto their target surfaces and cover nearly the entire binding site. These peptides have shapes similar to those found in natural protein-peptide complexes, and they show strong predicted binding affinity in computational experiments. Because ApexGen couples sequence and structure design at every step of Euler integration within a flow-matching sampler, it is much faster and more efficient than prior approaches. This unified method could greatly accelerate the discovery of new peptide-based therapeutics.

q-bio.BM↗

arXiv2025

Hyperbolic Graph Embeddings Reveal the Host-Pathogen Interactome

Infections depend on interactions between pathogen and host proteins, but comprehensively mapping these interactions is challenging and labor intensive. Many biological networks have hierarchical, scale-free structure, so we developed a deep learning framework, ApexPPI, that represents protein networks in hyperbolic Riemannian space to capture these features. Our model integrates multimodal biological data (protein sequences, gene perturbation experiments, and complementary interaction networks) to predict likely interactions between pathogen and host proteins through multi-task hyperbolic graph neural networks. Mapping protein features into hyperbolic space led to much higher accuracy than previous methods in predicting host-pathogen interactions. From tens of millions of possible protein pairs, our model identified thousands of high-confidence interactions, including many involving human G-protein-coupled receptors (GPCRs). We validated dozens of these predicted complexes using AlphaFold 3 structural modeling, supporting the accuracy of our predictions. This comprehensive map of host-pathogen protein interactions provides a resource for discovering new treatments and illustrates how advanced AI can unravel complex biological systems.

q-bio.MN↗

arXiv2025

Covering Multiple Objectives with a Small Set of Solutions Using Bayesian Optimization

In multi-objective black-box optimization, the goal is typically to find solutions that optimize a set of $T$ black-box objective functions, $f_1, \ldots f_T$, simultaneously. Traditional approaches often seek a single Pareto-optimal set that balances trade-offs among all objectives. In contrast, we consider a problem setting that departs from this paradigm: finding a small set of $K < T$ solutions, that collectively "cover" the $T$ objectives. A set of solutions is defined as "covering" if, for each objective $f_1, \ldots f_T$, there is at least one good solution. A motivating example for this problem setting occurs in drug design. For example, we may have $T$ pathogens and aim to identify a set of $K < T$ antibiotics such that at least one antibiotic can be used to treat each pathogen. This problem, known as coverage optimization, has yet to be tackled with the Bayesian optimization (BO) framework. To fill this void, we develop Multi-Objective Coverage Bayesian Optimization (MOCOBO), a BO algorithm for solving coverage optimization. Our approach is based on a new acquisition function reminiscent of expected improvement in the vanilla BO setup. We demonstrate the performance of our method on high-dimensional black-box optimization tasks, including applications in peptide and molecular design. Results show that the coverage of the $K < T$ solutions found by MOCOBO matches or nearly matches the coverage of $T$ solutions obtained by optimizing each objective individually. Furthermore, in in vitro experiments, the peptides found by MOCOBO exhibited high potency against drug-resistant pathogens, further demonstrating the potential of MOCOBO for drug discovery. All of our code is publicly available at the following link: https://github.com/nataliemaus/mocobo.

cs.LG↗

arXiv2025

From Supervision to Exploration: What Does Protein Language Model Learn During Reinforcement Learning?

Protein language models (PLMs) have advanced computational protein science through large-scale pretraining and scalable architectures. In parallel, reinforcement learning (RL) has broadened exploration and enabled precise multi-objective optimization in protein design. Yet whether RL can push PLMs beyond their pretraining priors to uncover latent sequence-structure-function rules remains unclear. We address this by pairing RL with PLMs across four domains: antimicrobial peptide design, kinase variant optimization, antibody engineering, and inverse folding. Using diverse RL algorithms and model classes, we ask if RL improves sampling efficiency and, more importantly, if it reveals capabilities not captured by supervised learning. Across benchmarks, RL consistently boosts success rates and sample efficiency. Performance follows a three-factor interaction: task headroom, reward fidelity, and policy capacity jointly determine gains. When rewards are accurate and informative, policies have sufficient capacity, and tasks leave room beyond supervised baselines, improvements scale; when rewards are noisy or capacity is constrained, gains saturate despite exploration. This view yields practical guidance for RL in protein design: prioritize reward modeling and calibration before scaling policy size, match algorithm and regularization strength to task difficulty, and allocate capacity where marginal gains are largest. Implementation is available at https://github.com/chq1155/RL-PLM.

