Blood pressure regulation is commonly addressed in terms of local mechanisms such as vascular resistance, compliance and neurohumoral control. However, the human vasculature encompasses multiple quasi-closed flow loops under both physiological and pathological conditions. To test whether these loops could influence pressure dynamics beyond local control, we address the role of vascular topology in blood pressure regulation. Using one dimensional flow simulation models, we compared pressure dynamics in open vascular segments and closed vascular loops. We found that in open segments pressure fades away and remains spatially localized, whereas in closed loops pressure can keep circulating around the loop even if resistance in one spot is modified. Since parallel pathways within loops are dynamically coupled rather than independent, pressure changes in one place can affect the entire closed loop, allowing system level pressure patterns to emerge. Also, we assessed the temporal evolution of pressure fluctuations within closed vascular loops in normotensive and hypertensive parameter regimes, before and after loop breaking intervention. This topological approach helps clarifying why drugs or local interventions may fail to lower blood pressure in looped vascular architectures, providing a theoretical interpretation of some forms of resistant hypertension. Because disrupting a loop restores pressure relaxation, it may also help explain the disproportionate pressure changes observed after topology altering events like thrombosis, vascular surgery or embolization of arteriovenous malformations and shunts. Therefore, vascular topology can influence cardiovascular physiology by coupling local pressure flow relations to global constraints on blood pressure regulation, with physiological, pathological and clinical implications.

Accurate modeling of drug release is essential for designing and developing controlled-release systems. Classical models (Fick, Higuchi, Peppas) rely on simplifying assumptions that limit their accuracy in complex geometries and release mechanisms. Here, we propose a novel approach using Physics-Informed Neural Networks (PINNs) and Bayesian PINNs (BPINNs) for predicting release from planar, 1D-wrinkled, and 2D-crumpled films. This approach uniquely integrates Fick's diffusion law with limited experimental data to enable accurate long-term predictions from short-term measurements, and is systematically benchmarked against classical drug release models. We embedded Fick's second law into PINN as loss with 10,000 Latin-hypercube collocation points and utilized previously published experimental datasets to assess drug release performance through mean absolute error (MAE) and root mean square error (RMSE), considering noisy conditions and limited-data scenarios. Our approach reduced mean error by up to 40% relative to classical baselines across all film types. The PINN formulation achieved RMSE <0.05 utilizing only the first 6% of the release time data (reducing 94% of release time required for the experiments) for the planar film. For wrinkled and crumpled films, the PINN reached RMSE <0.05 in 33% of the release time data. BPINNs provide tighter and more reliable uncertainty quantification under noise. By combining physical laws with experimental data, the proposed framework yields highly accurate long-term release predictions from short-term measurements, offering a practical route for accelerated characterization and more efficient early-stage drug release system formulation.

In this work we describe a new model for the evolution of a diploid structured population backwards in time that allows for large migrations and uneven offspring distributions. The model generalizes both the mean-field model of Birkner et al. [\textit{Electron. J. Probab.} 23: 1-44 (2018)] and the haploid structured model of Möhle [\textit{Theor. Popul. Biol.} 2024 Apr:156:103-116]. We show convergence, with mild conditions on the joint distribution of offspring frequencies and migrations, of gene genealogies conditional on the pedigree to a time-inhomogeneous coalescent process driven by a Poisson point process $Ψ$ that records the timing and scale of large migrations and uneven offspring distributions. This quenched scaling limit demonstrates a significant difference in the predictions of the classical annealed theory of structured coalescent processes. In particular, the annealed and quenched scaling limits coincide if and only if these large migrations and uneven offspring distributions are absent. The proof proceeds by the method of moments and utilizes coupling techniques from the theory of random walks in random environments. Several examples are given and their quenched scaling limits established.

