Ficus maxima (Moraceae, subgenus Pharmacosycea), known in Brazil as "caxinguba", occurs in northern Brazil and in Mato Grosso. Its leaves and fruits are important food sources for birds and mammals, and the species is traditionally used by Indigenous peoples of Central and South America to treat intestinal parasites, gingivitis, internal inflammation, and snakebites. Despite its medicinal relevance, studies on its chemistry and biological properties are still limited. This work aimed to characterize the chemical constituents of F. maxima leaves and stem bark using UPLC-MS-MS and to evaluate the antinociceptive activity and CYP1A inhibition of ethanolic extracts and fractions. Plant material collected in Abaetetuba, Para (October 2013), was extracted and fractionated by silica gel column chromatography. UPLC-MS-MS analyses in positive and negative ion modes enabled putative metabolite identification supported by MS/MS libraries and molecular networking on GNPS. A total of 45 metabolites were identified, including flavonoids, triterpenes, coumarins, polyunsaturated fatty acids, amino acids, and alkaloids. Major constituents in the leaf extract included triterpene 51, flavonoids 75 and 70, and alkaloid 29. The stem bark ethanolic extract showed marked antinociceptive activity in the inflammatory phase of the formalin test (muscarinic pathway) and 62.6 +/- 9.2% activity in the hot-plate test (opioid pathway). The dichloromethane fraction of the leaves exhibited potent CYP1A1 inhibitory activity in vitro. These findings support the traditional use of F. maxima as an anti-inflammatory resource and associate its major constituents with the observed activities, reinforcing its potential as a therapeutic source that deserves further investigation.

Bacterial chemotactic sensing converts noisy chemical signals into running and tumbling. We analyze the static sensing limits of mixed Tar/Tsr chemoreceptor clusters in individual Escherichia coli cells using a heterogeneous Monod-Wyman-Changeux (MWC) model. By sweeping a seven-dimensional parameter space, we compute three sensing performance metrics-channel capacity, effective Hill coefficient, and dynamic range. Across E. coli-like parameter regimes, we consistently observe pronounced local maxima of channel capacity, whereas neither the effective Hill coefficient nor the dynamic range exhibit comparable optimization. The capacity-achieving input distribution is bimodal, which implies that individual cells maximize information by sampling both low- and high concentration regimes. Together, these results suggest that, at the individual-cell level, channel capacity may be selected for in E. coli receptor clusters.

In this paper, we extend the demographic eco-evolutionary game approach, based on explicit birth and death dynamics instead of abstract "fitness" interpreted as an abstract "Malthusian parameter", by the introduction of the delay resulting from the juvenile maturation time. This leads to the application of the Delay Differential Equations (DDE). We show that delay seriously affects the resulting dynamics and may lead to the loss of stability of equilibria when critical delay is exceeded. We provide theoretical tools for the assessment of the critical delays and the parameter values when this may happen. Our results emphasize the importance of the mechanisms of density dependence. We analyze the impact of three different suppression modes based on: adult mortality, juvenile recruitment survival after the maturation period (without delay), and juvenile recruitment at birth (with the delay). The last mode leads to extreme patterns such as bifurcations, complex cycles, and chaotic dynamics. However, surprisingly, this mode leads to extension of the duration of the temporary transient metastable states known as "ghost attractors". In addition, we also focus on the problem of resilience of the analyzed systems against external periodic perturbations and feedback-driven factors such as additional predator pressure.

Neuronal systems often preserve their characteristic functions and signalling patterns, also referred to as regimes, despite parametric uncertainties and variations. For neural models having uncertain parameters with a known probability distribution, probabilistic robustness analysis (PRA) allows us to understand and quantify under which uncertainty conditions a regime is preserved in expectation. We introduce a new computational framework for the efficient and systematic PRA of dynamical systems in neuroscience and we show its efficacy in analysing well-known neural models that exhibit multiple dynamical regimes: the Hindmarsh-Rose model for single neurons and the Jansen-Rit model for cortical columns. Given a model subject to parametric uncertainty, we employ generalised polynomial chaos to derive mean neural activity signals, which are then used to assess the amount of parametric uncertainty that the system can withstand while preserving the current regime, thereby quantifying the regime's robustness to such uncertainty. To assess persistence of regimes, we propose new metrics, which we apply to recurrence plots obtained from the mean neural activity signals. The overall result is a novel, general computational methodology that combines recurrence plot analysis and systematic persistence analysis to assess how much the uncertain model parameters can vary, with respect to their nominal value, while preserving the nominal regimes in expectation. We summarise the PRA results through probabilistic regime preservation (PRP) plots, which capture the effect of parametric uncertainties on the robustness of dynamical regimes in the considered models.

