Biology subjects

Lakatos, A.

Publications and source records attributed to Lakatos, A..

10 recordsLinked to original sources

bioRxiv2026

scRepresenter: a workflow for computing, integrating and benchmarking cellular representations in single-cell transcriptomics

MotivationSingle-cell RNA sequencing (scRNA-seq) has become an attractive tool for studying complex diseases, in which transient cell states affecting diverse cell populations characterise disease development and progression. However, due to data sparsity and disease heterogeneity analysis is often challenging. With recent advances in machine learning, two widely used approaches have emerged for learning cellular representations: large-scale foundation models and biological knowledge-guided methods. Despite their complementary strengths, there is currently no unified workflow for systematically comparing and integrating these approaches. ResultsHere, we present scRepresenter, an open-source workflow for computing, integrating, and validating cellular embeddings derived from foundation models and biological knowledge-guided methods in the context of complex diseases. It consists of two components: a command-line workflow that computes cellular embeddings and performs downstream analyses, and an interactive Shiny application for visualizing and comparing the computed embeddings. scRepresenter supports four categories of cellular representations: (1) expression-based, (2) knowledge-guided, (3) foundation model-derived, and (4) hybrid embeddings that combine foundation model-derived representations with knowledge-guided representations. This approach takes a cell-by-gene count matrix as input and outputs an integrated object containing the computed embeddings. Then, this object can be uploaded into our interactive Shiny application to compare different embeddings. AvailabilityThe workflow is available at https://github.com/GuilhermePocas/scRepresenter ContactAL291@cam.ac.uk; MA2129@cam.ac.uk

bioinformatics↗

bioRxiv2026

Organoids reveal niche-specific mechanotransduction-guided human cortical patterning and cell fate acquisition

The highly dynamic mechanical environment of the developing cerebral cortex has the potential to encode information vital for neuronal differentiation. Changes in local biophysical properties have been implicated in processes ranging from the maintenance of neural progenitors to long-range axon guidance. However, whether cortical mechanics influence cellular organisation and specification during neurogenesis remains largely unexplored. Here, we leverage a 3D mosaic organoid model of human cortical development, in which we selectively disrupt a key component of nuclear mechanosensing, the LINC complex, decoupling cells from their mechanical environment. We show that LINC-decoupling alters nuclear morphology in a compartment-specific manner, driving preferential exclusion from the germinal zone and concomitant premature differentiation. Excluded cells exhibit a biased spatial distribution in the cortical plate and altered fate allocation, with loss of intermediate progenitors and upper-layer neuron populations. Combining this approach with single-cell transcriptomic profiling, we reveal a signature of impaired ERK activation and density sensing in LINC-decoupled cells. Furthermore, we show that altered density sensing contributes to mislocalisation of the nuclear envelope protein emerin and disruption of histone mark deposition during differentiation. Taken together, our findings illustrate how the dynamic mechanical environment of a complex tissue can dictate cell fate and pattern formation during development.

developmental biology↗

bioRxiv2026

Myelin basic protein is an RNA chaperone in microglial nuclear retro-transport

CNS oligodendrocytes generate myelin, an RNA-containing proteolipid substance that enhances axonal transmission. In multiple sclerosis (MS), myelin debris is phagocytosed by microglia (MG), and prior studies have detected myelin-derived mRNA in MG nuclei, suggesting a retrograde transport pathway. We report myelin basic protein (MBP) is a nucleic acid-binding and trafficking protein. We found that retro-transport of myelin RNA into the MG nucleus was phagocytosis and importin-dependent. Transcriptomic and proteomic analyses of MG nuclei revealed enrichment of myelin mRNAs and proteins, with MBP singularly detected in soluble and chromatin-associated fractions. MBP bound mRNA with high affinity (Kd {approx} 0.30 nM) and was sufficient to facilitate MG RNA nuclear import in vitro and in vivo. Functionally, MBP mediated the delivery of small interfering RNAs for targeted knockdown of toll-like receptor 4. These findings indicate MBP as an RNA-binding protein capable of MG nuclear import, providing insight into neuroinflammatory pathology of MS.

neuroscience↗

bioRxiv2026

An Intermediate Mesoderm Premyogenic Niche Supports Early Human Myogenic Lineage Progression