cs.LG↗

arXiv2025

Lightweight MSA Design Advances Protein Folding From Evolutionary Embeddings

Protein structure prediction often hinges on multiple sequence alignments (MSAs), which underperform on low-homology and orphan proteins. We introduce PLAME, a lightweight MSA design framework that leverages evolutionary embeddings from pretrained protein language models to generate MSAs that better support downstream folding. PLAME couples these embeddings with a conservation--diversity loss that balances agreement on conserved positions with coverage of plausible sequence variation. Beyond generation, we develop (i) an MSA selection strategy to filter high-quality candidates and (ii) a sequence-quality metric that is complementary to depth-based measures and predictive of folding gains. On AlphaFold2 low-homology/orphan benchmarks, PLAME delivers state-of-the-art improvements in structure accuracy (e.g., lDDT/TM-score), with consistent gains when paired with AlphaFold3. Ablations isolate the benefits of the selection strategy, and case studies elucidate how MSA characteristics shape AlphaFold confidence and error modes. Finally, we show PLAME functions as a lightweight adapter, enabling ESMFold to approach AlphaFold2-level accuracy while retaining ESMFold-like inference speed. PLAME thus provides a practical path to high-quality folding for proteins lacking strong evolutionary neighbors.

cs.LG↗

arXiv2025

A deep reinforcement learning platform for antibiotic discovery

Antimicrobial resistance (AMR) is projected to cause up to 10 million deaths annually by 2050, underscoring the urgent need for new antibiotics. Here we present ApexAmphion, a deep-learning framework for de novo design of antibiotics that couples a 6.4-billion-parameter protein language model with reinforcement learning. The model is first fine-tuned on curated peptide data to capture antimicrobial sequence regularities, then optimised with proximal policy optimization against a composite reward that combines predictions from a learned minimum inhibitory concentration (MIC) classifier with differentiable physicochemical objectives. In vitro evaluation of 100 designed peptides showed low MIC values (nanomolar range in some cases) for all candidates (100% hit rate). Moreover, 99 our of 100 compounds exhibited broad-spectrum antimicrobial activity against at least two clinically relevant bacteria. The lead molecules killed bacteria primarily by potently targeting the cytoplasmic membrane. By unifying generation, scoring and multi-objective optimization with deep reinforcement learning in a single pipeline, our approach rapidly produces diverse, potent candidates, offering a scalable route to peptide antibiotics and a platform for iterative steering toward potency and developability within hours.

cs.LG↗

arXiv2025

A Dataset for Distilling Knowledge Priors from Literature for Therapeutic Design

AI-driven discovery can greatly reduce design time and enhance new therapeutics' effectiveness. Models using simulators explore broad design spaces but risk violating implicit constraints due to a lack of experimental priors. For example, in a new analysis we performed on a diverse set of models on the GuacaMol benchmark using supervised classifiers, over 60\% of molecules proposed had high probability of being mutagenic. In this work, we introduce Medex, a dataset of priors for design problems extracted from literature describing compounds used in lab settings. It is constructed with LLM pipelines for discovering therapeutic entities in relevant paragraphs and summarizing information in concise fair-use facts. Medex consists of 32.3 million pairs of natural language facts, and appropriate entity representations (i.e. SMILES or refseq IDs). To demonstrate the potential of the data, we train LLM, CLIP, and LLava architectures to reason jointly about text and design targets and evaluate on tasks from the Therapeutic Data Commons (TDC). Medex is highly effective for creating models with strong priors: in supervised prediction problems that use our data as pretraining, our best models with 15M learnable parameters outperform larger 2B TxGemma on both regression and classification TDC tasks, and perform comparably to 9B models on average. Models built with Medex can be used as constraints while optimizing for novel molecules in GuacaMol, resulting in proposals that are safer and nearly as effective. We release our dataset at https://huggingface.co/datasets/medexanon/Medex, and will provide expanded versions as available literature grows.