Human cognition integrates information across nested timescales. While the cortex exhibits hierarchical Temporal Receptive Windows (TRWs), local circuits often display heterogeneous time constants. To reconcile this, we trained biologically constrained deep networks, based on scale-invariant hippocampal time cells, on a language classification task mimicking the hierarchical structure of language (e.g., 'letters' forming 'words'). First, using a feedforward model (SITHCon), we found that a hierarchy of TRWs emerged naturally across layers, despite the network having an identical spectrum of time constants within layers. We then distilled these inductive priors into a biologically plausible recurrent architecture, SITH-RNN. Training a sequence of architectures ranging from generic RNNs to this restricted subset showed that the scale-invariant SITH-RNN learned faster with orders-of-magnitude fewer parameters, and generalized zero-shot to out-of-distribution timescales. These results suggest the brain employs scale-invariant, sequential priors - coding "what" happened "when" - making recurrent networks with such priors particularly well-suited to describe human cognition.

DNA language models have emerged as powerful tools for decoding the complex language of DNA sequences. However, the performance of these models is heavily affected by their tokenization strategy, i.e., a method used to parse DNA sequences into a shorter sequence of chunks. In this work, we propose DNACHUNKER, which integrates a learnable dynamic DNA tokenization mechanism and is trained as a masked language model. Adopting the dynamic chunking procedure proposed by H-Net, our model learns to segment sequences into variable-length chunks. This dynamic chunking offers two key advantages: it's resilient to shifts and mutations in the DNA, and it allocates more detail to important functional areas. We demonstrate the performance of DNACHUNKER by training it on the human reference genome (HG38) and testing it on the Nucleotide Transformer and Genomic benchmarks. Further ablative experiments reveal that DNACHUNKER learns tokenization that grasps biological grammar and uses smaller chunks to preserve detail in important functional elements such as promoters and exons, while using larger chunks for repetitive, redundant regions.

Do transformers learn like brains? A key challenge in addressing this question is that transformers and brains are trained on fundamentally different data. Brains are initially "trained" on prenatal sensory experiences (e.g., retinal waves), whereas transformers are typically trained on large datasets that are not biologically plausible. We reasoned that if transformers learn like brains, then they should develop the same structure as newborn brains when exposed to the same prenatal data. To test this prediction, we simulated prenatal visual input using a retinal wave generator. Then, using self-supervised temporal learning, we trained transformers to adapt to those retinal waves. During training, the transformers spontaneously developed the same structure as newborn visual systems: (1) early layers became sensitive to edges, (2) later layers became sensitive to shapes, and (3) the models developed larger receptive fields across layers. The organization of newborn visual systems emerges spontaneously when transformers adapt to a prenatal visual world. This developmental convergence suggests that brains and transformers learn in common ways and follow the same general fitting principles.

This study explores the effects of electric charge on the dynamics of drug transport and absorption in subcutaneous injections of monoclonal antibodies (mAbs). We develop a novel mathematical and computational model, based on the Nernst-Planck equations and porous media flow theory, to investigate the complex interactions between mAbs and charged species in subcutaneous tissue. The model enables us to study short-term transport dynamics and long-term binding and absorption for two mAbs with different electric properties. We examine the influence of buffer pH, body mass index, injection depth, and formulation concentration on drug distribution and compare our numerical results with experimental data from the literature.

