Biology subjects

Jacobs, J.

Publications and source records attributed to Jacobs, J..

At least 19 recordsLinked to original sources

bioRxiv2026

Global population genomics and pangenome of the clubroot pathogen link clonal expansion and effector diversification to resistance breakdown in canola

Clubroot disease, caused by the obligate biotrophic protist Plasmodiophora brassicae, threatens Brassica crops worldwide. Novel pathogen virulence frequently breaks resistance in agricultural crops, and yet explanations for isolate evolution remain unknown. We combined whole genome sequencing of 192 isolates from 24 countries with a pangenome analysis of 55 high-quality complete or nearly complete genome assemblies and performed functional effector analyses to investigate potential mechanisms of virulence emergence. Three major phylogenetic clusters were identified, including two highly clonal lineages with contrasting virulence profiles, and a genetically diverse population retaining signatures of recombination. Our findings support a model in which genome-wide purifying selection maintained core genome integrity, while effector genes evolved under less constraint. Despite a compact closed pangenome, we identified chromosome 17 as a major hotspot of single nucleotide and structural variations, transposable elements and expansion of ankyrin repeat effectors. Within the variable chromosome 17, PbAnk2 alleles from virulent and avirulent isolates encoded proteins with distinct predicted structures. PbAnk2 interacted with the Arabidopsis co-chaperone DJC31 in yeast two-hybrid and BiFC assays, whereas co-immunoprecipitation did not detect an interaction for the allele found in isolates that break CRa-mediated resistance, suggesting that allelic variation may alter PbAnk2's association with DJC31. These findings reveal the clonal expansion of virulent isolates and connect it to effector diversification that led to resistance breakdown and establish a genomic framework for pathogen surveillance and durable clubroot resistance.

genomics↗

bioRxiv2026

Beyond Rate: Hippocampal Phase Codes for Space and Memory

Place cells and concept neurons are defined by a rise in firing rate at a particular location or for a particular person or object. Neurons without such a rise remain unlabeled, although they keep firing during the task. Such neurons could still carry information in the timing of their spikes. Hippocampal theta provides a reference for this timing. We therefore asked whether the theta phase of a neuron's spikes depends on location or on the viewed image, even when its firing rate does not. We analysed hippocampal neurons in humans navigating a virtual city, rats running on a linear track, and humans viewing images of anima ls, faces, and places. In both navigation datasets, more neurons than expected by chance fired at a consistent theta phase at particular locations. This phase tuning remained after we excluded neurons whose firing rate depended on location. In human navigation, the spike phases of these remaining neurons improved location decoding beyond the firing rates of all neurons. More phase-tuned neurons than expected by chance also fired at progressively earlier theta phases as participants passed each neuron's preferred location. This pattern resembles phase precession, which classically occurs as an animal crosses the firing-rate field of a place cell. The neurons that showed it had no firing-rate field, so their phase shift was anchored to the location defined by phase tuning alone. During image viewing, more neurons than expected by chance changed the theta phase of their spikes after a particular image appeared. This effect remained after we removed, for each neuron, the images that changed its firing rate. Spike phases alone also predicted the image category above chance. Theta-relative spike timing therefore carries in formation about location and images that firing rate alone misses. We propose that weak input changes when a neuron fires with in the theta cycle, whereas strong input also raises its firing rate.

neuroscience↗

bioRxiv2026

Polysaccharide release factors drive lipid-carrier recycling in bacteria

Gram-negative bacteria build their cell envelope by transporting various sugar-based precursors across the inner membrane on the shared lipid carrier undecaprenyl (also known as C55). Because the cellular pool of undecaprenyl is finite and used by multiple competing pathways, it must be efficiently regenerated to sustain envelope assembly. Yet the enzymes that release polysaccharide cargo from the lipid carrier have remained elusive, leaving a major gap in our understanding of bacterial envelope biogenesis. Here we identify three periplasmic enzymes in Escherichia coli - UprA, UprB, and UprC - that hydrolyze the bond linking polysaccharide precursors to undecaprenyl. UprA releases elongating peptidoglycan glycan strands from the inner membrane, catalyzing the long-sought terminal step of peptidoglycan synthesis. UprB cleaves the undecaprenyl-linked initiating unit common to the enterobacterial common antigen and O-antigen pathways, while UprC is the likely release factor for the colanic acid pathway. These enzymes define a widespread Upr-like protein family (formerly DUF218) as polysaccharide release factors, revealing a unifying mechanism by which bacteria coordinate envelope assembly. Because all three Upr-like proteins share a conserved catalytic core, this family represents a potential target for antibiotics that simultaneously disrupt multiple envelope pathways.