Traditional neural networks, while powerful, rely on biologically implausible learning mechanisms such as global backpropagation. This paper introduces the Structurally Adaptive Predictive Inference Network (SAPIN), a novel computational model inspired by the principles of active inference and the morphological plasticity observed in biological neural cultures. SAPIN operates on a 2D grid where processing units, or cells, learn by minimizing local prediction errors. The model features two primary, concurrent learning mechanisms: a local, Hebbian-like synaptic plasticity rule based on the temporal difference between a cell's actual activation and its learned expectation, and a structural plasticity mechanism where cells physically migrate across the grid to optimize their information-receptive fields. This dual approach allows the network to learn both how to process information (synaptic weights) and also where to position its computational resources (network topology). We validated the SAPIN model on the classic Cart Pole reinforcement learning benchmark. Our results demonstrate that the architecture can successfully solve the CartPole task, achieving robust performance. The network's intrinsic drive to minimize prediction error and maintain homeostasis was sufficient to discover a stable balancing policy. We also found that while continual learning led to instability, locking the network's parameters after achieving success resulted in a stable policy. When evaluated for 100 episodes post-locking (repeated over 100 successful agents), the locked networks maintained an average 82% success rate.

We propose a predator-prey model with an age-structured predator population that exhibits a functional role reversal. The structure of the predator population in our model embodies the ecological concept of an "ontogenetic niche shift," in which a species' functional role changes as it grows. This structure adds complexity to our model but increases its biological relevance. The time evolution of the age-structured predator population is motivated by the Kermack-McKendrick Renewal Equation (KMRE). Unlike KMRE, the predator population's birth and death rate functions depend on the prey population's size. We establish the existence, uniqueness, and positivity of the solutions to the proposed model's initial value problem. The dynamical properties of the proposed model are investigated via Latin Hypercube Sampling in the 15-dimensional space of its parameters. Our Linear Discriminant Analysis suggests that the most influential parameters are the maturation age of the predator and the rate of consumption of juvenile predators by the prey. We carry out a detailed study of the long-term behavior of the proposed model as a function of these two parameters. In addition, we reduce the proposed age-structured model to ordinary and delayed differential equation (ODE and DDE) models. The comparison of the long-term behavior of the ODE, DDE, and the age-structured models with matching parameter settings shows that the age structure promotes the instability of the Coexistence Equilibrium and the emergence of the Coexistence Periodic Attractor.

Epileptic seizures are generated in cerebral networks that propagate ictal and interictal activity. The structure of cerebral networks underpinning epileptic activity can be inferred from diffusion-weighted MRI (DWI). However, publicly available DWI data in individuals with epilepsy are scarce, and processing is technically challenging due to scan-specific artifacts, limiting research progress. Here, we release raw DWI data from 216 individuals with epilepsy and 98 healthy controls. Subject identifiers align with our previous data release (IDEAS), which includes T1-weighted and FLAIR MRI, surgical details, and long-term seizure outcomes after surgery. Preprocessing reduced distortions and artifacts, while fully processed data include diffusion metric maps in native and template space. We also provide parcellated structural connectomes using multiple atlases and connectivity measures. To illustrate the utility of this IDEAS II data, we replicated ENIGMA consortium findings, observing widespread reductions of fractional anisotropy, particularly ipsilateral to the area of seizure onset. We further demonstrate localised abnormality, and network connectivity using streamline tractography in a patient who subsequently underwent temporal lobe resection. This open dataset offers a comprehensive resource to advance research on structural connectivity and surgical outcomes in epilepsy.

Sensitivity to small changes in the environment is crucial for many real-world tasks, enabling living and artificial systems to make correct behavioral decisions. It has been shown that such sensitivity is maximized when a system operates near the critical point of a phase transition. However, proximity to criticality introduces large fluctuations and diverging timescales. Hence, to leverage the maximal sensitivity, it would require impractically long integration periods. Here, we analytically and computationally demonstrate how the optimal tuning of a recurrent neural network is determined given a finite integration time. Rather than maximizing the theoretically available sensitivity, we find networks attain different sensitivities depending on the available time. Consequently, the optimal dynamic regime can shift away from criticality when integration times are finite, highlighting the necessity of incorporating finite-time considerations into studies of information processing.