Transient cell states that precede and support human myogenic lineage commitment, and the intrinsic and extrinsic signals that control them, remain poorly defined in vitro. Here, we used longitudinal single-nucleus profiling, together with a SIX1:H2B-GFP hPSC reporter for lineage tracing, resolved previously uncaptured transient intermediates and sequential waves of human myogenesis across differentiation and in vivo. We show that hPSC-directed myogenesis gives rise in parallel to paraxial mesoderm and a transient PAX8+ intermediate mesoderm population that forms a 3-dimensional pre-myogenic niche supporting the PAX3-to-PAX7 myogenic progenitor transition. LIANA+ analysis further identified a temporally restricted BMP7-BMPR1B interaction, together with laminin-linked signaling, between PAX8+ niche cells and skeletal muscle progenitors before commitment. We further show that dynamic SIX1 cofactor switching, including EYA3 activity, is required for PAX3-to-PAX7 progression, and that disruption of this program compromises multi-lineage niche integrity. Together, these findings define how transient niche populations and intrinsic regulatory networks coordinate early human myogenic lineage progression and provide a human in vitro platform to study parallel intermediate and paraxial mesoderm development during myogenesis. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=129 SRC="FIGDIR/small/715044v1_ufig1.gif" ALT="Figure 1"> View larger version (44K): org.highwire.dtl.DTLVardef@189294dorg.highwire.dtl.DTLVardef@1f6b669org.highwire.dtl.DTLVardef@41cc68org.highwire.dtl.DTLVardef@1d2cb42_HPS_FORMAT_FIGEXP M_FIG C_FIG HighlightsO_LISIX1+PAX8+ niche progenitors promote myogenic differentiation via BMP7- and laminin-dependent signaling. C_LIO_LILoss of SIX1-EYA3 activity disrupts the pre-myogenic niche and impairs the PAX3-to-PAX7 transition C_LIO_LIMulti-omics single-cell optimal transport resolves previously uncaptured transient intermediates and sequential waves of human myogenesis C_LIO_LISIX1 lineage tracing identifies CREB5 as a top regulator of the PAX7+ state in human myogenesis C_LI

cell biology↗

bioRxiv2025

MYO1F interactome reveals the SH3-domain linked CASS complex at podosomes and the phagocytic cup

MYO1F, a long-tailed myosin of class I, is selectively expressed in immune cells and upregulated in microglia associated with neurodegenerative pathogenesis. The intracellular functions of myosin motors involve adaptor proteins, which regulate cargo attachment and intracellular motor recruitment. To define the MYO1F interactome, we performed an in situ proximity labelling-based proteomic analysis in human myeloid U937 cells. We identified a distinct SH3-domain-dependent adaptor module comprising CD2AP, ASAP1, SH3BP2, and SH3KBP1 (CASS complex), which localizes with MYO1F in podosomes and phagocytic cups. Structural modelling and mutagenesis confirmed multivalent proline-rich motif interactions of the CASS complex with the MYO1F SH3 domain. Further deletions revealed a second group of membrane-associated adaptor proteins that bind to the MYO1F pleckstrin homology (PH) domain. Immunofluorescence in macrophages and microglia confirmed the conserved localization of MYO1F and its adaptors at actin-rich podosomes and phagocytic cups. Functional assays demonstrated that MYO1F recruitment to the phagocytic cup requires motor activity and intact PH and SH3 domains. This study provides the first comparative interactome of MYO1F and its paralogue MYO1E and supports a role for MYO1F in podosomes and during phagocytosis in both peripheral and brain-resident myeloid cells.

cell biology↗

bioRxiv2025

Mitochondrial DNA heteroplasmy drives cortical neuronal disturbances in human organoids harbouring the common m.3243A>G mutation

Mitochondrial diseases frequently affect the brain leading to severe and disabling neurological symptoms. The heteroplasmic m.3243A>G mutation in MT-TL1, encoding mt-tRNALeu, is responsible for [~]80% of mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS), which is one of the most characteristic mitochondrial syndromes, leading to disability and early death. There are no animal models harbouring this mutation to provide precise mechanistic insights informing therapeutic interventions. Here, we generated a human iPSC-derived cerebral organoid slice model that recapitulates cortical architecture and mitochondrial pathology. Using biological assays and single-cell RNA sequencing, we uncovered heteroplasmy-dependent transcriptional shifts and changes in key cellular processes in cortical neurons. Organoids with high heteroplasmy showed a predominant impairment of deep-layer neurons triggered by mitochondrial stress, leading to axonal degeneration and apoptosis, similar to brain autopsy of a MELAS patient. Our findings provide insights into the vulnerability of long-range projection neurons in mitochondrial diseases, advancing our understanding of disease mechanisms with a view to potential therapeutic strategies.

neuroscience↗

bioRxiv2024

A chemically defined and xeno-free hydrogel system for regenerative medicine.