We present Connection-Aware Motif Sequencing (CamS), a graph-to-sequence representation that enables decoder-only Transformers to learn molecular graphs via standard next-token prediction (NTP). For molecular property prediction, SMILES-based NTP scales well but lacks explicit topology, whereas graph-native masked modeling captures connectivity but risks disrupting the pivotal chemical details (e.g., activity cliffs). CamS bridges this gap by serializing molecular graphs into structure-rich causal sequences. CamS first mines data-driven connection-aware motifs. It then serializes motifs via scaffold-rooted breadth-first search (BFS) to establish a stable core-to-periphery order. Crucially, CamS enables hierarchical modeling by concatenating sequences from fine to coarse motif scales, allowing the model to condition global scaffolds on dense, uncorrupted local structural evidence. We instantiate CamS-LLaMA by pre-training a vanilla LLaMA backbone on CamS sequences. It achieves state-of-the-art performance on MoleculeNet and the activity-cliff benchmark MoleculeACE, outperforming both SMILES-based language models and strong graph baselines. Interpretability analysis confirms that our multi-scale causal serialization effectively drives attention toward cliff-determining differences.

Credit assignment--how changes in individual neurons and synapses affect a network's output--is central to learning in brains and machines. Noise correlation, which estimates gradients by correlating perturbations of activity with changes in output, provides a biologically plausible solution to credit assignment but scales poorly as accurately estimating the Jacobian requires that the number of perturbations scale with network size. Moreover, isotropic noise conflicts with neurobiological observations that neural activity lies on a low-dimensional manifold. To address these drawbacks, we propose neural manifold noise correlation (NMNC), which performs credit assignment using perturbations restricted to the neural manifold. We show theoretically and empirically that the Jacobian row space aligns with the neural manifold in trained networks, and that manifold dimensionality scales slowly with network size. NMNC substantially improves performance and sample efficiency over vanilla noise correlation in convolutional networks trained on CIFAR-10, ImageNet-scale models, and recurrent networks. NMNC also yields representations more similar to the primate visual system than vanilla noise correlation. These findings offer a mechanistic hypothesis for how biological circuits could support credit assignment, and suggest that biologically inspired constraints may enable, rather than limit, effective learning at scale.

A number of simple chaotic three-dimensional dynamical systems (DSs) with quadratic polynomials on the right-hand sides are reported in the literature, containing exactly 5 or 6 monomials of which only 1 or 2 are quadratic. However, none of these simple systems are chemical dynamical systems (CDSs) - a special subset of polynomial DSs that model the dynamics of mass-action chemical reaction networks (CRNs). In particular, only a small number of three-dimensional quadratic CDSs with chaos are reported, all of which have at least 9 monomials and at least 3 quadratics, with CRNs containing at least 7 reactions and at least 3 quadratic ones. To bridge this gap, in this paper we prove some basic properties of chaotic CDSs, including that those in three dimensions have at least 6 monomials, at least one of which is negative and quadratic. We then use these results to computationally find 20 chaotic three-dimensional CDSs with 6 monomials and as few as 4 quadratics, or 7 monomials and as few as 2 quadratics. At the CRN level, some of these systems have 4 reactions of which only 3 are quadratic, or 5 reactions with only 2 being quadratic. These results quantify structural complexity of chaotic CDSs, and indicate that they are ubiquitous.

Defining agency is an extremely important challenge for cognitive science and artificial intelligence. Physics generally describes mechanical happenings, but there remains an unbridgeable gap between them and the acts of agents. To discuss the morality and responsibility of agents, it is necessary to model acts; whether such responsible acts can be fully explained by physical determinism has been debated. Although we have already proposed a physical "agent determinism" model that appears to go beyond mere mechanical happenings, we have not yet established a strict mathematical formalism to eliminate ambiguity. Here, we explain why a physical system can follow coarse-graining agent-level determination without violating physical laws by formulating supervenient causation. Generally, supervenience including coarse graining does not change without a change in its lower base; therefore, a single supervenience alone cannot define supervenient causation. We define supervenient causation as the causal efficacy from the supervenience level to its lower base level. Although an algebraic expression composed of the multiple supervenient functions does supervenes on the base, a sequence of indices that determines the algebraic expression does not supervene on the base. Therefore, the sequence can possess unique dynamical laws that are independent of the lower base level. This independent dynamics creates the possibility for temporally preceding changes at the supervenience level to cause changes at the lower base level. Such a dual-laws system is considered useful for modeling self-determining agents such as humans.

