Nature Neuroscience

Many neural recordings have revealed low-dimensional sets of behaviorally relevant variables encoded within large-scale neural activity patterns. However, dimensionality reduction analyses alone cannot yield causal explanations for how networks stably implement computations that are resilient to the substantial variability of single neuron dynamics. Further, existing methods for dimensionality reduction often rely on simplifying assumptions about network structure that limit their applicability and explanatory power. To provide a theoretical framework describing the dynamics of low-dimensional computation in high-dimensional neural networks, here we introduce the concept of latent processing units (LPUs), which are architecture-agnostic computational elements operating within biological neural circuitry. Six theorems governing coding and computation by LPUs collectively provide explanations for a range of common biological findings: low-dimensional sets of coding variables can generate high-dimensional neural dynamics; many neurons have activity patterns that represent behaviorally relevant variables but exert little influence on downstream circuits; linear readouts of neural population activity commonly permit near-optimal decoding; the drift of neural representations is often substantial even while network computations remain intact. Overall, our treatment of LPUs, as enacted in network dynamics, unifies the geometric and dynamical views of neural computation under a joint framework and provides systems neuroscience with a causal account of how the brain executes reliable computations.

Boredom - a pervasive mental state - promotes the pursuit of novel information by assigning negative value to monotonous conditions. Yet, how the brain extracts and represents the information content of ongoing sensory experience remains poorly understood. Here, we combine behavioral assays, neurophysiological recordings and computational modeling across humans and mice to investigate how sensory information shapes boredom-related behavior. In a cross-species choice task, both humans and mice robustly avoid monotonous sources of sensory stimulation. We formalize perceived monotony using empirical entropy as a measure of information content and show that monotony avoidance scales directly with low entropy and in humans correlates with boredom experience. Human electroencephalography and mesoscopic calcium imaging in mice reveal that the recruitment of neocortical activity tracks stimulus entropy. Two-photon calcium imaging in the auditory cortex of mice further uncovers a stimulus-invariant population code for entropy, supported by neurons tuned to information content. A recurrent network model reproduced this code through an interplay of afferent depression and recurrent facilitation. Together, we demonstrate how the information content of sensory experience is represented in cortical population activity, providing a basis for boredom-related avoidance behavior. Thus, our findings link synaptic and neuronal dynamics to boredom, acting as a safeguard mechanism to ensure high information input to the brain.

With recent advances in high-throughput proteomic technologies, population-scale plasma proteomics datasets, often linked to extensive genetic and phenotypic information, have become increasingly accessible. Yet the relationships between circulating protein levels, brain imaging phenotypes, and risk for neurological and psychiatric disorders remain largely unexplored. Proteome-wide association studies offer a promising approach for elucidating biological mechanisms that connect genetic variation to complex brain-related traits and diseases. In this study, we integrated protein quantitative trait loci (pQTLs) from the two largest plasma proteomic resources (the UK Biobank Pharma Proteomics Project [UKB-PPP] and Ferkingstad et al. [deCODE]) with genome-wide association studies of brain imaging-derived phenotypes in UK Biobank using Mendelian randomization and colocalization analyses. We identified 120 cis and 20 trans associations between plasma proteins and imaging phenotypes and validated these findings using brain tissue-derived proteomic and transcriptomic datasets. Multivariable Mendelian randomization revealed eleven plasma proteins (coding genes APOE, ARL3, MICB, NSF, RHOC, RSPO3, ENPP2, BTN2A1, EIF2AK3, MRVI1, and OPLAH) with significant direct effects on the risk of Alzheimer's disease, Parkinson's disease, multiple sclerosis, bipolar disorder, and schizophrenia. Single-cell expression and pathway enrichment analyses further revealed cell-type-specific effects and distinct biological processes underlying these protein-disease associations. Together, these findings demonstrate robust links between plasma protein variation and brain structure, delineate protein-disease pathways, and highlight the cellular and molecular mechanisms that contribute to neurobiological diversity and pathology.

