Hybrid artificial intelligence and quantum annealing as an optimization layer in drug discovery

Artificial intelligence has significantly expanded candidate generation in drug discovery, but selecting optimal compounds under real-world experimental constraints remains a major bottleneck. This review examines quantum annealing as a potential downstream optimization layer within hybrid computational workflows. By focusing on constraint-dominated decision tasks formulated as quadratic unconstrained binary optimization (QUBO) problems, the authors analyze how annealing architectures complement AI-generated scores. They further discuss practical hardware limitations, reporting considerations, and the critical role of rigorous benchmarking against classical baselines.

dlvr.it

Uncovering smooth structures in single-cell data with neighbor embeddings guided by predictability-computability-stability

Single-cell data often capture continuous biological processes, yet common visualization tools can introduce misleading structure. This study systematically evaluates neighbor embedding methods using the predictability-computability-stability (PCS) framework, revealing key limitations in reliability. The authors introduce NESS, a stability-guided approach that improves the recovery of smooth biological patterns, enabling more trustworthy interpretation of cell-state dynamics across diverse datasets.

dlvr.it

Revealing hidden material composition via physics-induced deep learning for precision clinical diagnosis

Liu et al. present a physics-induced deep-learning approach that extracts tissue composition information from routine single-energy CT scans. By incorporating X-ray imaging principles into model design, this approach improves material characterization and enhances performance across diverse clinical tasks. The framework expands the diagnostic value of standard CT imaging and supports more accessible precision medicine.

dlvr.it

Network-based disruption analysis of biological coordination

Biological systems rely on coordinated interaction networks between molecules and cells that can be disrupted by physiological challenges. NeDis is an open-source framework that measures how these networks change. By identifying groups of biomarkers with characteristic disruption patterns across conditions and time points, NeDis reveals system-level adaptations that are missed by standard analyses. Applications to immune data uncover coordinated disruptions during pregnancy and preeclampsia.

dlvr.it

Sampling bias corrections for discrete and Gaussian partial information decompositions

Lorenz, Engel et al. characterize the limited-sampling bias in partial information decomposition calculations. The bias is substantial when computed on empirical brain datasets, but the bias-correction methods developed here are demonstrated to work highly effectively to produce unbiased estimates of information synergy and redundancy.

dlvr.it

Interpretable machine learning reveals hemispheric asymmetry of state switching in the suprachiasmatic nucleus

Zhang et al. developed an interpretable machine learning framework to analyze population-level calcium ion signals in the suprachiasmatic nucleus (SCN), the central circadian clock of mammals. They uncover a hemispheric asymmetry in the SCN, particularly during the subjective day. Time prediction experiments further show that each hemisphere encodes differential time-encoding features. The work highlights hemispheric asymmetry as a fundamental property of the brain’s circadian clock.

dlvr.it

mlr3mbo: Bayesian optimization in R

Many problems in engineering, the sciences, and machine learning require optimizing a system with expensive evaluations. The authors present mlr3mbo, a rich R toolbox for Bayesian optimization (BO)—the standard methodology for such problems—that pairs a modular design of BO building blocks with well-performing, empirically chosen defaults and convenient entry points. On standard benchmarks, it performs on par with or better than HEBO, SMAC3, Ax, and Optuna.

dlvr.it

A tropical geometry for bounded biochemical state spaces

Many biological measurements are bounded: a protein site cannot be less than unmodified or more than fully modified. Cobley shows that treating such data with ordinary linear tools can hide biological structure. Using cysteine oxidation as an example, Oxi-Shapes applies tropical geometry to reveal how aging reorganizes the redox proteome despite small average changes, offering a broader framework for analyzing bounded biochemical state spaces.

dlvr.it

OscillomeR infers ultradian oscillations and targets of the Hes family

OscillomeR is a master-transcription-factor-anchored framework that infers ultradian gene oscillations from bulk and single-cell RNA sequencing. By reconstructing cyclic trajectories and phase-resolved gene expression, the method recovers Hes-driven oscillations in asynchronous muscle stem cells and pancreatic progenitors. It maps widespread rhythmic programs and phase shifts, revealing dynamic rewiring of gene-regulatory networks and helping to dissect the biological functions of gene-expression oscillations across development and disease.

