Visual decoding from brain signals is a key challenge at the intersection of computer vision and neuroscience, requiring methods that bridge neural representations and computational models of vision. A field-wide goal is to achieve generalizable, cross-subject models. A major obstacle towards this goal is the substantial variability in neural representations across individuals, which has so far required training bespoke models or fine-tuning separately for each subject. To address this challenge, we introduce a meta-optimized approach for semantic visual decoding from fMRI that generalizes to novel subjects without any fine-tuning. By simply conditioning on a small set of image-brain activation examples from the new individual, our model rapidly infers their unique neural encoding patterns to facilitate robust and efficient visual decoding. Our approach is explicitly optimized for in-context learning of the new subject's encoding model and performs decoding by hierarchical inference, inverting the encoder. First, for multiple brain regions, we estimate the per-voxel visual response encoder parameters by constructing a context over multiple stimuli and responses. Second, we construct a context consisting of encoder parameters and response values over multiple voxels to perform aggregated functional inversion. We demonstrate strong cross-subject and cross-scanner generalization across diverse visual backbones without retraining or fine-tuning. Moreover, our approach requires neither anatomical alignment nor stimulus overlap. This work is a critical step towards a generalizable foundation model for non-invasive brain decoding.

Recent advances in single-cell technologies have advanced our understanding of gene regulation and cellular heterogeneity at single-cell resolution. Single-cell data contain both gene expression levels and the proportion of expressing cells, which makes them structurally different from bulk data. Currently, methodological work on causal mediation analysis for single-cell data remains limited and often requires specific distributional assumptions. To address this challenge, we present QuasiMed, a mediation framework specialized for single-cell data. Our proposed method comprises three steps, including (i) screening mediator candidates through penalized regression and marginal models (similar to sure independence screening), (ii) estimation of indirect effects through the average expression and the proportion of expressing cells, (iii) and hypothesis testing with multiplicity control. The key benefit of QuasiMed is that it specifies only the mean functions of the mediation models through a quasi-regression framework, thereby relaxing strict distributional assumptions. The method performance was evaluated through the real-data-inspired simulations, and demonstrated high power, false discovery rate control, and computational efficiency. Lastly, we applied QuasiMed to ROSMAP single-cell data to illustrate its potential to identify mediating causal pathways. R package is freely available on GitHub repository at https://github.com/sjahnn/QuasiMed.

This study investigates the influence of different types of non-pharmaceutical interventions (NPIs) on epidemic progression using SIR compartmental models. We analyze the optimization of two distinct targets: the final epidemic size and the infection peak, particularly how they respond to variations in the initiation time of the NPIs. We derive analytical approximations for the critical points of the infection curve of the standard mean-field SIR model with NPIs, and for the epidemic size, enabling a systematic comparison. The analytical results reveal the existence of six different allowed scenarios for the evolution of the epidemic with a single NPI. Furthermore, by employing degree-based mean-field network models, we distinguish between NPIs that decrease the transmission rate (individual and environmental measures) and those that reduce social contacts (lock down measures). We find that, when assuming equal effects on the reproductive number, the former are more efficient in reducing the final epidemic size. Meanwhile, the effectivities of both types of NPIs differ in reducing primary and secondary peaks. The results for all models consistently confirm that minimizing the infection peak requires earlier implementation of the NPI than minimizing the epidemic size, offering new insights for strategic public health timing.

Many essential biological functions, such as breathing and locomotion, rely on the coordination of robust and adaptable rhythmic patterns, governed by specific network architectures known as connectomes. Rhythmic adaptation is often linked to slow structural modifications of the connectome through synaptic plasticity, but such mechanisms are too slow to support rapid, localized rhythmic transitions. Here, we propose a neuromodulation-based control architecture for dynamically reconfiguring rhythmic activity in networks with fixed connectivity. The key control challenge is to achieve reliable rhythm switching despite neuronal degeneracy, a form of structured variability where widely different parameter combinations produce similar functional output. Using equivariant bifurcation theory, we derive necessary symmetry conditions on the neuromodulatory projection topology for the existence of target gaits. We then show that an adaptive neuromodulation controller, operating in a low-dimensional feedback gain space, robustly enforces gait transitions in conductance-based neuron models despite large parametric variability. The framework is validated in simulation on a quadrupedal gait control problem, demonstrating reliable gallop-to-trot transitions across 200 degenerate networks with up to fivefold conductance variability.