AO_SCPLOWBSTRACTC_SCPLOWThere is a gap between neuroscientific theories of learning and the speed of learning observed in many experiments. Since the Cognitive Revolution of the 1950s, compositionality has played a central role in efforts to bridge this gap. Roughly, a compositional system is one where distinct modules are combined according to a set of rules in order to accomplish complex tasks. Recently, significant progress has been made in understanding the emergence of modules in both biological and artificial neural systems. How, and under what conditions, the rules of module recombination are represented in these systems remains an open question. Here we present a neural model that can leverage these rules to dramatically speed up learning. We first show that when faced with multiple tasks which share subcomponents, models learn a low-dimensional representation that captures how subcomponents are reused across the task set. These low-dimensional spaces encode the structure that governs how modules should be recombined. Restricting learning to these subspaces greatly reduces the amount of experience needed to acquire a novel task, even when learning from reinforcement on single trials. In some cases, we can leverage the geometric regularities of these representations to reduce learning to a form of hypothesis testing over a small set of discrete points. Finally, we use this theory to model both behavioral and neural data from non-human primates performing a compositional task, and show that key features in this data are consistent with a model in which exploration during learning is restricted to these low-dimensional spaces. Overall, this work shows that the advantages of modularity in neural systems can be greatly improved upon when models represent the structure of module reuse. Both these features working in tandem lead to learning on timescales similar to biological intelligences, and hence provide a model for how such fast, adaptable behavior can emerge from systems of neurons.

Speech perception requires transforming acoustic input into neural representations that support linguistic understanding, yet its underlying computational principles remain unclear. Classical efficient coding theories posit optimal compression of sensory input, whereas alternative accounts propose that neural systems preferentially encode information that supports prediction. A key open question is whether such predictive encoding operates on fixed inputs or on flexible internal representations. We instantiated three hypothesis models of speech processing: (i) optimal compression with deep autoencoders, (ii) predictive reconstruction with predictive autoencoders, and (iii) predictive information representation via latent-space prediction using contrastive learning. We compared resulting speech latent representations to electroencephalographic (EEG) activity during speech listening. Representations learned under the predictive information objective best explained neural latents. Crucially, only representations that selectively compressed predictive information predicted behavioral performance, suggesting that neural speech representations are structured to encode predictive information in latent space rather than to maximize compression or input prediction.

The increased vulnerability to mental health problems in adolescence is frequently reported but poorly understood, hampered by a rigid diagnostic system which fails to capture intertwining symptoms and only loosely aligns with biological axes of variability. Here, we reconceptualised the mental health symptoms of young adolescents in the ABCD cohort (N=11862) as a latent topology of overlapping symptom dimensions, using an unsupervised machine learning algorithm to establish how transdiagnostic dimensions co-occur and overlap within individuals. Combining this with a novel classification approach, we delineated zones within this landscape, within which specific profiles of symptoms were robustly represented. These data-driven profiles were leveraged to establish associated resting-state functional connectivity and genetic characteristics. In doing so we recaptured the commonly reported p-factor axis as well as further symptom-subtype dimensions. Gene ontology analysis revealed that shared neurobiological and cellular mechanisms embedded in both the genome and transcriptome may confer risk for psychopathology.

Schizophrenia and bipolar disorder are diagnostically distinct categories that overlap substantially in clinical features and genetic aetiology. Understanding genetic variants that contribute liability specifically to each disorder can offer insights into biological processes that differentiate them. Here we used Case-Case GWAS (CC-GWAS) to identify common genetic variants differentially associated with schizophrenia and bipolar disorder, analysing 67,390 schizophrenia cases and 41,917 bipolar disorder cases. We identified 19 genome-wide significant loci, of which 16 (84%) demonstrated divergent genetic effects with risk alleles showing opposite directions of association between disorders. The CC-GWAS summary statistics had detectable disorder-differentiating heritability (10.27%, SE=0.01) and showed genetic correlations indicating that SCZ-differentiating alleles were associated with lower educational attainment, lower cognitive performance, and increased risk of ADHD, anorexia, autism, BD1 (though not BD2), cannabis use disorder, and OCD. Four loci showed divergent effects despite not reaching genome-wide significance in either individual disorder GWAS, demonstrating enhanced power to detect opposite-direction effects. Functional annotation identified 102 mapped genes significantly enriched for expression across all 13 tested brain regions, with no significant enrichment in peripheral tissues, and gene set enrichment analysis implicated neuronal projection and synaptic compartments as the strongest biological themes differentiating the two disorders. Polygenic risk scores derived from these disorder-differentiating variants were associated with earlier age at onset and more severe negative symptoms in schizophrenia, consistent with these variants marking neurodevelopmental dimensions of illness. Our findings provide targets for understanding pathogenic differences between schizophrenia and bipolar disorder and demonstrate that genuine divergent genetic effects exist beyond the substantial shared liability.