dlvr.it

MedicalAgentsBench for complex medical reasoning: Comparing internalized reasoning models versus externalized agent-based frameworks

Shao et al. introduce MedicalAgentsBench, a benchmark designed to separate genuine multi-step clinical reasoning from pattern matching. Using this benchmark, they compare internalized reasoning models against externalized agent frameworks and find that the two approaches are complementary; layering agents onto an internalized reasoning model achieves the highest accuracy, exceeding either approach alone.

dlvr.it

BioMaster: Multi-agent system for automated bioinformatics analysis workflow

BioMaster is a knowledge-guided multi-agent framework that automates bioinformatics workflow planning, execution, debugging, and validation. Across 49 tasks and 102 tools, it improves workflow completion over existing automated systems and supports both proprietary and open-source language models.

dlvr.it

Efficient attention mechanisms for large language models

This survey reviews efficient attention mechanisms for scalable long-context language modeling, with a focus on linear and sparse attention. It summarizes their algorithmic foundations, hardware considerations, and integration into large-scale pre-trained language models, providing a systematic reference for designing efficient transformer-based architectures.

dlvr.it

Multimodal spatial omics: From data acquisition to computational integration

Multimodal spatial omics technologies enable mapping of molecular features within tissues, but integrating these diverse data remains challenging. This review outlines experimental strategies and computational frameworks, highlighting how combining modalities can reveal tissue organization and advance biological discovery, while also outlining key challenges and future directions.

dlvr.it

Design principles for integrated AI alignment

Aligning AI models with human preferences remains a central challenge, yet the AI alignment field is increasingly divided between behavioral and representational approaches, resulting in narrowly scoped alignment strategies that are more vulnerable to failure. Drawing lessons from immunology and cybersecurity, the authors propose a set of design principles for integrated alignment frameworks that combine the complementary strengths of diverse alignment approaches. They emphasize the role of strategic diversity and outline steps toward greater coordination within the AI alignment research field.

dlvr.it

Brain-AI convergence: Generative world models and hierarchical attention for human intelligence

Large language models now exhibit human-like abilities, raising a fundamental question: are the brain and AI similar only at the surface level, or do they share deeper computational principles? This perspective argues that both systems build predictive world models through prediction-error learning and reuse those models for sensory understanding and motor generation. This indicates that the brain and AI share computational principles underlying advanced cognitive functions and intelligence, despite the long-standing view that their internal processes are fundamentally distinct.

dlvr.it

OpenChallenges: A FAIR-aligned, centralized platform for crowdsourced challenges in biomedical research

Community challenges drive benchmarking in computational biomedicine, but more than 2,000 such challenges are scattered across numerous platforms, with no central registry. OpenChallenges introduces a FAIR-aligned, open-source hub that aggregates a growing collection of challenges into a searchable registry with standardized metadata, ontology integration, and a transparent four-tier completeness score, transforming a fragmented competition ecosystem into persistent, reusable benchmarking infrastructure.

dlvr.it

LGBTQ+ in science: Cell Press

cell.com

Density-based longitudinal neuron tracking in high-density electrophysiological recordings

Tracking the same neurons across days or weeks is essential for understanding learning, stability, and plasticity in the brain but remains technically challenging in chronic electrophysiology due to probe drift and unit turnover. Huang et al. present DANT, a framework using density-based clustering for longitudinal neuron tracking that improves matching yield while maintaining accuracy across Neuropixels recordings in cortex and striatum during trained motor behaviors in freely moving rats.

dlvr.it

Disentangling covariate effects on single-cell-resolved epigenomes with DeepDive

Disentangling true biological signals from confounding factors remains a major challenge in single-cell genomics. Here, Møller and Madsen present DeepDive, a deep generative framework that separates covariate effects in single-cell chromatin accessibility data. DeepDive improves reconstruction accuracy, enables counterfactual predictions, and enhances the detection of differential accessibility under complex confounding. Applied to human pancreatic islets, it reveals disease-associated regulatory programs and links chromatin variation to genetic risk and beta-cell functional states.

