Papers

To be conscious is to have an experience - not a collection of phenomenal atoms, but a structured whole composed of distinctions and the relations that bind them. Integrated Information Theory (IIT) identifies the essential properties of every experience (axioms), formulates them operationally as postulates that a substrate must satisfy, and unfolds the cause-effect power of the resulting complex into a $\Phi$-structure. Accounting for a content of experience is then a matter of identifying the phenomenal distinctions and relations that compose it, and showing them reflected one-to-one in the causal distinctions and relations of the corresponding $\Phi$-structure. We apply that method to three pervasive contents whose structure is partly open to introspection. Spatial extendedness is composed of spots, whose elemental signature is reflexivity, bound by reflexive inclusion, connection, and fusion; temporal flow is composed of moments, whose signature is directedness, bound by directed inclusion, connection, and fusion; objects bind a particular configuration of features to a general concept through relations bearing the signature of hierarchy. The endeavor is akin to chemistry, which accounts for an endless variety of compounds from a limited set of elements and the ways they bond. We assess these accounts against seven criteria of a good explanation - scope, synthesis, specificity, self-consistency, system consistency, simplicity, and scientific validation - and sketch the prediction that follows: altering the structure specified by a substrate should alter the corresponding content, even when activity and behavior are held comparable. The same method may reach narrow qualia, which resist introspection, and the compound contents that bind many qualia together, though there it remains a proof of concept and an open research program.

Omics datasets, particularly single-cell RNA sequencing data, are high-dimensional, sparse, noisy, and dominated by zero values, making faithful low-dimensional representation challenging. Existing dimensionality-reduction methods may distort local neighbourhoods, global organization, or the cohesion of meaningful populations, with similar limitations arising in genealogical data. We introduce Contrastive Manifold Approximation and Projection (CosMAP), a graph-based unsupervised dimensionality-reduction method for producing faithful and interpretable embeddings. CosMAP extends the graph-based framework of UMAP by combining cosine-similarity neighbourhoods with temperature-normalized contrastive affinities, which are optimized in the embedding space using an attractive–repulsive objective. It further employs a two-phase refinement strategy: an intermediate higher-dimensional representation is first learned and then used to reconstruct the neighbourhood graph and initialize the final low-dimensional embedding. We evaluate CosMAP on MNIST and USPS handwritten-digit datasets, mouse retina and cortex single-cell RNA-sequencing datasets, and a large genealogical kinship dataset derived from BALSAC-CARTaGENE. Compared with state-of-the-art dimensionality-reduction methods, CosMAP produces more coherent visual representations, improves neighbourhood preservation, and provides clearer global organization of digit classes, biological cell populations, and regional genealogical patterns. These results indicate that CosMAP offers a robust framework for exploratory analysis of complex, sparse, high-dimensional data. The implementation is publicly available at https://github.com/FenosoaRandrianjatovo/CosMAP-dr.

Predicting how molecular changes affect large-scale brain activity is a difficult task because of the lack of appropriate methods to link scales. In this perspective, we review a class of mean-field models that can integrate biophysical details such as synaptic receptors or membrane ion channels. This leads to a multi-scale modeling approach that can be used to evaluate how microscopic changes can impact macroscopic brain activity. This approach is illustrated here for the case of anesthesia, where changes at the level of specific synaptic receptors can lead to a global change in brain activity and a disconnection from external inputs. This is only possible using mean-field models that can include enough detail about the microscopic biophysical properties. This biophysically-based mean-field approach could be generalized to study cellular or molecular origins of brain diseases, or to better understand how drugs acting at microscopic scales can influence global brain activity. Biophysical mean-field models also link different fields of neuroscience, from molecular studies to brain imaging.

Active-quiescent switching occurs in biological populations in which growth is confined to a proliferative active state, while cells may reversibly enter a nonproliferative quiescent state. Experiments often observe only part of this process, through active-state markers, aggregate population measurements, or aggregate data supplemented by sparse active-state observations. Such measurements arise when marker panels are limited, active-state assays require fixation or endpoint sampling, or only total cell number, optical density, tumor burden, or aggregate fluorescence is reported. These observation choices complicate sparse regression methods such as sparse identification of nonlinear dynamics (SINDy), because the measured variable may satisfy a different equation from the underlying active-quiescent system. A library chosen for the wrong observable may therefore fit a trajectory without preserving mechanistic interpretation or transferability. We study this issue using a two-compartment ordinary differential equation model. We define observable reduction as the elimination of hidden states to obtain the differential equation satisfied by the measured variable. For several biologically relevant growth laws, we derive observation-specific reductions and use them to construct sparse-regression libraries. Using synthetic data, we compare these structured libraries with standard polynomial SINDy. Polynomial libraries can match training trajectories while failing coefficient-relation and transfer tests, whereas reduction-guided libraries recover interpretable coefficient maps when the observed regime is informative. These results show that interpretable equation learning in hidden-compartment systems requires matching both the regression target and candidate library to the observation process.