Discrete diffusion models have recently emerged as a powerful class of generative models for chemistry and biology data. In these fields, the goal is to generate various samples with high rewards (e.g., drug-likeness in molecules), making reward-based guidance crucial. Most existing methods are based on guiding the diffusion model using intermediate rewards but tend to underperform since intermediate rewards are noisy due to the non-smooth nature of reward functions used in scientific domains. To address this, we propose Clean-Sample Markov Chain (CSMC) Sampler, a method that performs effective test-time reward-guided sampling for discrete diffusion models, enabling local search without relying on intermediate rewards. CSMC constructs a Markov chain of clean samples using the Metropolis-Hastings algorithm such that its stationary distribution is the target distribution. We design a proposal distribution by sequentially applying the forward and backward diffusion processes, making the acceptance probability tractable. Experiments on molecule and biological sequence generation with various reward functions demonstrate that our method consistently outperforms prior approaches that rely on intermediate rewards.

The paper titled ''Global gridded population datasets systematically underrepresent rural population'' by Josias Láng-Ritter et al. provides a valuable contribution to the discourse on the accuracy of global population datasets, particularly in rural areas. We recognize the efforts put into this research and appreciate its contribution to the field. However, we feel that key claims in the study are overly bold, not properly backed by evidence and lack a cautious and nuanced discussion. We hope these points will be taken into account in future discussions and refinements of population estimation methodologies. We argue that the reported bias figures are less caused by actual undercounting of rural populations, but more so by contestable methodological decisions and the historic misallocation of (gridded) population estimates on the local level.

As single-cell RNA sequencing datasets grow in adoption, scale, and complexity, data analysis remains a bottleneck for many research groups. Although frontier AI agents have improved dramatically at software engineering and general data analysis, it remains unclear whether they can extract biological insight from messy, real-world single-cell datasets. We introduce scBench, a benchmark of 394 verifiable problems derived from practical scRNA-seq workflows spanning six sequencing platforms and seven task categories. Each problem provides a snapshot of experimental data immediately prior to an analysis step and a deterministic grader that evaluates recovery of a key biological result. Benchmark data on eight frontier models shows that accuracy ranges from 29-53%, with strong model-task and model-platform interactions. Platform choice affects accuracy as much as model choice, with 40+ percentage point drops on less-documented technologies. scBench complements SpatialBench to cover the two dominant single-cell modalities, serving both as a measurement tool and a diagnostic lens for developing agents that can analyze real scRNA-seq datasets faithfully and reproducibly.

Diabetes is a chronic metabolic disorder characterized by elevated blood glucose levels due to impaired insulin production or function. Two main forms are recognized: type 1 diabetes (T1D), which involves autoimmune destruction of insulin-producing \b{eta}-cells, and type 2 diabetes (T2D), which arises from insulin resistance and progressive \b{eta}-cell dysfunction. Understanding the molecular mechanisms underlying these diseases is essential for the development of improved therapeutic strategies, particularly those targeting \b{eta}-cell dysfunction. To investigate these mechanisms in a controlled and biologically interpretable setting, mouse models have played a central role in diabetes research. Owing to their genetic and physiological similarity to humans, together with the ability to precisely manipulate their genome, mice enable detailed investigation of disease progression and gene function. In particular, mouse models have provided critical insights into \b{eta}-cell development, cellular heterogeneity, and functional failure under diabetic conditions. Building on these experimental advances, this study applies machine learning methods to single-cell transcriptomic data from mouse pancreatic islets. Specifically, we evaluate two supervised approaches identified in the literature; Extra Trees Classifier (ETC) and Partial Least Squares Discriminant Analysis (PLS-DA), to assess their ability to identify T2D-associated gene expression signatures at single-cell resolution. Model performance is evaluated using standard classification metrics, with an emphasis on interpretability and biological relevance