Immunohistochemistry (IHC) is essential for assessing specific immune biomarkers like Human Epidermal growth-factor Receptor 2 (HER2) in breast cancer. However, the traditional protocols of obtaining IHC stains are resource-intensive, time-consuming, and prone to structural damages. Virtual staining has emerged as a scalable alternative, but it faces significant challenges in preserving fine-grained cellular structures while accurately translating biochemical expressions. Current state-of-the-art methods still rely on Generative Adversarial Networks (GANs) or standard convolutional U-Net diffusion models that often struggle with "structure and staining trade-offs". The generated samples are either structurally relevant but blurry, or texturally realistic but have artifacts that compromise their diagnostic use. In this paper, we introduce HistDiT, a novel latent conditional Diffusion Transformer (DiT) architecture that establishes a new benchmark for visual fidelity in virtual histological staining. The novelty introduced in this work is, a) the Dual-Stream Conditioning strategy that explicitly maintains a balance between spatial constraints via VAE-encoded latents and semantic phenotype guidance via UNI embeddings; b) the multi-objective loss function that contributes to sharper images with clear morphological structure; and c) the use of the Structural Correlation Metric (SCM) to focus on the core morphological structure for precise assessment of sample quality. Consequently, our model outperforms existing baselines, as demonstrated through rigorous quantitative and qualitative evaluations.

Scalable assessments of mental illness remain a critical roadblock toward accessible and equitable care. Here, we show that everyday human-computer interactions encode mental health with biomarker accuracy. We introduce MAILA, a MAchine-learning framework for Inferring Latent mental states from digital Activity. We trained MAILA on 18,200 cursor and touchscreen recordings labelled with 1.3 million mental-health self-reports collected from 9,500 participants. MAILA tracks dynamic mental states along 13 clinically relevant dimensions, resolves circadian fluctuations and experimental manipulations of arousal and valence, achieves near-ceiling accuracy at the group level, and captures information about mental health that is only partially reflected in verbal self-report. By extracting signatures of psychological function that have so far remained untapped, MAILA establishes human-computer interactions as a new modality for scalable digital phenotyping of mental health.

Plant-soil feedback is recognized as a causal mechanism for the emergence of vegetation patterns of the same species especially when water is not a limiting resource (e.g. humid environments). Nevertheless, in the field, plants rarely grow in monoculture but compete with other plant species. In these cases, plant-soil feedback was shown to play a key role in plant-species coexistence. Using a mathematical model consisting of four PDEs, we investigate mechanisms of inter- and intra-specific plant-soil feedback on the coexistence of two competing plant species. In particular, the model takes into account both negative and positive feedback influencing the growth of the same and the other plant species. Both the coexistence of the plant species and the dominance of a particular plant species is examined with respect to all model parameters together with the emergence of spatial vegetation patterns.