Reproductive hormones regulate a wide range of physiological processes throughout the human lifespan. Estrogen, in particular, varies substantially across the menstrual cycle and is widely used in contraceptives and hormone replacement therapies. Despite its physiological importance, few experimental studies and even fewer mathematical models explicitly investigate how estrogen regulates smooth muscle function. As smooth muscle lines our blood vessels, airways, uterus, and several other organs, understanding how estrogen impacts smooth muscle is important to improving the understanding of sex differences in lifelong health. Here we extend an established mathematical model of smooth muscle cell calcium signalling to incorporate estrogen-dependent modulation of intracellular calcium transport pathways. Numerical simulations, global sensitivity analysis, and numerical bifurcation analysis are then used to quantify the influence of estrogen on intracellular calcium dynamics and the resulting steady-state and oscillatory behaviours. Our results demonstrate that physiologically relevant changes in estrogen shift intracellular calcium concentrations while leaving the underlying bifurcation structure and qualitative dynamics largely unchanged, suggesting that estrogen acts primarily as a quantitative modulator of smooth muscle calcium signalling. This work also serves to establish a foundation for future mechanistic models of hormone-dependent cell physiology.

Neural tracking - the time-locking of neural responses to continuous stimuli such as speech, music, and video - is widely used to study how the brain processes natural input. Tracking strength is typically quantified as the correlation between the recorded neural response and the stimulus, decoded and/or encoded through data-driven models, and this correlation is routinely used to compare stimulus features, models, or settings. However, its magnitude depends not only on how strongly the brain tracks the stimulus, but also on the statistical properties of the signals being correlated. For example, a smallband speech envelope carrying almost no information about speech content yields among the highest correlations, simply because it is easier to reconstruct. Meaningful interpretation therefore requires comparing each correlation to its null distribution: the correlations expected without any stimulus-response relationship. We show that the randomization procedures commonly used to construct this null distribution are not interchangeable: each implicitly encodes a different null hypothesis, and we motivate stimulus-response misalignment as the most practical and appropriate choice. Because reliable null distributions require many permutations, we introduce a semi-parametric model using the normal distribution after the Fisher transform that yields accurate significance levels from only 3-5 min of data and predicts them across analysis window lengths. Building on this, we propose the null-normalized tracking score, an interpretable measure placing features and models on a common scale, which relates directly to the widely used match-mismatch accuracy. Applied to EEG from 121 participants listening to continuous speech, the framework reverses conclusions drawn from raw correlations, providing an efficient and principled methodology for interpreting neural tracking correlations.

An important question in pathology AI is how to allocate limited analytical resources between broad coverage of a whole slide image (WSI) and detailed analysis of selected regions. We compared a minimal WSI-AI model designed for broad search with Diagnostic Field Selection AI (DFS-AI), which uses a coarse overview to select diagnostically relevant fields for detailed analysis. Three minimal models introduced localization information, contextual discrimination, and spatial structure in sequence. In Model I, WSI-AI was favored when coarse localization was uninformative or costly. When coarse observation provided useful ranking information, DFS-AI was favored over an intermediate range of resources; WSI-AI again became preferable when resources allowed nearly complete fine observation. In Model II, background heterogeneity generated distractors. In a WSI control given the same local context as DFS-AI, most of the original DFS advantage was explained by local background correction, while coarse candidate selection provided a smaller additional benefit. In Model III, larger contiguous lesions were easier for WSI-AI to discover but more difficult to characterize completely with a finite budget for fine observation. The DFS advantage for complete characterization persisted when the comparator used the same field sized unit of fine analysis and when success was defined symmetrically as 100%, 90%, or 80% lesion unit coverage. Together, the models support a framework in which the preferred strategy changes with available resources and depends on the information available for selection, its acquisition cost, lesion structure, and the diagnostic endpoint. This framework clarifies when limited pathology AI resources should be devoted to broad WSI coverage and when they should be concentrated on selected Diagnostic Fields.