cs.LG↗

arXiv2025

Predicting and generating antibiotics against future pathogens with ApexOracle

Antimicrobial resistance (AMR) is escalating and outpacing current antibiotic development. Thus, discovering antibiotics effective against emerging pathogens is becoming increasingly critical. However, existing approaches cannot rapidly identify effective molecules against novel pathogens or emerging drug-resistant strains. Here, we introduce ApexOracle, an artificial intelligence (AI) model that both predicts the antibacterial potency of existing compounds and designs de novo molecules active against strains it has never encountered. Departing from models that rely solely on molecular features, ApexOracle incorporates pathogen-specific context through the integration of molecular features captured via a foundational discrete diffusion language model and a dual-embedding framework that combines genomic- and literature-derived strain representations. Across diverse bacterial species and chemical modalities, ApexOracle consistently outperformed state-of-the-art approaches in activity prediction and demonstrated reliable transferability to novel pathogens with little or no antimicrobial data. Its unified representation-generation architecture further enables the in silico creation of "new-to-nature" molecules with high predicted efficacy against priority threats. By pairing rapid activity prediction with targeted molecular generation, ApexOracle offers a scalable strategy for countering AMR and preparing for future infectious-disease outbreaks.

cs.LG↗

arXiv2025

Large Scale Multi-Task Bayesian Optimization with Large Language Models

In multi-task Bayesian optimization, the goal is to leverage experience from optimizing existing tasks to improve the efficiency of optimizing new ones. While approaches using multi-task Gaussian processes or deep kernel transfer exist, the performance improvement is marginal when scaling beyond a moderate number of tasks. We introduce a novel approach leveraging large language models (LLMs) to learn from, and improve upon, previous optimization trajectories, scaling to approximately 1500 distinct tasks. Specifically, we propose a feedback loop in which an LLM is fine-tuned on the high quality solutions to specific tasks found by Bayesian optimization (BO). This LLM is then used to generate initialization points for future BO searches for new tasks. The trajectories of these new searches provide additional training data for fine-tuning the LLM, completing the loop. We evaluate our method on two distinct domains: database query optimization and antimicrobial peptide design. Results demonstrate that our approach creates a positive feedback loop, where the LLM's generated initializations gradually improve, leading to better optimization performance. As this feedback loop continues, we find that the LLM is eventually able to generate solutions to new tasks in just a few shots that are better than the solutions produced by "from scratch" by Bayesian optimization while simultaneously requiring significantly fewer oracle calls.

cs.LG↗

arXiv2024

Interfacial Rheology of Lanthanide Binding Peptide Surfactants at the Air-Water Interface

Peptide surfactants (PEPS) are studied to capture and retain rare earth elements (REEs) at air-water interfaces to enable REE separations. Peptide sequences, designed to selectively bind REEs, depend crucially on the position of ligands within their binding loop domain. These ligands form a coordination sphere that wraps and retains the cation. We study variants of lanthanide binding tags (LBTs) designed to complex strongly with Tb$^{3+}$. The peptide LBT$^{5-}$ (with net charge -5) is known to bind Tb$^{3+}$ and adsorb with more REE cations than peptide molecules, suggesting that undesired non-specific Coulombic interactions occur. Rheological characterization of interfaces of LBT$^{5-}$ and Tb$^{3+}$ solutions reveal the formation of an interfacial gel. To probe whether this gelation reflects chelation among intact adsorbed LBT$^{5-}$:Tb$^{3+}$ complexes or destruction of the binding loop, we study a variant, LBT$^{3-}$, designed to form net neutral LBT$^{3-}$:Tb$^{3+}$ complexes. Solutions of LBT$^{3-}$ and Tb$^{3+}$ form purely viscous layers in the presence of excess Tb$^{3+}$, indicating that each peptide binds a single REE in an intact coordination sphere. We introduce the variant RR-LBT$^{3-}$ with net charge -3 and anionic ligands outside of the coordination sphere. We find that such exposed ligands promote interfacial gelation. Thus, a nuanced requirement for interfacial selectivity of PEPS is proposed: that anionic ligands outside of the coordination sphere must be avoided to prevent the non-selective recruitment of REE cations. This view is supported by simulation, including interfacial molecular dynamics simulations, and interfacial metadynamics simulations of the free energy landscape of the binding loop conformational space.

cond-mat.soft↗