During the formation of a thrombus, the architecture of the growing platelet aggregate is heterogeneous, with areas of dense and loosely packed platelets. The surface of activated platelets facilitate biochemical coagulation reactions that ultimately result in the formation of a fibrin network which stabilizes the thrombus. How platelet-plug microstructure and flow jointly govern the onset and development of fibrin is incompletely understood. We developed a novel 2D computational framework that integrates (1) a pre-adhered, discrete platelet aggregate, (2) a reduced coagulation model that generates thrombin, and (3) a fibrin polymerization model. Three platelet-plug configurations were constructed with prescribed interplatelet gaps and simulations were performed with various wall shear rates. We quantified spatiotemporal clotting metrics, including coagulation factor concentrations, fibrin evolution, and gelation onset. Across geometries, gelation initiation accelerated with increasing plug density. For more dense geometries, gelation emerged first near the plug periphery. As the platelet density increased, intraplug transport was increasingly restricted and the thrombin concentrations in between platelets increased. In contrast, the loose plug supported fibrinogen replenishment deeper into the plug core. Despite slower coagulation initiation due to reduced platelet surface area, monomer generation persisted in the interior, causing gelation to begin at the vessel wall. These results suggest a mechanistic tradeoff: rapid sealing of the injured vessel wall by early platelet contraction, i.e. plug densification, may impede the intraplug fibrin formation needed for durable stabilization. The proposed model provides a basis for studies of platelet-coagulation interactions under flow, including therapeutic developments relevant to prevention of cardiovascular disease.

Quantitative systems pharmacology (QSP) models are increasingly applied to inform decision making across drug development and to support regulatory interactions within model informed drug development (MIDD). QSP supports a broad range of applications across drug development and can be tailored to specific therapeutic areas, mechanisms of action, and contexts of use (CoU). While this diversity is a core strength of QSP, it also presents challenges for reporting for regulatory use. Despite the growing impact of QSP models, there is currently no established guidance on how QSP analyses should be documented and reported for regulatory purposes. This white paper, developed by the International Society of Pharmacometrics (ISoP) QSP Special Interest Group Working Group on Credibility Assessment of QSP for Regulatory Use, seeks to address this gap by proposing best practices for QSP model reporting in regulatory settings. The recommendations are grounded in collective real world experience from regulatory interactions and are aligned with reporting guidance established for physiologically based pharmacokinetic (PBPK) modeling and reporting principles outlined in ICH M15. Rather than prescribing a rigid, one size fits all template, this work proposes a flexible, tiered reporting framework that accounts for development phase and model impact. The proposed framework is intended to facilitate regulatory review and enhance transparency while accommodating the inherent diversity of QSP modeling.

LLMs gain competence by predicting words in human text, which often reflects how people perform tasks. Consequently, coupling an LLM to an engineered runtime turns prediction into control: outputs trigger interventions that enact goal-oriented behavior. We argue that a central design lever is where control resides in these systems. Brains embed prediction within layered feedback controllers calibrated by the consequences of action. By contrast, LLM agents implement Cartesian agency: a learned core coupled to an engineered runtime via a symbolic interface that externalizes control state and policies. The split enables bootstrapping, modularity, and governance, but can induce sensitivity and bottlenecks. We outline bounded services, Cartesian agents, and integrated agents as contrasting approaches to control that trade off autonomy, robustness, and oversight.

The world is full of systems of distributed agents, collaborating and competing in complex ways: firms and workers specialise within economies, neurons adapt their tuning across brain circuits, and species compete and coexist within ecosystems. In that context, individual research fields built theories explaining how comparative advantage drives trade specialisation, how balanced neural representations emerge from sensory coding, and how biodiversity sustains ecological productivity. Here we propose that many of these well-understood findings across fields can be captured in one simple joint cross-disciplinary model, which we call the Distributed Production System. It captures how agent heterogeneity, resource constraints, communication topology, and task structure jointly determine the productivity, efficiency, and robustness of distributed systems across biology, economics, neuroscience, and computing. This model reveals that a small set of underlying laws generates the complex dynamics observed across fields. These can be summarised in our Principle of Maximum Heterogeneity: any distributed production system optimising for performance will converge on an increasingly heterogeneous configuration; environmental demands place an upper bound on the degree of heterogeneity required; and the communication topology determines the spatial scale over which heterogeneity spreads, with this principle applying recursively across all layers of nested production systems. Beyond explaining existing systems, these principles act as a blueprint for constructing ideal ones. We demonstrate this by suggesting specific redesigns for compute systems executing large-scale AI. In total, The Principle of Maximum Heterogeneity reveals a unique convergence of complex phenomena across fields onto simple underlying design principles with important predictive value for future distributed production systems.