Somatic mechanical stimulation (e.g., acupuncture) exerts systemic immunomodulatory effects, yet the cellular bridge translating peripheral physical force into visceral repair remains elusive. Here, employing a custom interpretable deep learning framework (CARSS) on single-cell RNA sequencing data, we identify CD34$^{+}$PDGFR$α$$^{+}$ telocytes (CPTCs) as the primary mechanosensors in both fascia and colon during bacterial colitis. We show that somatic mechanotherapy triggers an AP-1/Hsp70-dependent transcriptional program in fascial CPTCs, inducing systemic Wnt elevation, which elicits a "transcriptional resonance" in colonic CPTCs, reprogramming their communication network from an inflammatory amplifier to a Wnt-driven regenerative hub. Mechanistically, this axis activates epithelial $β$-catenin/Myc signaling, suppressing apoptosis and restoring barrier integrity independent of immune cells. Our findings define a CPTC-Driven Mechano-Resonance Axis, where CPTCs serve as synchronized relay stations that convert local mechanical cues into systemic regenerative microenvironments.

This paper presents the study of a DNA replication model grounded in the biochemical kinetics of DNA polymerases, which copy each DNA strand into a complementary strand, except for rare point-like mutations caused by nucleotide substitution errors. Numerical simulations of many successive replications, starting from an arbitrary initial DNA sequence, show that the fractions of mono- and oligonucleotides converge toward compliance with Chargaff's second parity rule. The theoretical framework developed for this multireplication process demonstrates that the near-equalities of complementary nucleotide fractions arise from two key features: (1) the dominant role of base-pair complementarity in replication kinetics and (2) the low intrinsic error rate of DNA polymerases. Together, these two features yield a robust mechanistic basis for Chargaff's second parity rule. These considerations explain the existence of deviations with respect to the predictions of models assuming no-strand-bias conditions.

Accurate and generalisable segmentation of stroke lesions from magnetic resonance imaging (MRI) is essential for advancing clinical research, prognostic modelling, and personalised interventions. Although deep learning has improved automated lesion delineation, many existing models are optimised for narrow imaging contexts and generalise poorly to independent datasets, modalities, and stroke stages. Here, we systematically evaluated stroke lesion segmentation using the nnU-Net framework across multiple heterogeneous, publicly available MRI datasets spanning acute and chronic stroke. Models were trained and tested on diffusion-weighted imaging (DWI), fluid-attenuated inversion recovery (FLAIR), and T1-weighted MRI, and evaluated on independent datasets. Across stroke stages, models showed robust generalisation, with segmentation accuracy approaching reported inter-rater reliability. Performance varied with imaging modality and training data characteristics. In acute stroke, DWI-trained models consistently outperformed FLAIR-based models, with only modest gains from multimodal combinations. In chronic stroke, increasing training set size improved performance, with diminishing returns beyond several hundred cases. Lesion volume was a key determinant of accuracy: smaller lesions were harder to segment, and models trained on restricted volume ranges generalised poorly. MRI image quality further constrained generalisability: models trained on lower-quality scans transferred poorly, whereas those trained on higher-quality data generalised well to noisier images. Discrepancies between predictions and reference masks were often attributable to limitations in manual annotations. Together, these findings show that automated lesion segmentation can approach human-level performance while identifying key factors governing generalisability and informing the development of lesion segmentation tools.