microbiology↗

bioRxiv2026

Lack of adoption of computational gene-panel-selection tools in targeted spatial transcriptomics

Targeted spatial transcriptomics requires selecting a gene panel before any experiment is run. Choosing well across tens of thousands of candidate genes under platform-specific probe, panel-size, and sensitivity constraints is difficult to do manually, and impossible to correct once a panel has been used. A growing number of computational tools to automate this selection have been designed, yet it remains unclear whether they are adopted in practice. Several factors could limit uptake: validation uncertainty about panel performance, workflow friction between a ranked gene list and a submission-ready panel, limited platform awareness in tools developed for dissociated single-cell data, and the cost advantage of vendor panels. We analyzed 18 computational tools for gene-panel selection from two angles: their citation landscape and their use across a targeted spatial transcriptomics corpus. Across 540 full-text-classified records, only 35 (6.5%) reported using the cited tool for panel design; most came from computational, benchmark, review, or non-spatial papers, or papers that cited but did not use it. In an independent corpus, only 34 of 479 (7.1%) involved a computational tool, alone or in a hybrid workflow; most used manually curated, premade vendor, or modified vendor panels. Both lines of evidence point the same way: computational selection is rarely part of the reported experimental workflow, and it more often supplemented manual curation than replaced it. Adoption is unlikely to turn on further benchmark gains alone; it will depend on reducing the validation uncertainty and workflow friction that keep users on established panel-design practice.

bioinformatics↗

bioRxiv2026

An infection assay recapitulating Xanthomonas translucens pv. translucens natural entry into barley hydathodes

Pathogenic bacteria navigate biological host landscapes to enter and establish infection. How Xanthomonas phytopathogenic bacteria enter specific plant tissues is believed to directly determine the systemic vs. local outcome of infection. Non-vascular Xanthomonas infect leaf mesophyll locally from stomatal pores, while vascular ones enter through hydathode water pores on the leaf margin before invading the xylem for systemic movement. The pathovar Xanthomonas translucens pv. translucens (Xtt) is associated with barley infection, causing both bacterial leaf streak and bacterial blight symptoms by colonizing both the mesophyll spaces through stomatal guard cells and the plant vasculature, respectively. How Xtt reaches the vasculature in barley leaves is yet to be established. In this study, we determined the anatomy of barley hydathodes. A protocol allowing efficient hydathode infection by Xtt was established and caused bacterial blight symptoms, closely mimicking the natural infection process. This study establishes Xtt as one of the few demonstrated hydathode pathogens.

plant biology↗

bioRxiv2026

Dishevelled 2 coordinates aortic Notch signaling through primitive myeloid and hemogenic endothelial cells during hematopoietic stem cell specification

Hematopoietic stem cells (HSCs) arise from the hemogenic endothelium in the dorsal aorta through the endothelial-to-hematopoietic transition, a tightly regulated developmental process requiring coordinated signaling from the developing hematopoietic niche. Wnt and Notch signaling are essential for HSC specification, with Wnt functioning upstream of Notch; however, the molecular mechanisms by which these pathways are coordinated during HSC development remain poorly understood. Here, we identify Disheveled 2 (Dvl2), a key intracellular Wnt scaffold protein, as an essential regulator of HSC identity specification in zebrafish. Loss of dvl2 impaired HSC specification and reduced Notch signaling in the dorsal aorta, despite normal canonical Wnt activity in the developing endothelium, suggesting that Dvl2 regulates HSC specification independently of endothelial beta-catenin signaling. Hemogenic endothelial-specific expression of the Notch intracellular domain rescued HSC defects in dvl2 morphants, demonstrating that Dvl2 functions upstream of Notch activation in the hemogenic endothelium during HSC specification. In addition, loss of dvl2 reduced primitive macrophages and neutrophil populations. Restoration of primitive myeloid cells through gata1a knockdown rescued both endothelial Notch signaling and HSC defects, demonstrating that Dvl2 is also required for the supportive immune cells that induce HSC identity. Together, these findings reveal that Dvl2 coordinates hemogenic endothelial and immune niche cues to control HSC identity specification.