Biofabricated scaffolds facilitate bona fide cellular interactions, cell type specification, and the formation of three-dimensional tissue architecture from human pluripotent stem cells (hPSCs). However, xenogenic biomaterials are poorly defined, and synthetic biomaterials remain underdeveloped and understudied, hindering regulatory approval for clinical use and preventing the translation of lab-grown therapies. Here, we describe a protein screen-based hydrogel system biofabricated from clinical-grade human components. We show that "Alphagel", a base hydrogel comprising human embryonic matrices, supports the trilineage differentiation of hPSCs into neural, cardiac, and liver tissue. Alphagel is also shown to be biocompatible and biodegradable in vivo. Further, upon adding select proteins from maturing human foetal liver to Alphagel, we show that the resulting hydrogel (termed "Hepatogel") enhances the differentiation of hPSC-derived hepatocytes (H-iHeps) compared with Matrigel. Importantly, when injected into mice livers, Hepatogel significantly improves the retention of H-iHeps compared to standard aqueous cell injections. Altogether, our results provide proof of concept that customisable and organ-specific hydrogel systems are a valuable tool for developing clinically translatable therapies for regenerative medicine and tissue engineering.

bioengineering↗

bioRxiv2024

Predicting immune checkpoint therapy response in three independent metastatic melanoma cohorts

While Immune checkpoint inhibition (ICI) therapy shows significant efficacy in metastatic melanoma, only about 50% respond, lacking reliable predictive methods. We introduce a panel of six proteins aimed at predicting response to ICI therapy. Evaluating previously reported proteins in two untreated melanoma cohorts, we used a published predictive model (EaSIeR score) to identify potential proteins distinguishing responders and non-responders. Six proteins initially identified in the ICI cohort correlated with predicted response in the untreated cohort. Additionally, three proteins correlated with patient survival, both at the protein, and at the transcript levels, in an independent immunotherapy treated cohort. Our study identifies predictive biomarkers across three melanoma cohorts, suggesting their use in therapeutic decision-making. O_FIG O_LINKSMALLFIG WIDTH=164 HEIGHT=200 SRC="FIGDIR/small/592032v1_ufig1.gif" ALT="Figure 1"> View larger version (63K): org.highwire.dtl.DTLVardef@754668org.highwire.dtl.DTLVardef@1616263org.highwire.dtl.DTLVardef@827009org.highwire.dtl.DTLVardef@1899ba7_HPS_FORMAT_FIGEXP M_FIG C_FIG

cancer biology↗

bioRxiv2024

MEA-NAP compares microscale functional connectivity, topology, and network dynamics in organoid or monolayer neuronal cultures

Microelectrode array (MEA) recordings are commonly used to compare firing and burst rates in neuronal cultures. MEA recordings can also reveal microscale functional connectivity, topology, and network dynamics--patterns seen in brain networks across spatial scales. Network topology is frequently characterized in neuroimaging with graph theoretical metrics. However, few computational tools exist for analyzing microscale functional brain networks from MEA recordings. Here, we present a MATLAB MEA network analysis pipeline (MEA-NAP) for raw voltage time-series acquired from single- or multi-well MEAs. Applications to 3D human cerebral organoids or 2D human-derived or murine cultures reveal differences in network development, including topology, node cartography, and dimensionality. MEA-NAP incorporates multi-unit template-based spike detection, probabilistic thresholding for determining significant functional connections, and normalization techniques for comparing networks. MEA-NAP can identify network-level effects of pharmacologic perturbation and/or disease-causing mutations and, thus, can provide a translational platform for revealing mechanistic insights and screening new therapeutic approaches.

neuroscience↗

bioRxiv2018

Cerebral organoids at the air-liquid interface generate diverse nerve tracts with functional output

Three-dimensional neural organoids are emerging tools with the potential for improving our understanding of human brain development and neurological disorders. Recent advances in this field have demonstrated their capacity to model neurogenesis1,2, neuronal migration and positioning3,4, and even response to sensory input5. However, it remains to be seen whether these tissues can model axon guidance dynamics and the formation of complex connectivity with functional neuronal output. Here, we have established a longterm air-liquid interface culture paradigm that leads to improved neuronal survival and allows for imaging of axon guidance. Over time, these cultures spontaneously form thick axon tracts capable of projecting over long distances. Axon bundles display various morphological behaviors including intracortical projection within and across the organoid, growth cone turning, decussation, and projection away from the organoid. Single-cell RNA-sequencing reveals the full repertoire of cortical neuronal identities, and retrograde labelling demonstrates these tract morphologies match the appropriate molecular identities, namely callosal and corticofugal neuron types. We show that these neurons are functionally mature, generate active networks within the organoid, and that extracortical projecting tracts innervate and activate mouse spinal cord-muscle explants. Muscle contractions can be evoked by stimulation of the organoid, while axotomy of the innervating tracts abolishes the muscle contraction response, demonstrating dependence on connection with the organoid. Overall, these results reveal a remarkable selforganization of corticofugal and callosal tracts with a functional output, providing new opportunities to examine relevant aspects of human CNS development and response to injury.

neuroscience↗