Biology subjects

Cai, J.

Publications and source records attributed to Cai, J..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Microencapsulation of a novel Bacteroides thetaiotaomicron strain: a promising strategy to fortify intestinal barrier fortification in weaned pig model

Porcine Bacteroides thetaiotaomicron LYH5 demonstrated in vitro antimicrobial activity, suggesting probiotic potential. Due to poor gastric juice tolerance, LYH5 was encapsulated via extrusion using sodium alginate (SA) and gellan gum. Box-Behnken design optimization yielded optimal parameters: SA 1.5%, gellan gum 0.4%, CaCl2 0.9%, bacteria:glue ratio 1:4, achieving an encapsulation rate of 84.22{+/-}0.17%. Its effect on weaned piglet intestinal health was evaluated using 78 piglets (7.69{+/-}0.52 kg) randomly assigned to 4 groups for 40 days: CON (control), T (basal diet + LYH5 live bacteria, 1x10{superscript 1} CFU/mL), TJ (basal diet + LYH5 microcapsules, 1x10{superscript 1} CFU/mL, J (basal diet + empty capsules). The results of this experiment showed that compared with the control group, LYH5 microcapsule can improve the intestinal barrier function without affecting the growth performance of piglets, and provide ideas and references for the development of human next-generation probiotics (NGP). IMPORTANCEThis study addresses the key bottleneck of poor gastric acid tolerance of probiotics via microencapsulation and provides a practical reference for the development of human next-generation probiotics.

microbiology↗

bioRxiv2026

Diffusion-ACP39: A Decoder-Adaptive Latent Diffusion Framework for Generative Anticancer Peptide Discovery

Cancer remains a major global health threat, with its incidence and mortality rates consistently rising in recent years. Anticancer peptides (ACPs) are short amino acid chains that can inhibit the growth or spread of cancer cells. Compared to traditional treatments, ACPs are a promising class of potential cancer therapies due to their multiple mechanisms, potential for combination cancer therapy, enhanced immune function, lower toxicity to normal tissues, fewer side effects, and less drug resistance. Although it is necessary to explore novel ACPs, traditional wet-lab methods for selecting them are labor-intensive, time-consuming, and expensive. To accelerate the discovery of novel ACPs, we proposed Diffusion-ACP39, a latent diffusion-based generative model with synchronized seed autoencoder for anticancer peptide design, capable of generating novel peptides with lengths ranging from 5 to 39 amino acids. Furthermore, we developed RF-ACP39, a random forest classifier model to assess the generative power of Diffusion-ACP39. Finally, Diffusion-ACP39 achieved an accuracy of 94.5% when generating 10,000 peptides with RF-ACP39. We also qualitatively analyzed the differences among true ACPs, random sequences, random peptides, and generated ACPs, demonstrating that the generated ACPs are most similar to true ACPs.

bioinformatics↗

bioRxiv2026

Membrane Tension Integrates Physical and Signaling Cues to Gate Cell Fate Transitions

Physical forces shape cell behavior, yet how they integrate with signaling to control fate and disease remains unclear. The alveolar epithelium is patterned by FGF signaling and mechanical stretch, but how these cues specify AT1 and AT2 cells is poorly understood. Here we show that cell membrane tension (CMT) is a conserved regulator of epithelial fate in mouse and human lungs. CMT drops before differentiation and is spatially patterned, defining where bipotent progenitors acquire AT1 or AT2 identity. Lower CMT enhances FGFR2 endocytosis and ERK signaling to drive AT2 differentiation and permits architectural remodeling that enables stretch-mediated YAP/TAZ nuclear entry for AT1 maturation. {beta}-catenin elevates CMT cell-intrinsically independent of its role in canonical WNT transcription, while embedding, osmotic compression, or fibroblast wrapping elevate CMT extrinsically. Combined intrinsic and extrinsic tension traps alveolar epithelial cells in a KRT8 transitional state seen in fibrotic lungs. Membrane tension thus integrates physical and molecular cues linking morphogenesis to fibrosis. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=160 HEIGHT=200 SRC="FIGDIR/small/708749v1_ufig1.gif" ALT="Figure 1"> View larger version (52K): org.highwire.dtl.DTLVardef@1a9c04forg.highwire.dtl.DTLVardef@190bee2org.highwire.dtl.DTLVardef@1e41664org.highwire.dtl.DTLVardef@17a3ec7_HPS_FORMAT_FIGEXP M_FIG C_FIG In BriefWe identify a conserved drop in cell membrane tension (CMT) that gates epithelial fate transitions across mouse and human lungs. In bipotent progenitors, reduced CMT promotes Alveolar Type 2 (AT2) fate via FGFR2 endocytosis and ERK signaling, as well as Alveolar Type 1 (AT1) fate by enabling architectural remodeling required for YAP/TAZ nuclear entry. We further show that intrinsic {beta}-catenin and extrinsic confinement cues, including mesenchymal contact, converge to elevate CMT, restricting differentiation and--in adult AT2s--inducing a KRT8 transitional state associated with fibrosis. HighlightsO_LICell membrane tension drops before differentiation and is required for AT1 and AT2 fate acquisition. C_LIO_LIReduced tension enhances FGFR2 endocytosis and ERK signaling to drive AT2 specification. C_LIO_LIReduced tension permits architectural remodeling and YAP/TAZ nuclear entry required for AT1 maturation. C_LIO_LIIntrinsic {beta}-catenin and extrinsic confinement elevate CMT to restrict differentiation and together strongly induce a KRT8 transitional state associated with fibrosis. C_LI