developmental biology↗

bioRxiv2026

Kainate receptors coordinate primitive hematopoiesis and hemogenic niche organization to promote hematopoietic stem cell development

The adult hematopoietic system is established during embryogenesis through the generation of hematopoietic stem and progenitor cells (HSPCs) within specialized transient niches, yet the signaling mechanisms that coordinate this process remain poorly understood. Here, we identify a developmental role for kainate receptors (KARs), glutamate-gated ion channels classically associated with excitatory neurotransmission, in coordinating primitive hematopoiesis and hemogenic niche formation in zebrafish. Loss of the KAR subunits encoded by grik1b or grik5 depletes primitive blood cells and disrupts hemogenic niche organization. These defects are accompanied by endothelial oxidative stress and impaired Notch signaling, which together reduce HSPC emergence and limit the clonal diversity of the adult hematopoietic stem cell pool. Shifting primitive hematopoietic output away from erythropoiesis and toward myelopoiesis restored macrophage niche colonization, endothelial Notch activity, and HSPC formation in KAR-deficient embryos. Moreover, hemogenic cell-specific Notch reactivation rescued HSPC production, demonstrating that KAR-dependent primitive hematopoiesis regulates this process through endothelial Notch signaling. Lastly, conserved KAR expression in endothelial and hematopoietic populations across zebrafish, mouse, and human suggests that these receptors may regulate blood development across multiple cellular components. Together, these findings reveal that glutamate receptor signaling extends beyond its well-established functions in neuronal transmission to coordinate the developmental processes that establish the blood system.

developmental biology↗

bioRxiv2026

Hippocampal and cortical oscillations support the encoding of event memories during movie watching

Remembering temporally-extended experiences requires integrating information over time into coherent event representations that can be stored and later retrieved. Oscillatory dynamics may provide a flexible mechanism to support these operations by coordinating communication across regions and organizing neural activity in time. Yet most evidence for the oscillatory basis of episodic encoding comes from discrete-stimulus paradigms, leaving unclear whether these mechanisms can help organize memory during continuous experience. Using iEEG recorded while participants watched a 50-min film, we measured how power, synchrony and phase alignment support memory formation at the beginning, middle, and end of the events. At event onset, remembered events showed reduced gamma power with increased hippocampal-frontal synchrony, indicating that successful events begin with transient long-range coordination rather than heightened local processing. Within an event, memory was predicted by the direction of change, with recalled events showing increases in temporal gamma power and hippocampal-temporal gamma synchrony, consistent with the progressive binding of incoming information into a developing event representation. At the end of an event, memory was supported by an increase in hippocampal theta power and phase resetting. Temporal gamma power was likewise elevated and hippocampal-temporal gamma synchrony increased, suggesting that information bound locally over the course of the event might be transmitted to the hippocampus as the event closes. Frontal beta power decreased at the same moment, consistent with disruption of the outgoing event model. Naturalistic memory formation therefore does not rest on a sustained encoding state, but on the appropriate temporal coordination and stability of oscillatory states across brain regions and event phases, in which stable oscillatory states support ongoing representation while transient, boundary-driven reorganization of hippocampal-cortical networks switches the system between consolidating the completed event and encoding the next.

neuroscience↗

bioRxiv2026

LipoGrid: A High-Throughput Multi-omics Perturbation Screen Dissects the Genetic Architecture of Lipid Metabolism

Lipids constitute one of the largest and most diverse classes of cellular molecules, sustaining membrane architecture, energy storage, and signaling. Consequently, their dysregulation underlies a broad spectrum of human disease. However, the genetic mechanisms governing lipid homeostasis have remained largely inaccessible, owing to the lack of approaches capable of systematically linking defined genetic perturbations to large-scale changes in cellular lipidome composition. Here we introduce LipoGrid, a spatial mass spectrometry platform that resolves the genetic architecture of lipid metabolism at single-cell resolution. LipoGrid arrays CRISPR/Cas9-perturbed cells on a micropatterned grid and sequentially captures lipidomic and gRNA identity from the same cells, complemented by single-cell RNA sequencing of matched cell populations subjected to the same perturbations. Using this approach, we quantified the relative abundance of 158 distinct lipid species across 143 target genes in a rigorously controlled experimental framework. We find that most gene knockouts produced measurable alterations in lipid composition, often affecting specific lipid classes and molecular subspecies. The screen accurately recapitulated established gene-lipid relationships, including enzyme-substrate specificities, lipid pathway regulators, and disease-associated loss-of-function phenotypes, thereby demonstrating the sensitivity and accuracy of LipoGrid. By jointly profiling transcriptomic and lipidomic responses, we further uncover compensatory feedback mechanisms that buffer the impact of genetic perturbations on the cellular lipidome. Collectively, these findings establish LipoGrid as a scalable multimodal platform for systematically mapping gene-lipid interactions and reveal the regulatory networks linking gene perturbation, transcriptional adaptation, and lipidome remodeling. HighlightsO_LIMicropatterned single-cell growth enables spatial lipidomic perturbation screens C_LIO_LILipoGrid maps 143 gene knockouts to 158 lipid species and transcriptomic states C_LIO_LIPerturbed lipidomes reveal compensatory feedback and lipid-class-specific uptake C_LIO_LIRecovers enzyme substrate specificities and disease-linked lipid signatures C_LI