Neural encoding models aim to predict fMRI-measured brain responses to natural images. fMRI data is acquired as a 3D volume of voxels, where each voxel has a defined spatial location in the brain. However, conventional encoding models often flatten this volume into a 1D vector and treat voxel responses as independent outputs. This removes spatial context, discards anatomical information, and ties each model to a subject-specific voxel grid. We introduce the Neural Response Function (NRF), a framework that models fMRI activity as a continuous function over anatomical space rather than a flat vector of voxels. NRF represents brain activity as a continuous implicit function: given an image and a spatial coordinate (x, y, z) in standardized MNI space, the model predicts the response at that location. This formulation decouples predictions from the training grid, supports querying at arbitrary spatial resolutions, and enables resolution-agnostic analyses. By grounding the model in anatomical space, NRF exploits two key properties of brain responses: (1) local smoothness -- neighboring voxels exhibit similar response patterns; modeling responses continuously captures these correlations and improves data efficiency, and (2) cross-subject alignment -- MNI coordinates unify data across individuals, allowing a model pretrained on one subject to be fine-tuned on new subjects. In experiments, NRF outperformed baseline models in both intrasubject encoding and cross-subject adaptation, achieving high performance while reducing the data size needed by orders of magnitude. To our knowledge, NRF is the first anatomically aware encoding model to move beyond flattened voxels, learning a continuous mapping from images to brain responses in 3D space.

The computational role of imagination remains debated. While classical accounts emphasize reward maximization, emerging evidence suggests it accesses internal world models (IWMs). We employ psychological network analysis to compare IWMs in humans and large language models (LLMs) via imagination vividness ratings, distinguishing offline world models (persistent memory structures accessed independent of immediate goals) from online models (task-specific representations). Analyzing 2,743 humans across three populations and six LLM variants, we find human imagination networks exhibit robust structural consistency, with high centrality correlations and aligned clustering. LLMs show minimal clustering and weak correlations with human networks, even with conversational memory, across environmental and sensory contexts. These differences highlight disparities in how biological and artificial systems organize internal representations. Our framework offers quantitative metrics for evaluating offline world models in cognitive agents.

Background. Preeclampsia (PE) complicates 2-8% of pregnancies and involves placental hypoxia and HIF-pathway activation, especially in early-onset PE (eoPE). Chemical mimetics like cobalt (II) chloride (CoCl2) and oxyquinoline derivatives model trophoblast hypoxia in vitro, yet their fidelity in recapitulating PE gene profiles remains unclear. Integrating patient tissue analyses with experimental models may reveal common markers and validate physiologically relevant paradigms. Methods. We analyzed scRNA-seq data from 10 eoPE, 7 late-onset PE, and matched control placentas, identifying villous cytotrophoblast, syncytiotrophoblast, and extravillous trophoblast (EVT). BeWo b30 cells were treated for 24 h with CoCl2 (300 $μ$M) or the oxyquinoline derivative neuradapt (5 $μ$M) to induce hypoxia. RNA-seq with qPCR validation and small RNA-seq quantified mRNA and microRNA changes; PROGENy inferred pathway activities. Results. scRNA-seq revealed highest hypoxia activation in eoPE, with EVT showing maximum activity. Nine genes were upregulated across all trophoblast types (EBI3, CST6, FN1, RFK, COL17A1, LDHA, PKP2, RPS4Y1, RPS26). In vitro, neuradapt induced more specific hypoxia responses than CoCl2 (1284 vs. 3032 differentially expressed genes). Critically, EBI3, FN1, and COL17A1 showed concordant upregulation in tissue and neuradapt-treated cells, whereas CoCl2 produced opposite patterns. MicroRNAs hsa-miR-27a-5p and hsa-miR-193b-5p were consistently elevated in both models; 3'-isoforms of hsa-miR-9-5p and hsa-miR-92b-3p were identified as hypoxia-associated. Conclusions. EBI3, COL17A1, miR-27a-5p, and miR-193b-5p emerge as trophoblast hypoxia markers. Neuradapt (a selective HIF-prolyl hydroxylase inhibitor) provides a more physiologically relevant in vitro model than CoCl2, recapitulating transcriptomic signatures observed in PE placentas.