A subcritical age-structured branching population dies out almost surely. Conditioned on survival, it converges to its Yaglom limit, a quasi-stationary equilibrium that we take as the operating point for control. We model preventive removal (culling) as an age-dependent actuator that raises the mortality rate and leaves the offspring law untouched, and we show that its authority over this equilibrium is bounded for structural reasons. Two facts drive the result. First, the input is matched to the killing rate but unmatched with respect to the Foster–Lyapunov drift, so the transmission barrier $\Lbar$ set by reproduction alone is invariant under such actuation. Second, and this does not follow from invariance alone,} the supremum of the reachable decay rates is $\Lbar+\nu^\star$, where $\nu^\star\le0$ is the Malthusian parameter of the lineage conditioned never to die childless; the gap $|\nu^\star|$ is given in closed form and vanishes exactly when no individual has two or more offspring. Consequently no removal law of this class reaches the barrier, and along the admissibility boundary the achievable decay rate is governed by the shape of the actuator rather than by its size. We illustrate these results on a model calibrated to the 2001 Cumbrian foot-and-mouth outbreak.

Protein backbone generation models are often credited with exploring novel fold space based solely on low full-chain similarity to known proteins, yet this cannot distinguish a genuinely new fold from a novel assembly of known structural units. We first ask whether this granularity mismatch alone explains the reported rates, and introduce the Domain Retrieval Rate (DRR), the fraction of generated backbones for which any constituent domain matches a known domain in CATH S40. Applied to eight backbone generation models spanning diffusion and flow-matching paradigms, DRR finds locally alignable known structure in most outputs, while the fraction containing a substantially covered complete domain is considerably smaller and depends on the scoring convention. To calibrate what retrieval alone can achieve, we propose RetFold, a zero-training baseline that constructs backbones by retrieving CATH domains and refining inter-domain connections through geometry-based helix-linker optimization, at two orders of magnitude lower cost on CPU alone.

Proteolysis-targeting chimeras (PROTACs) induce protein degradation by recruiting a target protein to an E3 ubiquitin ligase, making degradation a joint outcome of the degrader molecule and its biological context. Although public databases contain thousands of structured molecule-target-E3 records, degradation measurements are available for only a small fraction of them. Existing supervised approaches therefore leave most recorded chemical-biological relationships unused. We introduce DegradeQuery, a context-aware prediction framework that converts these label-missing records into a pretraining signal. Its counterfactual tuple pretraining objective contrasts recorded tuples with alternatives formed by replacing the target, the E3 ligase, or both, enabling the model to learn contextual associations without assigning activity pseudo-labels. The resulting representation is then fine-tuned to predict degradation from the complete molecule-target-E3 context. On the official PROTAC-8K benchmark, DegradeQuery achieves an area under the receiver operating characteristic curve of 0.9065 and an accuracy of 0.8500, outperforming the compared methods. Controlled analyses further show that the improvement is primarily attributable to tuple-level pretraining, can be recovered using only label-missing records, and remains complementary to protein language model representations. These findings demonstrate that incompletely labeled PROTAC databases contain useful relational supervision and provide a practical route for learning context-aware degradation predictors from scarce experimental labels.

How many distinguishable labels can a biological oscillator carry? Proposals invoking collective vibrational modes, endogenous electromagnetic fields, microtubule excitations and oscillatory phase codes are each debated on grounds particular to themselves, with no shared standard for comparison. We show that spectral distinguishability alone bounds the number of labels by the quality factor, M <= Q = 2 pi nu tau. This follows from the relation between linewidth and coherence time, so it is independent of substrate, of mechanism, and of any position on quantum effects in biology, and it can be evaluated from two published quantities. Applied to a recently proposed 30 GHz intracolumnar microwave field in cortex, it gives Q = 0.19: the linewidth exceeds the carrier five-fold. The obvious rescue, that a driven emitter can be spectrally narrower than its gain medium, requires a resonant cavity, and the model's own geometry forbids one. An independent bound on metabolic power is exceeded by five to nine orders of magnitude. Six further criteria follow from the same standpoint, including a two-sided persistence window requiring a label to be both readable and rewritable. Screening eleven carriers, only the low-frequency neural rhythms pass. High-frequency molecular carriers are eliminated by brevity, not by the fragility the debate has assumed.