Brightfield time-lapse imaging is widely used in cardiac tissue engineering, yet the absence of standardized, interpretable analytical frameworks limits reproducibility and cross-platform comparison. We present an open, scalable computational pipeline for quantifying spatiotemporal contractile dynamics in microscopy videos of human induced pluripotent stem cell-derived cardiac microbundles. Building on our open-source tools "MicroBundleCompute" and "MicroBundlePillarTrack," we define a suite of 16 interpretable structural, functional, and spatiotemporal metrics that capture tissue deformation, synchrony, and heterogeneity. The framework integrates full-field displacement tracking, strain reconstruction, spatial registration, dimensionality reduction, and topology-based vector-field analysis within a unified workflow. Applied to a dataset of 670 cardiac microbundles spanning 20 experimental conditions, the pipeline reveals continuous variation in contractile phenotypes rather than discrete condition-specific clustering, with intra-condition variability often exceeding inter-condition differences. Redundancy analysis identifies a reduced core set of 10 metrics that retain most informational content while minimizing multicollinearity. Analysis of denoised displacement fields shows that contraction is dominated by a global isotropic mode, with localized saddle-type deformation patterns present in approximately half of the samples. All software and workflows are released openly to enable reproducible, scalable analysis of dynamic tissue mechanics.

Activity cliff prediction - identifying positions where small structural changes cause large potency shifts - has been a persistent challenge in computational medicinal chemistry. This work focuses on a parsimonious definition: which small modifications, at which positions, confer the highest probability of an outcome change. Position-level sensitivity is calculated using 25 million matched molecular pairs from 50 ChEMBL targets across six protein families, revealing that two questions have fundamentally different answers. "Which positions vary most?" is answered by scaffold size alone (NDCG@3 = 0.966), requiring no machine learning. "Which are true activity cliffs?" - where small modifications cause disproportionately large effects, as captured by SALI normalization - requires an 11-feature model with 3D pharmacophore context (NDCG@3 = 0.910 vs. 0.839 random), generalizing across all six protein families, novel scaffolds (0.913), and temporal splits (0.878). The model identifies the cliff-prone position first 53% of the time (vs. 27% random - 2x lift), reducing positions a chemist must explore from 3.1 to 2.1 - a 31% reduction in first-round experiments. Predicting which modification to make is not tractable from structure alone (Spearman 0.268, collapsing to -0.31 on novel scaffolds). The system is released as open-source code and an interactive webapp.

Small longitudinal clinical cohorts, common in maternal health, rare diseases, and early-phase trials, limit computational modeling: too few patients to train reliable models, yet too costly and slow to expand through additional enrollment. We present multiplicity-weighted Stochastic Attention (SA), a generative framework based on modern Hopfield network theory that addresses this gap. SA embeds real patient profiles as memory patterns in a continuous energy landscape and generates novel synthetic patients via Langevin dynamics that interpolate between stored patterns while preserving the geometry of the original cohort. Per-pattern multiplicity weights enable targeted amplification of rare clinical subgroups at inference time without retraining. We applied SA to a longitudinal coagulation dataset from 23 pregnant patients spanning 72 biochemical features across 3 visits (pre-pregnancy baseline, first trimester, and third trimester), including rare subgroups such as polycystic ovary syndrome and preeclampsia. Synthetic patients generated by SA were statistically, structurally, and mechanistically indistinguishable from their real counterparts across multiple independent validation tests, including an ordinary differential equation model of the coagulation cascade. A downstream utility test further showed that a mechanistic model calibrated entirely on synthetic patients predicted held-out real patient outcomes as well as one calibrated on real data. These results demonstrate that SA can produce clinically useful synthetic cohorts from very small longitudinal datasets, enabling data-augmented modeling in small-cohort settings.