Objective: To compare first-line emergency department (ED) treatment classes for acute headache on short-term all-cause ED return and index admission across two independent health systems. Background: ED trials of acute headache treatment are judged on in-ED pain relief, a documented endpoint that is recorded incompletely and shifts with the scoring rule, and is a weak surrogate for what happens after discharge. All-cause ED return after an index headache visit (any subsequent ED encounter within the window) has not been used to compare first-line treatments at scale, and society guidance favors dopamine-receptor antagonists while recommending against routine opioids. Methods: Retrospective two-center cohort of adults treated for headache in the ED, using MIMIC-IV-ED (Beth Israel Deaconess Medical Center, 2011-2019) and MC-MED (Stanford, 2020-2022). The first-line class was the earliest qualifying acute agent. The primary contrast was opioids versus dopamine-receptor antagonists (the guideline-preferred class). Outcomes were 72-hour and 7-day all-cause ED return (among discharged patients; any subsequent ED encounter within the window) and index hospital admission. Confounding by indication was addressed with propensity overlap weighting; associations are reported as adjusted risk ratios (RRs) with bootstrap 95% CIs and E-values. Estimates were pooled with a site term and examined per site. Results: Among 13,285 treated adults (10,799 MIMIC-IV-ED; 2,486 MC-MED), opioid recipients were older and higher-acuity than dopamine-antagonist recipients (index admission 38.1% vs 16.4%). In the MIMIC-IV-ED discharged primary-contrast population, overlap weighting reduced the maximum standardized mean difference from 0.35 to 0.002; pooled and site-specific balance diagnostics are provided in the Supplement. First-line opioids remained associated with a higher 72-hour all-cause ED return (6.8% vs 3.8%; adjusted RR 1.79; 95% CI 1.31 to 2.33), 7-day return (10.7% vs 6.6%; RR 1.62; 95% CI 1.28 to 1.98), and index admission (RR 2.32; 95% CI 2.11 to 2.58, consistent with strong residual severity differences in patients selected for opioids). The direction of association was concordant across both health systems, although MC-MED return estimates were imprecise given the smaller opioid-treated discharged sample. In MIMIC-IV-ED, the cumulative all-cause return incidence by treatment class separated by day 3 and persisted through 30 days. The direction was consistent, though attenuated and no longer statistically significant, when the outcome was restricted to a headache-specific return (72-hour RR 1.31; 95% CI 0.91 to 1.88); the direction persisted for the composite of admission or 72-hour return, which does not condition on discharge but is influenced by the more confounded admission component (RR 2.16; 95% CI 1.98 to 2.39). Conclusion: Across two health systems, first-line opioid treatment for ED headache was associated with higher all-cause short-term ED return among discharged patients and higher index admission than dopamine antagonists. These observational associations reflect downstream all-cause ED utilization after an index headache visit rather than confirmed headache recurrence or treatment failure; they are consistent with guideline-concordant, opioid-sparing first-line treatment and warrant prospective confirmation. Plain Language Summary: Emergency departments treat headaches with several different medicines, but the usual way of judging which works, the pain score recorded during the visit, is often missing or inconsistent. Using two large hospital systems and a clearer outcome, whether patients came back to the emergency department for any reason, we found that patients first treated with opioids returned within 72 hours about 1.8 times as often as those given the guideline-preferred dopamine-blocking medicines and were admitted more than twice as often. These patterns pointed the same direction in both hospital systems after adjustment for the measured differences available in both databases. Because this was an observational comparison and returns were counted for any reason, the findings are consistent with using guideline-preferred non-opioid medicines first, rather than proof that opioids worsen headache.