dlvr.it

CLEAR-HPV: Interpretable concept discovery for human-papillomavirus-associated morphology in whole-slide histology

Understanding how AI models interpret pathology images remains a major challenge. Here, the authors introduce CLEAR-HPV, an interpretable framework that identifies meaningful tissue patterns linked to human papillomavirus status directly from whole-slide images without manual annotation. By transforming complex model representations into compact, biologically meaningful concepts, the approach maintains predictive accuracy while improving transparency. This work highlights how interpretable AI can bridge prediction and biological understanding in computational pathology.

dlvr.it

Graphlet- and motif-based link prediction in large networks

Networks underpin systems from biology to society, yet predicting missing connections remains challenging. Here, the authors show that small, recurring structural patterns alone contain powerful signals for link prediction, enabling a scalable and general approach across diverse networks. This work highlights a unifying principle: complex systems leave consistent structural signatures, allowing accurate and interpretable predictions without domain-specific inputs and advancing our ability to understand and navigate incomplete networked data.

dlvr.it

SelfCheck-Eval: A multi-module framework for zero-resource hallucination detection in large language models

Large language models are powerful but tend to generate convincing yet incorrect content, a problem called hallucination. While tools exist to catch these errors in general knowledge, they fail in mathematical reasoning, unable to reliably distinguish correct solutions from inferior ones. Muhammed, Tuccari, Rabby, et al. introduce a new mathematical benchmark and a detection framework, revealing that this failure persists across all tested approaches, signaling a fundamental gap that demands purpose-built solutions for AI reliability in technical domains.

dlvr.it

Tina: A diffusion neural network for generating personalized AI models from text prompts

Li et al. introduce Tina, a text-to-model generative AI that produces personalized neural network classifiers from natural-language descriptions. Powered by a conditional diffusion transformer with CLIP embeddings, Tina generates lightweight, task-specific models across diverse domains, including medical imaging, and generalizes to unseen tasks, offering non-expert users a direct channel to create customized AI tools.

dlvr.it

EnzymeHunter: Achieving fine-grained enzyme function prediction with a hierarchically aware contrastive learning framework

EnzymeHunter is a deep-learning framework for fine-grained enzyme function annotation based on hierarchically aware contrastive learning. It captures subtle catalytic differences, achieves state-of-the-art performance in low-homology scenarios, and accurately predicts rare enzyme classes. At the proteome scale, it can reveal novel enzymatic functions and provide interpretable insights into catalytic mechanisms.

dlvr.it

DuoMod-Net: Logarithmic balancing and geometric refinement for imbalanced semi-supervised medical image segmentation

Training AI on medical images is challenging because rare organs are often overlooked by standard models. Bo et al. introduce DuoMod-Net to address this severe class imbalance. By decoupling background noise to stabilize learning gradients and dynamically expanding feature spaces to create a geometric safety margin for rare structures, this framework improves the reliable detection of underrepresented organs, advancing robust and comprehensive medical image analysis under limited data conditions.

dlvr.it

SHIELD: A weakly supervised graph attention neural network for decoding disease-relevant cell-cell interactions

How do cells communicate in health and disease? Sehra et al. introduce SHIELD, a machine learning model decoding intercellular communication directly from multiplexed tissue images. Unlike traditional “black-box” models, SHIELD is inherently interpretable and operates without biological priors. Validated across multiple cancer entities and type 1 diabetes using different imaging technologies, this tool enables researchers to map the hidden cell communication rules of tissue function in health and disease, ultimately turning complex tissue imaging data into potentially actionable medical insights.

dlvr.it

GSFM: A gene set foundation model pre-trained on a massive collection of diverse gene sets

Large language foundation models learn patterns from text by predicting missing words after being trained on a massive and diverse body of text. Applying this concept to a functional genomic context, the authors train a gene set foundation model (GSFM), using a massive collection of diverse gene sets, to predict missing genes from these sets. By assigning high-dimensional contexts to genes, GSFM can predict gene function and enhance other key biological and biomedical data-analysis tasks.

dlvr.it

Cell Press: Patterns

Patterns is calling for submissions presenting compelling and creative reanalyses of prior works of high importance and broad impact.

cell.com