De novo genome assembly is challenging in highly repetitive regions; however, reference-guided assemblers often suffer from bias. We propose a framework for pangenome-guided sequence assembly, which can resolve short-read data in complex regions without bias towards a single reference genome. Our primary contribution is to frame the assembly as a graph traversal optimisation problem, which can be implemented classically or on a quantum computer. The workflow involves first annotating pangenome graphs with estimated copy numbers for each node, then finding a path on the graph that best explains those copy numbers. On simulated data, our approach significantly reduces the number of contigs compared to de novo assemblers. While they introduce a small increase in inaccuracies, such as false joins, our optimisation-based methods are competitive with current exhaustive search techniques. They are also designed to scale more efficiently as the problem size grows and will run effectively on future quantum computers; a small experiment on a real quantum device showcases this behaviour. Moreover, they are more resilient to noise in copy number estimation inherent in short-read-based assembly. We also develop novel tools for creating realistic synthetic pangenomes, aligning reads to pangenomes and for evaluating assembly quality.

A grand challenge in modern neuroscience is to bridge the gap between the detailed mapping of microscale neural circuits and mechanistic understanding of cognitive functions. While extensive knowledge exists about neuronal connectivity and biophysics, how these low-level phenomena eventually produce abstract behaviors remains largely unresolved. Here, we propose that a model based on State Space Models, an emerging class of deep learning architectures, can be a potential biological model for analysis. We suggest that the differential equations governing elements in a State Space Model are conceptually consistent with the dynamics of biophysical processes, while the model offers a scalable framework to build on the dynamics to produce emergent behaviors observed in experimental neuroscience. We test this model by training a network employing a diagonal state transition matrix on temporal discrimination tasks with reinforcement learning. Our results suggest that neural behaviors such as time cells naturally emerge from two fundamental principles: optimal pre-configuration and rotational dynamics. These features are shown mathematically to optimize history compression, and naturally generate structured temporal dynamics even prior to training, mirroring recent findings in biological circuits. We show that learning acts primarily as a selection mechanism that fine-tunes these pre-configured oscillatory modes, rather than constructing temporal codes de novo. The model can be readily scaled to abstract cognitive functions such as event counting, supporting the use of State Space Models as a computationally tractable framework for understanding neural activities.

Intrinsically disordered regions (IDRs) play central roles in cellular function, yet remain poorly evaluated by existing protein structure prediction benchmarks. Current evaluations largely focus on well-folded domains, overlooking three fundamental challenges in realistic biological settings: the structural complexity of proteins, the resulting low availability of reliable ground truth, and prediction uncertainty that can propagate into high-risk downstream failures, such as in drug discovery, protein-protein interaction modeling, and functional annotation. We present DisProtBench, an IDR-centric benchmark that explicitly incorporates prediction uncertainty into the evaluation of protein structure prediction models (PSPMs). To address structural complexity and ground-truth scarcity, we curate and unify a large-scale, multi-modal dataset spanning disease-relevant IDRs, GPCR-ligand interactions, and multimeric protein complexes. To assess predictive uncertainty, we introduce Functional Uncertainty Sensitivity (FUS), a novel prediction uncertainty-stratified metric that quantifies downstream task performance under prediction uncertainty. Using this benchmark, we conduct a systematic evaluation of state-of-the-art PSPMs and reveal clear, task-dependent failure modes. Protein-protein interaction prediction degrades sharply in IDRs, while structure-based drug discovery remains comparatively robust. These effects are largely invisible to standard global accuracy metrics, which overestimate functional reliability under prediction uncertainty. We have open-sourced our benchmark and the codebase at https://github.com/Susan571/DisProtBench.