systems biology↗

bioRxiv2026

FOCUS: end-to-end preprocessing, alignment and resolution-matched integration of spatial multi-omics data

SummaryIntegrating spatial multi-omics data requires coordinated preprocessing, cross-modality alignment and feature registration across modalities that differ in file format, coordinate system and spatial resolution. No existing tool addresses this pipeline end-to-end from raw experimental files till aligned data object. We present FOCUS, an open-source Python package that takes raw data from spatial transcriptomics, mass spectrometry imaging, Raman spectroscopy imaging and brightfield or fluorescence microscopy through modality-specific preprocessing, interactive spatial alignment and resolution-matching registration to a unified MuData object, driven by a single configuration file. Its modular, registry-based architecture allows straightforward extension to additional modalities. FOCUS is accessible via a command-line interface, a browser-based GUI and a Python API. Availability and implementationFOCUS is implemented in Python 3.11, with a browser-based GUI built on a Vue.js 3 frontend served by a Flask backend. Source code, documentation and container recipes are available at https://github.com/sifrimlab/FOCUS; a versioned release is archived on Zenodo (10.5281/zenodo.21700038). Outputs use the AnnData and MuData formats and are directly compatible with the scverse ecosystem.

bioinformatics↗

bioRxiv2026

Time cells lead neural reinstatement of episodic memory

The episodic memory system provides humans with a unique ability to form and retrieve rich and detailed memories. This capacity requires representing temporal context at encoding and recovering it at retrieval to drive successful recall. Time cells and ramping cells in human medial temporal cortex have been proposed as substrates of temporal context, but whether they participate in context recovery during retrieval remains unknown. Using microelectrode recordings from neurosurgical patients across two classical episodic paradigms, free recall and serial reconstruction, we identify time sensitive neurons and demonstrate that these neuron populations participate in neuronal assemblies and organized sequences on gamma and theta time scales respectively. Consistent with recovery of temporal context predicted by behavioral models of episodic processing, time cell firing precedes the broader assembly population and initiates sequential activation retrieval. These properties predict contextually-mediated recall behavior. We also identify phase coding by complementary populations as convergent mechanisms whose recruitment depends on task demands. Finally, we demonstrate the emergence of context-sensitive neurons in a naturalistic viewing dataset lacking the predictable temporal scaffold of canonical assays. These findings identify cellular mechanisms by which the human medial temporal cortex represents temporal context during episodic encoding and recovers it during retrieval.

neuroscience↗

bioRxiv2026

Thalamic Nuclei Differentially Coordinate Propagation of Cortical Slow Oscillations