The emotional content of suicide notes is typically examined using categorical coding, where each labeled passage is treated in isolation from its surrounding language. In contrast, dimensional models of psychopathology propose that affective content varies along continuous gradients. We evaluated this proposition directly. Excerpts from 884 annotated suicide notes were embedded in a semantic space defined solely by their linguistic properties, and we investigated whether human-assigned emotion labels changed smoothly across this space. They did: affective tone showed clear spatial autocorrelation (Moran's $I = 0.18$, $z = 19.68$, $p < 0.001$), an effect that replicated across three different encoders and remained after removing all within-note dependencies. Emotions occupied recognizable yet overlapping regions rather than forming distinct clusters and varied substantially in how tightly they were concentrated: love and hopelessness appeared with similar frequency, but love was far more localized ($z = 15.7$ versus $10.8$). Among all emotions, hopelessness was the most linguistically diffuse, implying that a single categorical label is capturing multiple, qualitatively different manifestations of suicidal distress.

Developmental competence allows tissues to respond to inductive cues before committing to specialized forms, but how this potential is encoded at clustered gene-family loci is poorly understood. We use vertebrate skin to address this problem. Epidermis responds to regional dermal signals before committing to feather, scale, or differentiated programs, and -keratin loci provide a stringent genomic test: separated type-I/type-II clusters show coordinated transcriptional pairing, yet individual keratin genes are selectively deployed across appendage, differentiation, and disease states. Using chicken developmental genomics with comparative mouse and human epidermal datasets, we show that -keratin clusters are organized before commitment as scaffolded chromatin domains. Within these domains, regulatory elements remain broadly accessible but acquire state-specific activity during commitment and differentiation. Inter-cluster contacts and chromatin-factor perturbation link this architecture to keratin output and morphology. These findings reveal a locus-level chromatin basis for developmental competence, enabling domain-level coordination with gene-level selectivity during epidermal diversification.

Hippocampal place cell activity represents an animals location in space; yet, how hippocampal neuronal population dynamics change with spatial learning and the mechanisms underlying these activity changes, which drive allocentric navigation to a learned goal, are poorly understood. To address these questions, we performed calcium imaging with a novel wire-free waterproof miniaturized microscope to image the activity of large populations of hippocampal CA1 neurons during spatial learning of a two-dimensional navigational task, the Morris water maze. We followed the same cells during learning and were able to directly examine how each neuron in the ensemble, and the ensemble as a whole, changes its response properties. We found that neuronal spatial selectivity increased and population decoding of spatial location improved as mice learned to navigate to the goal. Viral CRISPR knock out of Grin1 (encoding the essential GluN1 NMDA receptor subunit) in dorsal hippocampal neurons, dramatically reduced long-term potentiation in CA1. This manipulation also prevented the increase in spatial selectivity and improvement of population decoding with spatial learning and resulted in learning deficits in the Morris water maze. Together, our results show that dorsal hippocampus NMDAR-dependent synaptic plasticity is essential for the learning-dependent refinement of CA1 place selectivity and improvement in population decoding of space.

Comprehensive biomarkers of multiple sclerosis (MS) capable of simultaneously diagnosing the condition, capturing symptom severity and predicting treatment efficacy remain elusive. Although several studies have highlighted the pivotal role played by demyelinating lesions in determining MS structural pathology, their relationship with symptom severity is limited. Here, we combined personalized computational brain modeling with magnetoencephalography (MEG) recordings from 17 MS patients and 20 healthy controls (CTR) to derive personalized brain network excitability parameters, which we tested as MS biomarkers. Personalized parameters discriminated between CTR and MS participants with high accuracy, also classifying between progressing and remitting MS patients. Notably, they also predicted MS clinical scales across multiple domains. In all clinical tasks, personalized parameters consistently outperformed standard clinical measures and total lesion loads. Together, these results highlight the potential of personalized brain modelling in deriving integrative MS biomarkers, capable of simultaneously identifying the condition, classifying MS subtypes and predicting symptom severity. d brain modelling in deriving integrative MS biomarkers, capable of simultaneously identifying the condition, classifying between MS subtypes and predicting the severity of symptomatology.