Early initiation of alcohol, nicotine, cannabis, and other substances predicts later substance use disorders and related psychopathology. We integrate time-varying environmental factors with polygenic risk scores (PRS) in a longitudinal framework to identify determinants of substance initiation in adolescence. Using data from the Adolescent Brain Cognitive Development (ABCD) Study with repeated assessments over approximately four years, we defined time-to-event outcomes for first use of alcohol, nicotine, cannabis, and any substance. We constructed high-dimensional panels of time-varying environmental covariates across family, school, neighborhood, behavioral, and health domains, alongside time-invariant covariates and PRS for alcohol, cannabis, nicotine, and general substance use disorders. Time-varying Cox models with clustered standard errors were applied. Univariate analyses showed broad associations between earlier initiation and multiple environmental domains, including impulsivity, sleep disturbance, parental monitoring, caffeine use, and school functioning. In multivariable models, a smaller set of predictors remained robust, particularly impulsivity traits, parental monitoring, and selected health and lifestyle factors. PRS were positively associated with earlier initiation, with the strongest and most consistent effects for nicotine-related genetic risk. Secondary analyses using marginal structural models suggested that higher parental monitoring is protective, whereas higher impulsivity and caffeine exposure are associated with increased risk. These results demonstrate that integrating dynamic environmental exposures with genetic liability can identify key risk factors for adolescent substance initiation and highlight actionable targets for prevention.

Recent advances at the intersection of control theory, neuroscience, and machine learning have revealed novel mechanisms by which dynamical systems perform computation. These advances encompass a wide range of conceptual, mathematical, and computational ideas, with applications for model learning and training, memory retrieval, data-driven control, and optimization. This tutorial focuses on neuro-inspired approaches to computation that aim to improve scalability, robustness, and energy efficiency across such tasks, bridging the gap between artificial and biological systems. Particular emphasis is placed on energy-based dynamical models that encode information through gradient flows and energy landscapes. We begin by reviewing classical formulations, such as continuous-time Hopfield networks and Boltzmann machines, and then extend the framework to modern developments. These include dense associative memory models for high-capacity storage, oscillator-based networks for large-scale optimization, and proximal-descent dynamics for composite and constrained reconstruction. The tutorial demonstrates how control-theoretic principles can guide the design of next-generation neurocomputing systems, steering the discussion beyond conventional feedforward and backpropagation-based approaches to artificial intelligence.

Inferring cell-cell communication (CCC) from single-cell transcriptomics remains fundamentally limited by reliance on curated ligand-receptor databases, which primarily capture co-expression rather than the system-level effects of signaling on cellular states. Here, we introduce QuantumXCT, a hybrid quantum-classical generative framework that reframes CCC as a problem of learning interaction-induced state transformations between cellular state distributions. By encoding transcriptomic profiles into a high-dimensional Hilbert space, QuantumXCT trains parameterized quantum circuits to learn a unitary transformation that maps a baseline non-interacting cellular state to an interacting state. This approach enables the discovery of communication-driven changes in cellular state distributions without requiring prior biological assumptions. We validate QuantumXCT using both synthetic data with known ground-truth interactions and single-cell RNA-seq data from ovarian cancer-fibroblast co-culture model. The QuantumXCT model accurately recovered complex regulatory dependencies, including feedback structures, and identified dominant communication hubs such as the PDGFB-PDGFRB-STAT3 axis. Importantly, the learned quantum circuit is interpretable: its entangling topology was translated into biologically meaningful interaction networks, while post hoc contribution analysis quantified the relative influence of individual interactions on the observed state transitions. Notably, by shifting CCC inference from static interaction lookup to learning data-driven state transformations, QuantumXCT provides a generative framework for modeling intercellular communication. This work establishes a new paradigm for de novo discovery of communication programs in complex biological systems and highlights the potential of quantum machine learning in the context of single-cell biology.