Validation studies that assess the applicability and reliability of analytical and predictive methods rely on reference sets whose composition implicitly encodes multiple competing design objectives. Because these trade-offs are typically addressed through expert judgment, their structure often remains implicit, making it difficult to systematically examine how design choices shape evaluation outcomes. Here, we formulate reference set construction as an explicit multi-objective design problem. We define interpretable objective functions capturing structural, physicochemical, and response-related diversity, and employ a genetic algorithm as an exploratory solver to visualize the resulting trade-off structure. Rather than prescribing optimal or recommended reference sets, this framework enables systematic exploration of feasible designs and explicit comparison of their positions within a multi-dimensional design space. We apply this formulation to validation studies of toxicity assays as a representative real-world case. Using illustrative analyses under fixed modeling protocols, we show that reference set selection functions as an independent design axis that determines what properties of model behavior are observed under evaluation, without attributing such effects to model performance itself. Together, this work provides a general framework for making implicit trade-offs in reference set construction explicit. By complementing established expert-driven practices, the proposed approach supports more transparent discussion and interpretation of evaluation design choices across experimental validation settings.

Human cooperation persists among strangers in large, well-mixed populations despite theoretical predictions of difficulties, leaving a fundamental evolutionary puzzle. While upstream (pay-it-forward: helping others because you were helped) and downstream (rewarding-reputation: helping those with good reputations) indirect reciprocity have been independently considered as solutions, their joint dynamics in multiplayer contexts remain unexplored. We study public goods games without self-return (often called "others-only" PGGs) with benefit b and cost c and analyze evolutionary dynamics for three strategies: unconditional cooperation (ALLC), unconditional defection (ALLD), and an integrated reciprocity strategy combining unconditional forwarding with reputation-based discrimination. We show that integrating upstream and downstream reciprocity can yield a globally asymptotically stable mixed equilibrium of ALLD and integrated reciprocators when b/c > 2 in the absence of complexity costs. We analytically derive a critical threshold for complexity costs. If cognitive demands exceed this threshold, the stable equilibrium disappears via a saddle-node bifurcation. Otherwise, within the stable regime, complexity costs counterintuitively stabilize the equilibrium by preventing not only ALLC but also alternative conditional strategies from invading. Rather than requiring uniformity, our model reveals one pathway to stable cooperation through strategic diversity. ALLD serves as "evolutionary shields" preventing system collapse while integrated reciprocators flexibly combine open and discriminative responses. This framework demonstrates how pay-it-forward broadcasting and reputation systems can jointly maintain social polymorphism including cooperation despite cognitive limitations and group size challenges, offering a potential evolutionary foundation for behavioral diversity in human societies.

Understanding how receptive fields emerge and organize within brain networks and how neural dynamics couple with stimuli space is fundamental to neuroscience. Models often rely on fine-tuning connectivity to match empirical data, which may limit biological plausibility. Here we propose a physiologically grounded alternative where receptive fields and population-level attractor dynamics arise naturally from the effective hyperbolic geometry of scale-free networks. By associating stimulus space with the boundary of a hyperbolic embedding, we simulate neural dynamics using rate-based and spiking models, revealing localized activity patterns that reflect stimulus space structure without synaptic fine-tuning. The resulting receptive fields follow experimentally observed statistics and properties, and their sizes depends on neuron's connectivity degree. The model generalizes across stimuli dimensionalities and various modalities, such as orientation and place selectivity. Experimental analyses of hippocampal place fields recorded on a linear track support these findings. This framework offers a novel organizing principle linking network structure, stimulus space encoding, and neural dynamics, providing insights into receptive field formation across diverse brain areas.

Previous studies have compared neural activities in the visual cortex to representations in deep neural networks trained on image classification. Interestingly, while some suggest that their representations are highly similar, others argued the opposite. Here, we propose a new approach to characterize the similarity of the decision strategies of two observers (models or brains) using decision variable correlation (DVC). DVC quantifies the image-by-image correlation between the decoded decisions based on the internal neural representations in a classification task. Thus, it can capture task-relevant information rather than general representational alignment. We evaluate DVC using monkey V4/IT recordings and network models trained on image classification tasks. We find that model-model similarity is comparable to monkey-monkey similarity, whereas model-monkey similarity is consistently lower. Strikingly, DVC decreases with increasing network performance on ImageNet-1k. Adversarial training does not improve model-monkey similarity in task-relevant dimensions assessed using DVC, although it markedly increases the model-model similarity. Similarly, pre-training on larger datasets does not improve model-monkey similarity. These results suggest a divergence between the task-relevant representations in monkey V4/IT and those learned by models trained on image classification tasks.