Insulin signalling contributes to neurodevelopment and brain function, and insulin resistance (IR)-related traits are associated with cognitive performance. However, the genetic architecture shared across specific cognitive domains and IR-related phenotypes remains insufficiently defined. We analysed large-scale genome-wide association study summary statistics for 11 IR-related traits (N=53,334-933,970) and 10 cognitive measures (N=28,156-436,853) to quantify global and local genetic correlations, fine-map shared association signals, and annotate implicated genes and drug-gene interactions. Pairwise global and local genetic correlations were estimated, and shared high-confidence variants were prioritised using the multivariate Sum of Single Effects model. Positional and expression quantitative trait locus mapping was performed, and implicated genes were examined through functional annotation, tissue enrichment, and drug-gene interaction analyses. Low-to-moderate genetic correlations were observed between six IR-related traits and seven cognitive measures (|rg|=0.08-0.34), with predominantly opposite directions, except for correlations involving visual declarative short-term memory. Local genetic correlations showed mixed effect directions across most trait pairs, and multivariate fine-mapping prioritised 696 shared likely causal variants with high posterior support. Gene annotation indicated enrichment in several pathways, including immune-related, signal transduction, neurogenesis, neurotransmitter metabolism, receptor regulation, and lipid and cholesterol metabolism regulation. Implicated genes were expressed across various brain regions and showed prior associations with neuropsychiatric and cardiometabolic conditions. Several drug-gene interactions were identified, involving immunomodulatory and anti-inflammatory compounds. These findings indicate widespread heterogeneous genetic overlap between IR-related traits, particularly body mass index and waist-to-hip ratio, and cognitive measures of general intelligence, processing speed, and short-term visual declarative memory. The findings prioritise apolipoprotein-related lipid transport and inflammatory and oxidative stress pathways as candidate mechanisms linking cognitive, cardiometabolic, and neuropsychiatric phenotypes.

Dementia is a progressive condition that impairs cognitive processes such as memory, decision making, and the ability to manage daily activities. Recent estimates suggest that more than half of all dementia cases could be preventable by addressing their risk factors, including disease comorbidities such as diabetes and vision loss. Yet, we lack a comprehensive molecular map of dementia comorbidities. In this work, we analyzed Austrian nationwide hospital claims data, comprising 13 million hospital stays from 2015 to 2019, to systematically assess dementia-related risk across disease comorbidity patterns, covering both their molecular relationships and their epidemiological overrepresentation. We identified disease trajectories occurring before and at the time of dementia diagnosis, revealing both sex-specific and shared comorbidity patterns. Overall, we identified 51 potential risk factors, with a prominent contribution from endocrine and metabolic disorders. While Parkinson's disease emerged as a strong molecularly related driver of dementia, we also identified emerging and previously under chracterized risk factors, including vitamin D deficiency. This integrative framework provides a comprehensive view of dementia associated disease networks and identifies novel, potentially modifiable risk factors. These results offer new opportunities for targeted prevention strategies and advance our understanding of the complex interplay between comorbidities and dementia development.