Aquaporins (AQPs) are an ancient group of channels that arose early in phylogeny and radiated throughout the kingdoms of life. Classified as members of the broad Membrane Intrinsic Protein family, AQPs originally were envisioned as strictly water pores that were constitutively open. International research interest is driving new views of AQPs as multi-functional channels with complex control mechanisms that enable adaptive responses to physiological challenges. Expanding numbers of AQP classes are being discovered to allow permeation of diverse solutes in parallel to the archetypal substrate water, with channel activities that show subtype-specific regulation by intracellular and extracellular signals. To date, at least fifteen classes of AQPs from mammals, plants, insects and algae have been proposed to carry ions, enabling amino acid sequence comparisons that suggest ion channel functional domains arise from conserved transmembrane, loop and teminal regions of the tetrameric AQP protein, serving as selectivity filters, gates, and sites for modulation. More AQP ion channel classes likely await discovery, pending identification of relevant activators. Subtype-specific control of dual water-and-ion AQP classes orchestrate diverse roles for AQPs across the domains of life, promoting homeostasis, cell motility, nutrient acquisition, redox protection, sensory detection, and intercellular signaling in cells and tissues.

Characterizing the dissimilarity of neural representations between experimental conditions, and tracking it across time, is a central goal of multivariate pattern analysis. Guggenmos et al. (2018) assessed the reliability of many of the measures that can be used for that purpose on MEG data and recommended the use of either the cross-validated Euclidean distance or the within-class-corrected Pearson distance. In this commentary, we show that we can improve upon these distances. First, we show that the cross-validated Euclidean distance is equivalent to a sum of between-partition distances and that this equivalence can be leveraged to obtain a generalized variant, with increased reliability and accuracy. Second, we use the relationship between Euclidean distance and Pearson correlation to define a cross-validated correlation distance in a similar way. The resulting distance is more accurate and interpretable than a formulation proposed by Guggenmos and colleagues. Finally, we discuss the relationship between our generalized cross-validation and within-class correction, another strategy often used to increase reliability, and we show that generalized cross-validation results in higher accuracy for the correlation distance.

Computational head models are typically applied to isolated impacts, leaving repeated head loading largely unexplored. An Ogden-Roxburgh Mullins damage formulation was implemented in a high-fidelity finite element head model to represent loading-history-dependent softening during cyclic brain-tissue deformation. Repeated-loading histories derived from mixed martial arts head-impact data were applied and compared with damage-free hyperelastic (HE) and linear visco-hyperelastic (LVHE) model variants. Under five identical single-axis cycles, Mullins-type softening progressively increased strain and strain rate metrics relative to the HE model. Mullins-based injury probabilities progressively exceeded strain-based HE predictions and diverged from unchanged kinematics-based predictions, indicating that neglecting prior softening may underestimate injury risk. In a randomized twenty-cycle multiaxial sequence, cycles of similar kinematic intensity produced different deformation and injury-risk estimates depending on prior softening. HE and LVHE models predicted higher injury probabilities initially, whereas the Mullins-based model produced the largest later-cycle estimates and highest probability of at least one injury over the sequence. Regional amplification depended on loading direction and prior softening, with no direction-independent trend among brain substructures. Gyral elements exhibited higher cumulative maximum principal strain than sulcal elements, which showed greater amplification relative to initial responses. These findings demonstrate that short-term damage-mediated softening can substantially amplify tissue deformation and injury-risk estimates beyond damage-free head models under the same loading histories. Further experimental characterization of cyclic brain-tissue softening is needed to improve models of repeated head loading and traumatic brain injury.

Background: Brain information processing deteriorates as cortical neurons gradually transition from a rested state into fatigue. Subjectively, fatigue is experienced as a state of weariness, tiredness, or lack of energy that reduces the ability to work safely and effectively. Objective: In this theoretical paper, we pinpoint the physical origin of brain fatigue in the gradual deterioration of biochemical reaction quotients and transmembrane ion concentration gradients, which increase neuronal excitability and decrease the signal-to-noise ratio in the brain cortex. Methods: Brain performance in a rested state versus fatigue is examined by well-established, data-driven computer models for energy transport inside protein $\alpha$-helices in the presence of thermal noise for different ATP energy states or for pyramidal neuron firing of action potentials under electric stimulation in a rested membrane state versus fatigue. Results: We found that reduced Gibbs free energy supply from ATP hydrolysis impairs the cooperative effect between amide I excitons propagating inside protein $\alpha$-helices, with resulting decreased thermal stability of molecular solitons. Concurrent changes in Nernst reversal potentials for Na+ or K+ ions further led to neuronal hyperexcitability and a higher risk of neuronal depolarization block during mental fatigue. Conclusions: Detailed computational modeling showed that inefficient protein function due to diminished ATP energy status, alters the electrophysiological properties of individual neurons, thereby impairing their information processing capacity for the proper execution of cognitive tasks. Scheduling practices aimed at intermittent recovery of the rested brain state during intellectually challenging work could protect physical and mental wellbeing, prevent burnout, and enhance long-term productivity.