Sleep slow oscillations (SOs) vary in the spatial extent of their cortical propagation, ranging from widespread Global events to spatially restricted Frontal events. These distinctions could have functional consequences for memory consolidation. It remains unknown whether individual thalamic nuclei are differentially engaged across SO propagation types, and whether thalamic activity before SO onset predicts subsequent propagation. We analyzed simultaneous scalp EEG and thalamic stereo-EEG from 24 full-night recordings in 6 epilepsy patients, sampling 11 thalamic nuclei and enabling direct characterization of nucleus-specific thalamocortical interactions during naturally occurring sleep slow oscillations. Cortical SOs were classified by propagation type and thalamocortical coupling was characterized via peri-event histograms, phase-locking analysis, waveform morphology comparisons, and pre-onset spectral and cross-frequency coupling features. All well-sampled thalamic nuclei showed broad temporal co-occurrence with prefrontal SOs, yet occupied distinct phase positions within the cortical SO cycle. Medial pulvinar (PuM) showed the strongest phase locking across subjects, with significant preference for the ascending post-trough phase. Among the examined nuclei, Frontal SOs were associated with significant phase locking in PuM, whereas Global SOs preferentially synchronized the centromedian nucleus near the cortical down-state trough. Thalamic waveform morphology differed systematically with propagation extent, with opposing effects between pulvinar and intralaminar nuclei. Pre-onset PuM activity showed suppressed alpha, sigma, and beta power and delta-phase cross-frequency coupling before Global SOs (all p < 0.05). The human thalamus engages NREM sleep SOs through nucleus-specific, propagation-sensitive dynamics rather than as a functionally uniform structure. Pre-onset PuM activity contains predictive signatures of cortical SO propagation extent. This finding has implications for closed-loop neuromodulation targeting specific SO states, although replication in larger cohorts is needed.

neuroscience↗

bioRxiv2026

Investigating the Dynamic Relationship Between Anxiety and Spatial Memory Using Autonomous Ecological Momentary Assessment

Anxiety has been extensively studied in relation to memory, yet its dynamic association with spatial episodic memory in naturalistic clinical settings remains largely unexplored. We developed an anxiety-spatial-memory EMA protocol (asm-EMA) and deployed it in 30 epilepsy patients undergoing inpatient EEG monitoring, delivering combined momentary anxiety ratings and a validated spatial memory task pseudo-randomly every 90-150 minutes across multiple days. Subject-level asm-EMA means and session-to-session variability both correlated significantly with standard neuropsychological assessments, supporting the clinical validity of our design. Elevated within-person STAI-6 was selectively associated with faster retrieval responses, yet spatial memory accuracy was independent of all three anxiety measures, suggesting a shift in response strategy rather than memory impairment. Within-day anxiety showed short-term carryover between consecutive sessions, with little persistence beyond the next session. The asm-EMA protocol provides a feasible, autonomous framework for capturing moment-to-moment anxiety-memory dynamics in naturalistic settings.

neuroscience↗

bioRxiv2026

Polysemanticity in human hippocampal neurons

To comprehend language, the brain must navigate a high-dimensional semantic landscape while seamlessly contextualizing meaning. Inspired by recent advances in the mechanistic interpretability of large language models (LLMs), we hypothesized that the brain utilizes polysemanticity, a coding strategy wherein individual neurons represent multiple semantically unrelated features through high-dimensional superposition (Elhage et al., 2022; Olah et al., 2020). We recorded single-unit activity from the human hippocampus during podcast listening. We found that hippocampal neurons exhibit dense semantic codes characterized by multiple tuning peaks with an overdispersed, isotropic geometry. This geometry satisfies the theoretical requirements for interference minimization in superimposed codes. Furthermore, semantic responses are strongly modulated by lexical and speaker-identity context; nonetheless, the underlying population geometry remains stable. This coding strategy permits rapid contextualization without requiring specialized, context-specific neurons. Indeed, we show clear pattern separation of similar terms, along with pattern completion for held-out words. Together, these results demonstrate that the human brain leverages superposition to solve a universal computational problem: maximizing semantic capacity within a constrained representational space.

neuroscience↗

bioRxiv2026

Dolichol biosynthesis in yeast traces the expanded reaction pathway of higher eukaryotes

The recent reassignment of Saccharomyces cerevisiae Dfg10 as a polyprenal reductase left two unresolved steps in yeast dolichol biosynthesis: the conversion of polyprenol to polyprenal and dolichal to dolichol. In humans, both of these steps are carried out by DHRSX, an enzyme with a unique dual specificity. We found that yeast Env9 catalyzes an NADPH-dependent reduction of dolichal to dolichol. However, in contrast to DHRSX, Env9 does not catalyze the conversion of polyprenol to polyprenal. Instead, our data indicates that this reaction is catalyzed by another yeast short chain oxidoreductase, Tda5. Thus, we provide evidence that the dual role of DHRSX in human dolichol synthesis is fulfilled by two dedicated yeast enzymes, Env9 and Tda5. Accordingly, deletion of ENV9 and TDA5 led to the accumulation of polyisoprenoid intermediates, transfer of immature lipid-linked oligosaccharides onto nascent proteins, and defective N-glycosylation. This is similar to what had been observed in DHRSX-deficient mammalian cells and yeast cells with DFG10 deleted. Furthermore, we discovered that loss of Dfg10 results in a deficiency of cell wall -mannan, revealing a critical sensitivity of yeast mannan biosynthesis to the quality of nascent N-linked glycans.