Background: Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by death of upper and lower motor neurons, usually presented with clinical heterogeneity. Fluid biomarker development remains dominated by neurofilament light chain (NEFL), a marker of neuroaxonal injury. NEFL is however unspecific to ALS and its phenotypes and there is currently a lack of biomarkers that capture ALS heterogeneity such as onset site and ALS-frontotemporal spectrum disorder (ALS-FTSD). Therefore, we investigated whether plasma proteomics could reveal pathway-level signatures that stratify and explain ALS heterogeneity. Methods: We profiled ~5,400 plasma proteins (Olink Explore HT) in 299 patients with ALS and 50 age- and sex comparable healthy controls. We used two complementary analytic frameworks: (i) differential protein abundance analysis to identify altered proteins in ALS and across clinical subgroups, and (ii) weighted gene correlation network analysis (WGCNA) to identify coordinated protein modules and relate them to ALS diagnosis and to ALS-specific clinical traits (site of onset, ALS-FTSD, ALS functional rating scale-revised (ALSFRS-R) score, and plasma NEFL). Results: Differential abundance analysis identified 56 proteins altered in ALS versus controls, of which 40 were increased. WGCNA identified 11 co-expression modules, with ALS samples having the strongest correlation to a protein module (n=51) highly enriched for muscle-related proteins. Out of the 40 proteins that had increased expression levels, 29 overlapped with the muscle-enriched protein module, indicating that muscle related proteins are the dominant circulating proteomic signature in ALS. This signal extended to clinical stratification: spinal-onset patients showed a strong positive association with the muscle-module. Further, differential abundance analysis of spinal- versus bulbar-onset ALS identified changes that mapped predominantly to the same module, supporting a molecular signature of onset phenotype. In contrast, cognitive status (ALS-FTSD) mapped to distinct modules enriched for extracellular matrix/cell-adhesion pathways, consistent with a separable biological axis of disease heterogeneity. Although multiple modules correlated with NEFL, trait-specific signatures were not fully explained by neuroaxonal injury. Notably, the muscle-enriched module increased with higher NEFL and lower ALSFRS-R, supporting its interpretation as a severity-linked, muscle-involvement proxy. Conclusions: Large-scale plasma proteomics reveals that heterogeneity in ALS reflects underlying biological structures. We identified a dominant muscle-associated protein network that distinguished ALS patients from controls and correlated with disease onset phenotype and severity, alongside distinct protein networks linked to ALS-FTSD. By integrating differential protein abundance with network-based analysis, we defined pathway-level biomarker signatures that extend beyond NEFL, enabling biologically informed patient stratification and improved therapeutic monitoring.

Automated interpretation of neuroimaging studies requires simultaneous assessment of multiple imaging evidence variables, each tied to distinct anatomical structures. Vision-language models (VLMs) offer a unified framework for multi-task analysis, but adapting pre-trained VLMs remains challenging. Full fine-tuning is computationally prohibitive, and joint multi-task training requires simultaneous access to all task data, which is often infeasible in clinical settings. Although model merging enables multi-task composition without joint re-training, existing methods focus on post-hoc algorithms with limited extension to VLMs and minimal application to neuroimaging. Here, we present GRadient-guided Adapter Merging (GRAM), a layer-selective low-rank adaptation (LoRA)-based fine-tuning and merging framework for multi-task neuroimaging visual question-answering (VQA). GRAM uses a gradient ratio that contrasts class-specific gradients to identify task-discriminative layers, and applies subspace-constrained projected gradient descent to restrict LoRA updates to directions consistent with the geometry of the pre-trained model. We leveraged a structured VQA benchmark, developed from the National Alzheimer's Coordinating Center (NACC) dataset, that pairs multi-sequence brain MRI studies with question-answer pairs across clinically relevant imaging evidence variables. Experiments on the VQA benchmark showed that GRAM outperformed or matched all-layer LoRA fine-tuning and a standard merging baseline while reducing inter-task interference during merging, and approached or surpassed the performance of joint multi-task training without joint re-training.

Collective animal behavior occurs in high-stakes contexts where failing to coordinate effectively with group-mates can spell disaster for individuals. Yet, identifying instances of coordination failure is challenging, meaning their evolutionary effects remain mysterious. Synchronous calls in alternating frog choruses (i.e., inadvertent signal collisions) are unambiguous failure events that impose steep attractiveness costs. We modeled tungara frog chorusing dynamics to reveal the sensorimotor and social mechanisms underpinning synchrony. Ultimately, inter-male variation in two key sensorimotor attributes, the periods of male calling rhythms and call latencies, generated divergent synchrony engagement patterns. Modeling female preferences revealed that these varied behavioral outcomes then yielded disparate attractiveness consequences. By mechanistically linking the causes and consequences of coordination failure, we demonstrate that non-random failure patterns in collectives generate selection gradients that refine sensorimotor tuning.