Predicting future neural activity is a core challenge in modeling brain dynamics, with applications ranging from scientific investigation to closed-loop neurotechnology. While recent models of population activity emphasize interpretability and behavioral decoding, neural forecasting-particularly across multi-session, spontaneous recordings-remains underexplored. We introduce POCO, a unified forecasting model that combines a lightweight univariate forecaster with a population-level encoder to capture both neuron-specific and brain-wide dynamics. Trained across five calcium imaging datasets spanning zebrafish, mice, and C. elegans, POCO achieves state-of-the-art accuracy at cellular resolution in spontaneous behaviors. After pre-training, POCO rapidly adapts to new recordings with minimal fine-tuning. Notably, POCO's learned unit embeddings recover biologically meaningful structure-such as brain region clustering-without any anatomical labels. Our comprehensive analysis reveals several key factors influencing performance, including context length, session diversity, and preprocessing. Together, these results position POCO as a scalable and adaptable approach for cross-session neural forecasting and offer actionable insights for future model design. By enabling accurate, generalizable forecasting models of neural dynamics across individuals and species, POCO lays the groundwork for adaptive neurotechnologies and large-scale efforts for neural foundation models. Code is available at https://github.com/yuvenduan/POCO.

As large language models (LLMs) continue to revolutionize AI research, there is a growing interest in building large-scale brain foundation models to advance neuroscience. While most existing brain foundation models are pre-trained on time-series signals or connectome features, we propose a novel graph-based pre-training paradigm for constructing a brain graph foundation model. In this paper, we introduce the Brain Graph Foundation Model, termed BrainGFM, a unified framework that leverages graph contrastive learning and graph masked autoencoders for large-scale fMRI-based pre-training. BrainGFM is pre-trained on a diverse mixture of brain atlases with varying parcellations, significantly expanding the pre-training corpus and enhancing the model's ability to generalize across heterogeneous fMRI-derived brain representations. To support efficient and versatile downstream transfer, we integrate both graph prompts and language prompts into the model design, enabling BrainGFM to flexibly adapt to a wide range of atlases, neurological and psychiatric disorders, and task settings. Furthermore, we employ meta-learning to optimize the graph prompts, facilitating strong generalization to previously unseen disorders under both few-shot and zero-shot learning conditions via language-guided prompting. BrainGFM is pre-trained on 27 neuroimaging datasets spanning 25 common neurological and psychiatric disorders, encompassing 2 types of brain atlases (functional and anatomical) across 8 widely-used parcellations, and covering over 25,000 subjects, 60,000 fMRI scans, and a total of 400,000 graph samples aggregated across all atlases and parcellations.

We study the equilibrium phases of a generalized Lotka-Volterra model characterized by a species interaction matrix which is random, sparse and symmetric. Dynamical fluctuations are modeled by a demographic noise with amplitude proportional to the effective temperature T. The equilibrium distribution of species abundances is obtained by means of the cavity method and the Belief Propagation equations, which allow for an exact solution on sparse networks. Our results reveal a rich and non-trivial phenomenology that deviates significantly from the predictions of fully connected models. Consistently with data from real ecosystems, which are characterized by sparse rather than dense interaction networks, we find strong deviations from Gaussianity in the distribution of abundances. In addition to the study of these deviations from Gaussianity, which are not related to multiple-equilibria, we also identified a novel topological glass phase, present at both finite temperature, as shown here, and at T=0, as previously suggested in the literature. The peculiarity of this phase, which differs from the multiple-equilibria phase of fully-connected networks, is its strong dependence on the presence of extinctions. These findings provide new insights into how network topology and disorder influence ecological networks, particularly emphasizing that sparsity is a crucial feature for accurately modeling real-world ecological phenomena.