Graph-theoretical analysis offers a principled framework for quantifying emergent dynamics in neuronal cultures. Here, we present an analytical pipeline for inferring network-level properties of in vitro cortical cultures from multichannel electrophysiological recordings. The approach builds on a recently proposed Reservoir Computing (RC) framework (Auslender et al., 2025), which enables direct extraction of an Intrinsic Connectivity Map (ICM) from neural activity. We interpret the ICM as an effective adjacency matrix and apply graph-theoretic centrality measures to quantify node- and edge-level contributions to the culture's collective dynamics. We systematically evaluate both local and global graph metrics and examine their relationships with experimentally measured activity features, including firing rates and network-level descriptors. To validate the inference procedure, we also simulate the experimental environment, enabling controlled benchmarking of the RC-derived connectivity against a known ground-truth adjacency matrix and assessment of model performance as a function of graph structure. Our results demonstrate statistically robust associations, of varying strength, between graph-theoretic measures and experimentally observed activity patterns. These findings additionally support the validity of the RC-based connectivity inference and establish a scalable, data-driven framework for functional network characterization in neuronal culture systems.

Cardiac digital twin research is moving from subject-specific anatomical replicas toward virtual cohorts that represent clinically relevant population subgroups. Yet access to representative imaging-derived anatomy datasets remains limited by cohort size, subgroup sparsity, and data-sharing constraints. Conditional generative models could help address this gap, but virtual cohorts are useful only if they preserve realistic, metadata-dependent anatomical variability. Existing cardiac anatomy generators largely rely on conditional variational autoencoders (cVAEs), which couple representation learning and metadata conditioning through a shared regularized latent prior. We introduce CAN-FLOW, a two-step Conditional ANatomy generation framework based on normalizing FLOWs that first learns geometry-only latent representations of diffeomorphic cardiac shape momenta and then models their sex-, age-, and body-mass-index-dependent distribution with a conditional normalizing flow. We trained CAN-FLOW on 2,208 healthy UK Biobank subjects and compared it with cVAEs across regularization strengths. CAN-FLOW generated plausible stochastic biventricular anatomies that better reproduced clinical phenotype distributions, metadata-dependent trends, subgroup variability, point-cloud coverage, and high-dimensional shape variability. Together, these results establish CAN-FLOW as a shareable framework for generating realistic, stochastically varying, metadata-conditioned biventricular anatomies for virtual cohort construction and in silico clinical trial workflows.

Here, we present an overview of how archaeogenomic data can be used to investigate past mobilities and how it can be integrated with archaeological research to reconstruct models of human mobility at different scales. We first seek to explain the importance of a clear terminological and theoretical approach to mobility, discussing terms such as migration, admixture, ancestry, and replacement. We then describe the theoretical principles and state-of-the-art tools for inferring past mobility from ancient DNA data, including exploratory approaches and formal modeling, and the relevance of using diverse lines of evidence including allele-frequency patterns, haplotype-sharing, genetic relatedness, and uniparental markers. We further discuss the power and limitations of these methods for inferring different modes of mobility, such as large-scale and long-distance mobility, which we refer to as migration, and also small-scale and short-distance movements, which we argue constituted the bulk of past human mobility and may have had the major share in shaping cultural landscapes. We also describe inference of sex-biased mobility events. Finally, we present examples of joint analyses of genomic data with archaeological, bioarchaeological and historical evidence, demonstrating the potential of interdisciplinary analyses in building comprehensive models of past mobilities. We finish by calling for further integration across fields.

Biological systems not only infer states of the external world from sensory input but also vary the expression of their responses according to the information they have acquired. Here, without altering the standard inferential and policy-evaluation schemes of the Free Energy Principle (FEP) and active inference, we propose a minimal dynamics that links the mutual information formed between an external state and an internal representation through inference to a response-expression variable distinct from policy selection. In this formulation, established information is positioned as a state signal that modulates response expression. The model introduces a piecewise nonlinear term in which response activation is driven only when mutual information exceeds an information threshold. Monte Carlo simulations using a two-state Markov environment showed that increasing observation accuracy increased the mutual information between the external state and the internal representation and, in turn, increased response activation. By contrast, in a control condition without information-response coupling, the response remained at its baseline level. The basic pattern of response activation was preserved when observation accuracy, the information-response coupling coefficient, the information threshold, and the closed-loop coefficient linking response to sensory sampling were varied. These results show that coupling information formed by FEP-consistent inference to an independent response-expression dynamics can generate a nonlinear response that depends on an information threshold.