Sensory processing differences are a core feature of autism, affecting 60-95% of individuals, yet the associated neural mechanisms remain unclear. An excitation-inhibition (E/I) imbalance in brain circuits has been proposed, but in vivo evidence linking genetic variation in E/I pathways, regional neurochemistry, neural circuit function, and sensory behaviour has been lacking. Here we performed a multimodal investigation in 206 individuals (130 autistic), integrating gene-set polygenic scores for excitatory glutamatergic and inhibitory gamma-aminobutyric acid (GABA)-ergic pathways, magnetic resonance spectroscopy (MRS) measures of regional GABA and Glx (glutamate + glutamine) levels, vibrotactile psychophysical measures of tactile perception, and questionnaire measures of behavioural sensory reactivity. We found that glutamatergic polygenic scores predicted thalamic glutamate levels in neurotypical but not autistic individuals, suggesting altered genotype-neurochemistry coupling in autism. Thalamic Glx:GABA levels associated with tactile perception in both groups, but with opposing directions of effect, indicating that autistic and neurotypical individuals achieve similar perceptual outcomes with potentially differing thalamocortical circuit mechanisms. Within autistic individuals, tactile perceptual differences further related to behavioural sensory reactivity. Together, these findings suggest that autistic sensory processing potentially relies on distinct circuit mechanisms linking genetic variation, neurochemistry and perception. This work thus has important implications for how sensory differences are conceptualised, studied, and interpreted, and ultimately for how interventions and support are developed.

Unlike most mammals, which lack the capacity to regenerate complex tissues following injury, other vertebrates such as the axolotl rebuild complete limbs throughout life, yet the regulatory mechanisms underlying this striking difference have remained elusive. Here, we define the core cis-regulatory motif grammar driving axolotl limb regeneration and demonstrate that this grammar is conserved within the syntenic neighborhoods of regeneration-gene orthologs in human and mouse genomes, despite being epigenetically sealed in adult mammalian tissues. Integrating this cross-species grammar projection with AlphaGenome, a multimodal genomic AI capable of predicting epigenomic states from long sequence context, we find that the highest-ranking candidate loci are predicted to occupy a state of bivalent dormancy marked by the co-enrichment of poising and repressive histone modifications alongside suppression of transcriptional activity and chromatin accessibility. Systematic in silico motif perturbation further predicts that this dormant state is actively enforced by specific dormancy-stabilizing sequence elements, and that disrupting these elements shifts candidate loci toward a regeneration-competent chromatin configuration. Our findings support a model in which the regenerative blueprint has not been erased from the mammalian genome but locked within it, opening new avenues for understanding the evolution of regenerative competence and for the rational reactivation of latent regenerative programs.

Multimorbidity is highly prevalent in ageing populations, yet its shared molecular basis remains poorly defined, limiting the development of therapies that target multiple conditions. We systematically integrated measurements of 1,954 circulating proteins from 54,219 individuals in discovery and 35,559 in replication, focusing on ten common age-related diseases: coronary artery disease, chronic kidney disease, chronic obstructive pulmonary disease, dementia, heart failure, major depressive disorder, osteoarthritis, Parkinson's disease, stroke, and type 2 diabetes. Coronary artery disease emerged as a central condition in the multimorbidity network, sharing circulating protein signatures with seven other diseases. Through genetic causal-inference analyses, we identified 40 circulating proteins with cross-disease relevance, of which four were further supported by colocalization of genetic variant associations. Among these, complement C1r, encoded by C1R, emerged as a key link between coronary artery disease and dementia, supported by independent colocalization evidence (PP.H4 = 0.86). Phenome-wide association analyses of C1R variants suggested that this signal was not driven by widespread unrelated genetic effects, but instead may reflect a more specific contribution to coronary artery disease-dementia pathogenesis. In vitro experiments further suggested that fibroblast-derived C1R promotes endothelial inflammation and neuronal apoptosis, providing mechanistic plausibility. Together, these findings position C1R as a biologically plausible and therapeutically relevant molecular link between coronary artery disease and dementia.