cell biology↗

bioRxiv2026

Benchmarking niche identification via domain segmentation for spatial transcriptomics data

Tissue niches are spatially organized microenvironments in which coordinated multicellular interactions shape cellular states and biological functions. Currently, niche identification is routinely performed using domain segmentation frameworks. While interrelated, spatial domains and niches are not fundamentally equivalent. The former emphasizes intra-domain compositional consistency and transcriptomic homogeneity, whereas the latter is defined by the emergent properties of localized signaling gradients and the functional reciprocity between key cell lineages. Here, we present a high-resolution reference by thoroughly annotating single-cell resolution CosMx ST data of a human follicular lymphoid hyperplasia lymph node, a dynamic, non-compartmentalized tissue containing several critical immune niches defined by specific lineage architectures. We systematically benchmarked 16 contemporary domain segmentation algorithms, demonstrating that most methods in their default configurations fail to recapitulate biologically defined niche boundaries. Our analysis reveals that the definitive, disjoint spatial distributions of key functional lineages are frequently obscured by the stochastic infiltration of peripheral cell types. Such reduction in the spatial signal-to-noise ratio represents a primary bottleneck for existing algorithms, which prioritize local transcriptomic variance over global architectural logic. Following this observation, we demonstrate that strategic weighting of core functional lineages can restore the resolution of spatial niches in select domain segmentation frameworks. Cross-comparison against compartmentalized tissues further underscores the unique challenges of niche identification in non-mechanically separated environments and clarifies the fundamental divergence between structural domain segmentation and functional niche discovery. Our work delineates the limitations of current paradigms and advocates for the development of specialized computational approaches tailored specifically to the complexity of functional microenvironments.

bioinformatics↗

bioRxiv2026

Information Geometry Reconciles Discrete and Continuous Variation in Single-Cell and Spatial Transcriptomic Analysis

Single-cell and spatial transcriptomics provide high-resolution cellular characterization, yet standard analytical approaches remain theoretically misaligned with the probabilistic nature of the data. After UMI normalization, current pipelines rely on Euclidean or log-transformed Euclidean distance for similarity measurement. Both are fundamentally ill-suited to model the multinomial count data. Euclidean distance in normalized space overemphasizes high-variance genes, while log-transformation inverts this bias but at the cost of distorting subtle, continuous expression modulations. Neither approach naturally captures the dual nature of gene expression: both discrete presence/absence transitions and continuous quantitative variation. To overcome these limitations, we introduce GAIA (Geometric Analysis from an Information Aspect), an information-geometric framework for cell representation learning and inter-cell similarity measurement. By anchoring analysis in the true probabilistic model, treating cells as multinomial distributions over genes and projecting cells to a statistical manifold, GAIA organically reconciles both the presence/absence effect and the more continuous expression modulations. Mathematically, GAIA exploits the equivalence between Fisher-Rao distance in multinomial space and geodesic distance on the unit hypersphere, a property that enables both theoretical guarantees and computational efficiency. Experiments in synthetic and real scRNA-seq and spatial transcriptomic datasets demonstrate that GAIA preserves robust and consistent cell-to-cell relationships, delineates biologically nuanced sub-types, mitigates batch effects arising from sequencing depth variation, and eliminates the dependence on knowledge-restricted gene selection for learning meaningful cell representations. Overall, GAIA offers a knowledge-lean, variance-stabilizing framework for analyzing single-cell and spatial transcriptomic data, enhancing discrimination between nuanced cell sub-type and -states.

bioinformatics↗

bioRxiv2026

Gene-First Identity Construction for Robust Cell Identification in Single-Cell Transcriptomics