Inverse decision modeling infers latent properties of decision processes from observed behavior, but existing formulations rely primarily on action trajectories. In verbalized cognitive tasks, task execution also produces response dynamics that action-only formulations leave unmodeled, such as verbal production, interaction, and hesitation. We propose a factorized inverse decision model (FIDM) that decomposes each individual's task-execution likelihood into an action factor and an effort factor, governed by separate individual-specific parameters. From raw verbal transcripts, a language model produces structured task-execution traces for factorized inference. On data from 400 older adults performing a grocery-shopping dialog task for cognitive screening, controlled recovery shows selective estimation of the intended factors, while matched semi-synthetic conditions show that FIDM preserves action-execution distinctions even when aggregate behavioral summaries are matched. Action evidence further localizes task-defined deviations across participants. In cognitive-status classification, FIDM provides information complementary to clinical scores, trajectory summaries, and frozen language representations, with consistent gains across all evaluated baselines in the binary setting.

Many computational problems possess the following three features: (1) the problem could be solved efficiently if a linear operator could be diagonalized, (2) direct diagonalization is infeasible due to scale, but, (3) the operator is known to commute with a permutation since the problem is (a) symmetric with respect to a change of coordinates, or (b) the operator is block circulant. For example, high-resolution template matching demands repeated convolution of large, high resolution images with an exhaustive list of related templates. The associated linear operator could be compressed, via a low rank approximation, if diagonalized. The required decomposition expensive, but, the entire problem is symmetric to in-plane rotations. In these cases, the eigenfunctions are constrained by the symmetry. These constraints allow efficient decomposition. We illustrate a parallelized algorithm that allows fast diagonalization of any block-circulant matrix. When the index space can be partitioned into $l$ classes of $m$ interchangeable elements, the algorithm reduces storage costs by a factor of $m$ and, if $w$ workers are available, the algorithm demands $\mathcal{O}(l^2m\log(m)/w) + \mathcal{O}(ml^3/w)$ floating point operations per worker yielding a $m^2$ speedup. We use this procedure to decompose a high precision template matching matrix. We compare runtime on a reduced problem, where the fast procedure ran 205 times faster per recovered feature, for 22.5 times as many features. On a typical template matching matrix, the decomposition required 25 times less time than computing the matrix entries. We also demonstrate decomposition of a $\sim$30 times larger matrix that covers the complete set of possible projections represented in a cryo-EM image at 2~Å resolution in 14 minutes. This procedure is more stable and allows recovery of in-plane rotations to arbitrary precision.

Tree Search-based test-time scaling of LLMs is a powerful tool for automated scientific coding. However, pure Tree Search sometimes struggles with systematic exploration, becoming trapped in local optima, or unproductive loops, especially in the vast search space of scientific methods. To address this limitation, we propose Idea Search, a framework that systematically integrates a dynamic "Idea Bank" into Tree Search. Idea Search involves three steps: (1) decomposing existing methods into atomic ideas, (2) sampling from this bank of ideas to guide branches of code mutations, and (3) dynamically updating the bank with new ideas discovered through execution. On single-cell RNA-sequencing (scRNA-seq) batch integration, Idea Search reliably breaks the plateau of a strong pure Tree Search baseline, improving the mean score from 0.678 to 0.697 and reaching a best score of 0.728. We then characterize which design choices drive these gains: bank augmentation helps bandit sampling but not random sampling, "Exploratory" prompting that prioritizes new ideas surfaces the rare best-performing solutions, while increasing sampling-level exploration is counterproductive.

Gene expression is an intrinsically stochastic process that generates phenotypic heterogeneity within genetically identical cell populations. While the exact statistical moments of the protein count can be obtained for a broad range of complex models, the corresponding distributions are significantly harder to obtain and intractable in many cases. The classical three-stage model of gene expression, which predicts fluctuations in protein levels as a function of promoter switching, transcription, translation, and degradation events all occurring with linear propensities, illustrates this perfectly; deriving its exact protein distribution remains elusive. Here, using the partitioning property of time-inhomogeneous Poisson processes, we develop an exact mapping of the three-stage model onto a simplified model. The simplified model allows us to formulate two analytical approximations for the full protein distribution of the three-stage model based on a beta-mixture representation of the exact solution for a simpler model. We show that the two approximations are asymptotically exact in different limiting cases and verify their accuracy against simulations for a broad range of parameters. Although approximate, these are the first analytical expressions for protein distributions for the three-stage model that are highly accurate in intermediate regimes.