Proteins are essential biological macromolecules that execute life functions. Local structural motifs, such as active sites, are the most critical components for linking structure to function and are key to understanding protein evolution and enabling protein engineering. Existing computational methods struggle to identify and compare these local structures, which leaves a significant gap in understanding protein structures and harnessing their functions. This study presents PLASMA, a deep-learning-based framework for efficient and interpretable residue-level local structural alignment. We reformulate the problem as a regularized optimal transport task and leverage differentiable Sinkhorn iterations. For a pair of input protein structures, PLASMA outputs a clear alignment matrix with an interpretable overall similarity score. Through extensive quantitative evaluations and three biological case studies, we demonstrate that PLASMA achieves accurate, lightweight, and interpretable residue-level alignment. Additionally, we introduce PLASMA-PF, a training-free variant that provides a practical alternative when training data are unavailable. Our method addresses a critical gap in protein structure analysis tools and offers new opportunities for functional annotation, evolutionary studies, and structure-based drug design. Reproducibility is ensured via our official implementation at https://github.com/ZW471/PLASMA-Protein-Local-Alignment.git.

The simulation of whole-brain dynamics should reproduce realistic spontaneous and evoked neural activity across different scales, including emergent rhythms, spatio-temporal activation patterns, and macroscale complexity. Once a mathematical model is selected, its configuration must be determined by properly setting its parameters. A critical preliminary step in this process is defining an appropriate set of observables to guide the selection of model configurations (parameter tuning), laying the groundwork for quantitative calibration of accurate whole-brain models. Here, we address this challenge by presenting a framework that integrates two complementary tools: The Virtual Brain (TVB) platform for simulating whole-brain dynamics, and the Collaborative Brain Wave Analysis Pipeline (Cobrawap) for analyzing simulation outputs using a set of standardized metrics. We apply this framework to a 998-node human connectome, using two configurations of the Larter-Breakspear neural mass model: one with the TVB default parameters, the other tuned using Cobrawap. The results reveal that the tuned configuration exhibits several biologically relevant features, absent in the default model for both spontaneous and evoked dynamics. In response to external perturbations, the tuned model generates non-stereotyped, complex spatio-temporal activity, as measured by the perturbational complexity index. In spontaneous activity, it exhibits robust alpha-band oscillations, infra-slow rhythms, scale-free characteristics, greater spatio-temporal heterogeneity, and asymmetric functional connectivity. This work demonstrates how combining TVB and Cobrawap can guide parameter tuning and lays the groundwork for data-driven calibration and validation of accurate whole-brain models.

Kinetic modeling of microbial growth is essential for the design, optimization, and scale-up of industrial bioprocesses. Classical empirical models often lack biologically interpretable parameters or fail to capture complex multiphasic (polyauxic) behaviors, while fully mechanistic models are impractical for systems involving complex substrates and mixed cultures. This study proposes a unified mathematical framework that reformulates the canonical Boltzmann and Gompertz equations into semi-mechanistic forms, explicitly defining the maximum specific reaction rate and lag phase duration. Polyauxic growth is represented as a weighted sum of sigmoidal phases, subject to stringent constraints that ensure parameter identifiability, temporal consistency, and biological plausibility. The methodology integrates a workflow to address nonlinear regression in high-dimensional parameter spaces. A two-stage optimization strategy using Differential Evolution for global search followed by L-BFGS-B for local refinement avoid bias and heuristic parameter initialization. A Charbonnier loss function and the Robust Regression and Outlier Removal procedure are employed to identify and mitigate experimental outliers. Model parsimony is enforced using Akaike (AIC, AICc) and Bayesian (BIC) information criteria to select the optimal number of growth phases and avoid overparameterization. The framework was evaluated using experimental anaerobic digestion datasets, demonstrating that conventional single-phase models can obscure relevant metabolic transitions in co-digestion systems.