biochemistry↗

bioRxiv2026

Astrocyte Reactivity by Alcohol Dependence in the Central Amygdala

Astrocytes play essential roles in maintaining brain homeostasis and in contributing to synaptic functions, but, in response to injury, infection, or disease, astrocytes can downregulate their homeostatic and physiological functions while increasing neuroinflammatory responses. The central amygdala (CeA) is important for stress responsivity and the development of alcohol (ethanol) dependence. Using a multi-omics approach in Aldh1l1-EGFP/Rpl10a mice and the chronic intermittent ethanol two-bottle choice (CIE-2BC) model, we have characterized the translational response of CeA astrocytes, as well as the proteomic and phosphoproteomic changes in ethanol dependent, non-dependent, and naive mice. We identified astrocyte-specific alterations in neuroimmune functions and antioxidant/oxidative stress pathways in ethanol dependent mice as well as cytoskeletal plasticity related pathways in non-dependent mice. Proteomic analysis showed down-regulation of astrocyte physiological functions in dependent animals while phosphoproteomic analysis identified pathways associated with cytoskeleton remodeling in both dependent and non-dependent mice. Reconstructions of astrocyte morphologies demonstrated increased CeA astrocyte complexity in dependent and non-dependent groups compared to naive mice. The astrocyte-specific activation of neuroimmune and antioxidant pathways, down-regulation of homeostatic functions, alteration in protein phosphorylation-mediated cytoskeleton remodeling, and increased astrocyte morphological complexity demonstrate that ethanol dependence induces astrocyte reactivity in the CeA consistent with both adaptive and maladaptive changes. These findings highlight the role of CeA astrocytes in the progression from alcohol intake to dependence and represent a first step toward identifying astrocyte-specific therapeutic strategies to treat Alcohol Use Disorder (AUD) aimed at potentiating reactive astrocyte adaptive changes and inhibiting maladaptive responses.

neuroscience↗

bioRxiv2025

Hidden Spirals Reveal Neurocomputational Mechanisms of Traveling Waves in Human Memory

While traveling waves are often described as planar propagations across cortex, recent theoretical work predicts that more complex spatial patterns, including spiral dynamics, could organize large-scale neural computations but remain difficult to detect in the human brain. To investigate this, we analyzed direct brain recordings from humans performing a working memory task. To characterize traveling wave patterns, we used independent component analysis, and showed that traveling waves propagated along the cortex in complex spatial patterns that correlated with behaviors such as memory encoding, maintenance, and retrieval. We then developed a novel computational framework based on coupled phase oscillators to model these distinct wave patterns. This computational approach revealed hidden spirals that were not visible in the original recordings. The center of these hidden spirals shifted across the cortex to distinguish separate behavioral states, such as memory encoding and retrieval. Together, these findings reveal that cortical traveling waves are governed by latent spiral attractor dynamics and suggest that rotating wave architectures provide a fundamental neurocomputational mechanism for flexible human memory processing.

neuroscience↗

bioRxiv2025

Multi-omics integration and batch correction using a modality-agnostic deep learning framework

State-of-the-art biotechnologies allow the detection of different molecular species on the same biological sample, generating complex highly-dimensional multi-modal datasets. Gaining a holistic understanding of biological phenomena, such as oncogenesis or aging, requires integrating these diverse modalities into low-dimensional data representations while correcting for technical artifacts. Here we present MIMA, a modular, unsupervised AI framework for multi-omics data integration and batch correction. Applied to complex spatial and single-cell datasets, MIMA effectively removes batch effects, while preserving biologically relevant information, and learns representations predictive of expert pathologist annotations. Additionally, it enables cross-modal translation, uncovers molecular patterns not captured by manual annotations, and despite being modality-agnostic performs on par with specialized state-of-the-art tools. MIMAs flexibility and scalability make it a powerful tool for multimodal data analysis. MIMA provides a foundation for AI-based, multi-omics augmented digital pathology frameworks, offering new opportunities for improved patient stratification and precision medicine through the comprehensive integration of high-dimensional molecular data and histopathological imaging.

bioinformatics↗