Cells often estimate chemical concentrations from only a few diffusing molecules, whose stochastic motion limits sensing precision. We consider a perfect monitoring instrument that records the instantaneous molecular density throughout a finite region without perturbing the concentration field or distinguishing molecular identities. The standard Berg-Purcell estimator weights all positions uniformly. Here we derive the optimal spatial weighting within a large class of unbiased estimators for both concentration and gradient sensing. Variance minimization maps to an electrostatic problem. Surprisingly, although the instrument monitors the entire volume, the optimal estimator assigns all weight to its boundary, reducing the uncertainty in both concentration and gradient sensing. We extend the results to arbitrary geometries and spatial dimensions.

Biological recognition rarely rests on one strong bond. It works by forming many weak ones at once, between crowded, deformable surfaces in water. This review develops that process as a problem in statistical mechanics. Counting the ways two multivalent objects can bind proves to be the classical monomer-dimer problem on a graph, with a rigorous consequence: the apparent switching of multivalent binding is always a smooth crossover, never a phase transition. Three constraints follow. A repulsive surface layer is obligatory rather than a design choice; bonds do not act independently; and since free energies enter rates exponentially, small changes in receptor number shift binding lifetimes by orders of magnitude. One set of equations then covers antibodies, lipoproteins, and T cell recognition. In each, what decides the outcome is not the strength of any single bond but how a fixed total is spread over many: affinity is a property of a molecule, selectivity a property of an assembly.

In a preceding paper, we used the graph-theoretic linear framework to show how transient properties of continuous-time Markov processes – splitting probabilities and the moments of first-passage time (FPT) distributions – could be expressed as rational algebraic functions of the transition rates, by using spanning forests of the underlying graph. This contrasts with the related rational formulas for steady-state (s.s.) probabilities, which use only spanning trees. The biophysicist Terrell Hill sketched a procedure for calculating mean FPTs and splitting probabilities in terms of the s.s. probabilities of a modified Markov process, thereby converting calculations using ensembles of trajectories to those using a single trajectory. Similarly, Mark Kac showed that the mean recurrence time to a state of a Markov process could be expressed in terms of the s.s. probability of that state. Here, we explore further the relationships between transient and s.s. properties, forests and trees, and ensemble and single-trajectory calculations. We formalise Hill's procedure by introducing a Hill operator, $H_u[G]$, on a graph, $G$, and use it to calculate all moments of the conditional and unconditional FPTs from $u$ as rational functions of the s.s. probabilities of $H_u[G]$, which arise from trees and "exchange factors" which arise from forests. We then combine this with an unravelling operator, $U_v[G]$, to calculate all moments of the recurrence time distribution as rational functions of the s.s. probabilities of $G$ and related exchange factors. Surprisingly, the Hill operator turns out to be a left-inverse to the unravelling operator, suggesting that the algebra of operators on linear framework graphs may be of broader interest. Our results integrate and generalise previously disparate findings into a common repertoire of rational algebraic formulas for FPTs of Markov processes.

We propose IM-LEPP (Integrated Multimodal Latent Energy-based Predictive Processing), a hierarchical, energy-based model of multimodal cognition that extends a previously proposed single-modality model (LEPP) to integrate vision and language. Following the view that generative neural networks are effective theories of cognitive dynamics, analogous to how statistical mechanics relates to thermodynamics, IM-LEPP models cognition as latent states flowing through learned energy landscapes rather than as an account of neural circuitry. The architecture is a hub-and-spoke hierarchy, grounded in the controlled semantic cognition framework of Lambon Ralph et al., in which predictive-coding pipelines for visual objects, scenes, and linguistic units converge on a shared amodal hub modeled on the anterior temporal lobe. Each pipeline's own prediction is conditioned by, rather than overwritten by, the current hub state, preserving pipeline-specific identity while letting every prediction reflect the full multimodal context. We show this architecture gives a mechanistic account of attentional phenomena such as inattentional blindness and Necker-cube bistability, and that its structure recovers or motivates independently established findings in psycholinguistics, including surprisal theory, the N400/P600 ERP components, and garden-path reanalysis, alongside a falsifiable contrast with transformer language models on trajectory-sensitivity in next-word prediction. We also discuss data-efficient language acquisition relative to LLMs, outline a semantic/episodic memory subsystem, situate the model against predictive coding, the free-energy principle, JEPA, and Hierarchical Temporal Memory, and propose concrete experimental predictions to test its central claims Key Words: predictive processing; predictive coding; energy-based models; diffusion models; effective theory; computational neuroscience.