Inferring protein production kinetics for dividing cells is complicated due to protein inheritance from the mother cell. For instance, fluorescence measurements -- commonly used to assess gene activation -- may reflect not only newly produced proteins but also those inherited through successive cell divisions. In such cases, observed protein levels in any given cell are shaped by its division history. As a case study, we examine activation of the glc3 gene in yeast involved in glycogen synthesis and expressed under nutrient-limiting conditions. We monitor this activity using snapshot fluorescence measurements via flow cytometry, where GFP expression reflects glc3 promoter activity. A naïve analysis of flow cytometry data ignoring cell division suggests many cells are active with low expression. Explicitly accounting for the (non-Markovian) effects of cell division and protein inheritance makes it impossible to write down a tractable likelihood -- a key ingredient in physics-inspired inference, defining the probability of observing data given a model. The dependence on a cell's division history breaks the assumptions of standard (Markovian) master equations, rendering traditional likelihood-based approaches inapplicable. Instead, we adapt conditional normalizing flows (a class of neural network models designed to learn probability distributions) to approximate otherwise intractable likelihoods from simulated data. In doing so, we find that glc3 is mostly inactive under stress, showing that while cells occasionally activate the gene, expression is brief and transient.

We constructed a computational model of the driver's brain for steering tasks using the active inference framework, grounded in the free energy principle - a theory from computational neuroscience. This model enables quantitative estimation of how accurately the brain learns vehicle dynamics and performs appropriate steering, using a measure called variational free energy. Through driving simulator experiments, we observed strong correlations between variational free energy and both expert drivers' subjective "as-intended" scores and general participants' objective control performance. These results suggest that variational free energy provides a promising quantitative metric for evaluating whether a vehicle behaves "as-intended."

In genome-scale constraint-based metabolic models, gene deletion strategies are essential for achieving growth-coupled production, where cell growth and target metabolite synthesis occur simultaneously. Despite the inherently networked nature of genome-scale metabolic models, existing computational approaches rely primarily on sequential data and lack graph representations that capture their complex relationships, as both well-defined graph constructions and learning frameworks capable of exploiting them remain largely unexplored. To address this gap, we present a twofold solution. First, we introduce a systematic pipeline for constructing graph representations from constraint-based metabolic models. Second, we develop a deep learning framework that integrates these graph representations with gene and metabolite sequence data to predict growth-coupled gene deletion strategies. Across three metabolic models, our approach consistently outperforms established baselines, with improvements in overall accuracy of 14.04%, 16.26%, and 13.18% over a deep feedforward neural network baseline, 6.17%, 4.96%, and 5.31% over a sequence-learning baseline, and 5.10%, 4.36%, and 4.70% over a topology-aware graph aggregation baseline on the same metabolite graph, respectively. The source code and example datasets are available at: https://github.com/MetNetComp/GraphGDel.

Data fusion modeling can identify common features across diverse data sources while accounting for source-specific variability. Here we introduce the concept of a \textit{coupled generator decomposition} and demonstrate how it generalizes sparse principal component analysis (SPCA) for data fusion. Leveraging data from a multisubject, multimodal (electro- and magnetoencephalography (EEG and MEG)) neuroimaging experiment, we demonstrate the efficacy of the framework in identifying common features in response to face perception stimuli, while accommodating modality- and subject-specific variability. Through split-half cross-validation of EEG/MEG trials, we investigate the optimal model order and regularization strengths for models of varying complexity, comparing these to a group-level model assuming shared brain responses to stimuli. Our findings reveal altered $\sim170ms$ fusiform face area activation for scrambled faces, as opposed to real faces, particularly evident in the multimodal, multisubject model. Model parameters were inferred using stochastic optimization in PyTorch, demonstrating comparable performance to conventional quadratic programming inference for SPCA but with considerably faster execution. We provide an easily accessible toolbox for coupled generator decomposition that includes data fusion for SPCA, archetypal analysis and directional archetypal analysis. Overall, our approach offers a promising new avenue for data fusion.