Task-trained recurrent neural networks (RNNs) are widely used in neuroscience and machine learning to model dynamical computations. To gain mechanistic insight into how neural systems solve tasks, prior work often reverse-engineers individual trained networks. However, different RNNs trained on the same task and achieving similar performance can exhibit strikingly different internal solutions, a phenomenon known as solution degeneracy. Here, we develop a unified framework to systematically quantify and control solution degeneracy across three levels: behavior, neural dynamics, and weight space. We apply this framework to 3,400 RNNs trained on four neuroscience-relevant tasks: flip-flop memory, sine wave generation, delayed discrimination, and path integration, while systematically varying task complexity, learning regime, network size, and regularization. We find that higher task complexity and stronger feature learning reduce degeneracy in neural dynamics but increase it in weight space, with mixed effects on behavior. In contrast, larger networks and structural regularization reduce degeneracy at all three levels. These findings empirically validate the Contravariance Principle and provide practical guidance for researchers seeking to tune the variability of RNN solutions, either to uncover shared neural mechanisms or to model the individual variability observed in biological systems. This work provides a principled framework for quantifying and controlling solution degeneracy in task-trained RNNs, offering new tools for building more interpretable and biologically grounded models of neural computation.

Diffusion models have achieved promising results for Structure-Based Drug Design (SBDD). Nevertheless, high-quality protein subpocket and ligand data are relatively scarce, which hinders the models' generation capabilities. Recently, Direct Preference Optimization (DPO) has emerged as a pivotal tool for aligning generative models with human preferences. In this paper, we propose DecompDPO, a structure-based optimization method aligns diffusion models with pharmaceutical needs using multi-granularity preference pairs. DecompDPO introduces decomposition into the optimization objectives and obtains preference pairs at the molecule or decomposed substructure level based on each objective's decomposability. Additionally, DecompDPO introduces a physics-informed energy term to ensure reasonable molecular conformations in the optimization results. Notably, DecompDPO can be effectively used for two main purposes: (1) fine-tuning pretrained diffusion models for molecule generation across various protein families, and (2) molecular optimization given a specific protein subpocket after generation. Extensive experiments on the CrossDocked2020 benchmark show that DecompDPO significantly improves model performance, achieving up to 95.2% Med. High Affinity and a 36.2% success rate for molecule generation, and 100% Med. High Affinity and a 52.1% success rate for molecular optimization. Code is available at https://github.com/laviaf/DecompDPO.

Structure-based drug design (SBDD) aims to generate ligands that bind strongly and specifically to target protein pockets. Recent diffusion models have advanced SBDD by capturing the distributions of atomic positions and types, yet they often underemphasize binding affinity control during generation. To address this limitation, we introduce \textbf{\textnormal{\textbf{BADGER}}}, a general \textbf{binding-affinity guidance framework for diffusion models in SBDD}. \textnormal{\textbf{BADGER} }incorporates binding affinity awareness through two complementary strategies: (1) \textit{classifier guidance}, which applies gradient-based affinity signals during sampling in a plug-and-play fashion, and (2) \textit{classifier-free guidance}, which integrates affinity conditioning directly into diffusion model training. Together, these approaches enable controllable ligand generation guided by binding affinity. \textnormal{\textbf{BADGER} } can be added to any diffusion model and achieves up to a \textbf{60\% improvement in ligand--protein binding affinity} of sampled molecules over prior methods. Furthermore, we extend the framework to \textbf{multi-constraint diffusion guidance}, jointly optimizing for binding affinity, drug-likeness (QED), and synthetic accessibility (SA) to design realistic and synthesizable drug candidates.

This article presents an artificial intelligence (AI) architecture intended to simulate the iterative updating of the human working memory system. It features several interconnected neural networks designed to emulate the specialized modules of the cerebral cortex. These are structured hierarchically and integrated into a global workspace. They are capable of temporarily maintaining high-level representational patterns akin to the psychological items maintained in working memory. This maintenance is made possible by persistent neural activity in the form of two modalities: sustained neural firing (resulting in a focus of attention) and synaptic potentiation (resulting in a short-term store). Representations held in persistent activity are recursively replaced resulting in incremental changes to the content of the working memory system. As this content gradually evolves, successive processing states overlap and are continuous with one another. The present article will explore how this architecture can lead to iterative shift in the distribution of coactive representations, ultimately leading to mental continuity between processing states, and thus to human-like thought and cognition. Taken together, these components outline a biologically motivated route toward synthetic consciousness or artificial sentience and subjectivity.