Tumor progression is an inherently multiphysical phenomenon in which interstitial fluid dynamics, biochemical transport, and cellular mechanics interact across multiple spatiotemporal scales. Classical mesh-based solvers, although accurate, impose prohibitive computational costs for the repeated evaluations demanded by inverse parameter identification and future patient-specific predictive pipelines. In this work we introduce a Physics-Informed Neural Network (PINN) framework for a tractable chemo-fluidic continuum model of tumor growth that couples an advection-diffusion-reaction (ADR) equation for the tumor volume fraction with a quasi-static Darcy pressure equation for the interstitial fluid pressure. By intentionally decoupling the solid-mechanical equilibrium, we obtain a three-equation system whose gradient structure is stable under automatic differentiation, enabling robust deep-learning optimization. The network simultaneously learns both state variables from physics constraints alone (forward problem) and recovers hidden transport parameters from sparse, noisy synthetic measurements (Data-Assimilation PINN, DA-PINN, inverse problem). We verify the forward solver against a high-resolution finite-difference (FD) reference, achieving a mean absolute error below 0.002. For the inverse problem, starting from an initial permeability estimate of 0.08 (a factor of 4x above the true value of 0.02) with only 5% spatially sparse observations corrupted by 5% Gaussian noise, the DA-PINN recovers the permeability with a relative error below 5%. These results demonstrate that physics-informed deep learning constitutes a viable, computationally efficient route to multiphysics oncology modeling and lays the mathematical groundwork for future integration into clinical data assimilation pipelines.

Continuous hemodynamic monitoring guides treatment decisions in surgery and intensive care. However, gold-standard signals are only measured in severe cases due to risks associated with invasive measurement. In this work, we introduce HIPNO (Hemodynamic Inference via Physics-informed Neural Operators) to recover hemodynamic state from ubiquitous, non-invasive signals and expand access to advanced monitoring. HIPNO addresses a problem of scale symmetry in physics-informed hemodynamic inference, where different combinations of flow, resistance, and compliance can generate the same observed pressure. We identify the symmetry group of the observation model and parameterize the network in its quotient space. For the 3-element Windkessel model, the quotient coordinates are the compliance-normalized flow $U=Q/C$, the decay time constant $\tau_{WK}=R_2 C$, and the characteristic-impedance coordinate $\kappa=R_1 C$. Across 945499 intraoperative windows from 2562 patients, HIPNO predicts $\tau_{wave}$, a proxy for vascular decay derived from pressure, with 32% lower error on the log scale than a population baseline while preserving mean arterial pressure accuracy. Because vascular decay and flow drive occupy separate coordinates, counterfactual perturbations produce the expected directional responses in at least 90% of windows in almost all prespecified scenarios, a separation unavailable to pressure-only baselines. The coordinates are also used as inputs to a calibration model for monitored cardiac output. Finally, the formulation identifies the external compliance or flow reference required to recover absolute physical scale.

Spatial transcriptomics can resolve gene expression at single-cell resolution, but it is costly, limited to targeted panels of a few hundred to a few thousand genes, and applicable to only a small number of samples. H&E imaging, by contrast, is cheap and collected routinely at scale. This makes predicting single-cell expression directly from morphology a practical way to bring molecular analysis to large tissue archives. We therefore present VOICE, a multimodal foundation model that predicts single-cell gene expression from H&E images using paired Xenium data. VOICE first aligns cell centered H&E morphology from a pathology foundation model with single-cell expression embeddings from a transcriptome foundation model, trained using contrastive learning over 23 million cells. Next it predicts expression through two branches. One branch directly regresses expression from morphology. The other branch retrieves measured expression from similar reference cells, recovering genes that do not have morphological signal. Because genes vary in morphological predictability, VOICE fuses the two branches with a per-gene weight. After training, VOICE generalizes to heldout patients, slides, and partially overlapping gene panels from Xenium, and it consistently outperforms prior single-cell expression prediction methods on seven metrics.