Research profiles highlight scientists' research focus, enabling talent discovery and collaborations, but are often outdated. Automated, scalable methods are urgently needed to keep profiles current. We design and evaluate two Large Language Models (LLMs)-based methods to generate scientific interest profiles--one summarizing PubMed abstracts and the other using Medical Subject Headings (MeSH) terms--comparing them with researchers' self-summarized interests. We collected titles, MeSH terms, and abstracts of PubMed publications for 595 faculty at Columbia University Irving Medical Center, obtaining human-written profiles for 167. GPT-4o-mini was prompted to summarize each researcher's interests. Manual and automated evaluations characterized similarities between machine-generated and self-written profiles. The similarity study showed low ROUGE-L, BLEU, and METEOR scores, reflecting little terminological overlap. BERTScore analysis revealed moderate semantic similarity (F1: 0.542 for MeSH-based, 0.555 for abstract-based), despite low lexical overlap. In validation, paraphrased summaries achieved a higher F1 of 0.851. Comparing original and manually paraphrased summaries indicated limitations of such metrics. Kullback-Leibler (KL) Divergence of TF-IDF values (8.56 for MeSH-based, 8.58 for abstract-based) suggests machine summaries employ different keywords than human-written ones. Manual reviews showed 77.78% rated MeSH-based profiling "good" or "excellent," with readability rated favorably in 93.44% of cases, though granularity and accuracy varied. Panel reviews favored 67.86% of MeSH-derived profiles over abstract-derived ones. LLMs promise to automate scientific interest profiling at scale. MeSH-derived profiles have better readability than abstract-derived ones. Machine-generated summaries differ from human-written ones in concept choice, with the latter initiating more novel ideas.

Smartphone-based tele-dermatology assumes that colorimetric calibration ensures clinical reliability, yet this remains untested for underrepresented skin phototypes. We investigated whether standard calibration translates to reliable clinical biomarkers using 43,425 images from 965 Korean subjects (Fitzpatrick III-IV) across DSLR, tablet, and smartphone devices. While Linear Color Correction Matrix (CCM) normalization reduced color error by 67-77% -- achieving near-clinical accuracy (Delta E < 2.3) -- this success did not translate to biomarker reliability. We identify a phenomenon termed "color-clinical decoupling": despite perceptual accuracy, the Individual Typology Angle (ITA) showed poor inter-device agreement (ICC = 0.40), while the Melanin Index achieved good agreement (ICC = 0.77). This decoupling is driven by the ITA formula's sensitivity to b* channel noise and is further compounded by anatomical variance. Facial region accounts for 25.2% of color variance -- 3.6x greater than device effects (7.0%) -- challenging the efficacy of single-patch calibration. Our results demonstrate that current colorimetric standards are insufficient for clinical-grade biomarker extraction, necessitating region-aware protocols for mobile dermatology.

We introduce a nonparametric model for time-evolving, unobserved probability distributions from discrete-time data consisting of unlabelled partitions. The latent process is a two-parameter Poisson-Dirichlet diffusion, and observations arise via exchangeable sampling. Applications include social and genetic data where only aggregate clustering summaries are observed. To address the intractable likelihood, we develop a tractable inferential framework that avoids label enumeration and direct simulation of the latent state. We exploit a duality between the diffusion and a pure-death process on partitions, together with coagulation operators that encode the effect of new data. These yield closed-form, recursive updates for forward and backward inference. We compute exact posterior distributions of the latent state at arbitrary times and predictive distributions of future or interpolated partitions. This enables online and offline inference and forecasting with full uncertainty quantification, bypassing MCMC and sequential Monte Carlo. Compared to particle filtering, our method achieves higher accuracy, lower variance, and substantial computational gains. We illustrate the methodology with synthetic experiments and a social network application, recovering interpretable patterns in time-varying heterozygosity.