The precise delineation of cell types is fundamental to single-cell transcriptomics, yet current clustering pipelines often violate an axiomatic principle: hierarchical consistency. Existing methods measure cell-to-cell distances within a fixed global feature space, disregarding the fact that biological distinctions are inherently context-dependent lineage separation requires different gene programs than subtype resolution. Mathematically, this implies that the similarity metric itself should not be a static functional, but a pair-dependent energy functional evaluated within a specific Hilbert subspace determined by the biological comparison at hand. The challenge lies in the fact that allowing pair-dependent metrics typically destroys the global geometric consistency required for downstream analysis, unless the family of Hilbert subspaces is given strong biological structure. To resolve this geometric dilemma, we introduce GeCCo (Gene Co-expression Constructed identity), which constructs identities by projecting cells onto a rigorously derived hierarchy of gene programs. To construct this hierarchy, GeCCo first quantifies Boolean regulatory logic via the{phi} coefficient, and subsequently employs a greedy topological inference to organize genes based on their synergistic and antagonistic relationships. Benchmarking on human immune atlases demonstrates that GeCCo achieves superior hierarchical consistency, ensuring that globally inferred cell identities rigorously match locally refined subtypes. Furthermore, in pancreatic endocrine progenitors, GeCCo resolves a hidden mitotic bridge state, suggesting a concentrated division phase prior to differentiation. Ultimately, GeCCo shifts the paradigm from ad hoc clustering to programmatic cell typing, offering a mathematically grounded framework for scalable atlases of cellular discovery.

bioinformatics↗

bioRxiv2026

BRD8 Disruption Unlocks Retrotransposon-Driven Tumor-Intrinsic Immunogenicity via Redistribution of Histone Acetylation in Liver Cancer

Overall response rate for immune checkpoint blockade (ICB) therapy remains limited in advanced hepatocellular carcinoma (HCC). Epigenetic perturbation has been shown to enhance tumor-intrinsic immunogenicity and thus reverse resistance to immunotherapy in multiple cancer types. Here, through domain-focused CRISPR/Cas9 genetic screens targeting chromatin regulators, we identified the bromodomain-containing protein 8 (BRD8) as a novel epigenetic factor that functions to suppress tumor-intrinsic immunity through BRD8/EP400 chromatin remodeling complex in HCC. Depletion of BRD8 potently induces anti-tumor immune responses and sensitizes tumors to ICB therapy in mouse tumor models. Mechanistically, BRD8 knockout leads to partial redistribution of H3K27ac to repetitive elements, activating double-stranded RNA (dsRNA) accumulation and subsequent type I interferon (IFN) response, leading to tumor rejection via remodeling the tumor microenvironment. The first-in-class selective chemical probe DN02 targeting the bromodomain of BRD8 also exhibits immunomodulatory and anti-tumor effects. In line with these findings, low BRD8 expression correlated with enriched interferon signatures, elevated immune cell infiltration, and prolonged survival in HCC patients. Together, this study reveals BRD8 as a selective epigenetic vulnerability for promoting tumor immunogenicity, and its bromodomain as a druggable target to overcome immunotherapy resistance in liver cancer patients. SignificanceThe induction of viral mimicry by epigenetic interference could trigger tumor-intrinsic immune response and enhance immunotherapy efficacy. Here, through domain-focused CRISPR/Cas9 screening, we identify BRD8 complex as a regulator involved in the suppression of repetitive elements by controlling H3K27ac distribution and propose BRD8 as a novel target to enhance anti-tumor immunity.

cancer biology↗

bioRxiv2025

An obligatory role for AgRP neurons in maintaining body temperature during time-restricted feeding

Homeotherms maintain a steady body temperature through thermoregulation, a process critical for survival during fasting, in which the brain has to defend energy-costly body temperature while reducing energy expenditure to conserve energy reserve; however, the neural basis for defending body temperature remains unclear. Here, we demonstrated that AgRP neuron lesion led to lethality during time-restricted feeding but no obvious impact on HFD-induced obesity or obesity-reducing responses to glucagon-like peptide-1 receptor agonism. The lesion disrupted adaptive feeding behaviors during time-restricted feeding and reduced motivational feeding. Notably, the lethality was caused by hypothermia instead of reduced food intake. The lesion also caused failure in body temperature maintenance during acute fasting in cold. Fasting-induced activation in AgRP neurons was abrogated when mice were placed in a warm environment. Our results identify the physiological role for AgRP neurons in defending body temperature during restricted feeding but dispensable for body weight regulation with food ad libitum.

neuroscience↗

bioRxiv2025

ClsDiff-AMP30: Generating Antimicrobial Peptides by a Classifier Guidance Noise Predictor

Antimicrobial peptides (AMPs) represent a promising therapeutic strategy to combat the increasing challenge of multidrug-resistant pathogens, a crisis intensified by the overuse of conventional antibiotics. In addition to their broad-spectrum antimicrobial activity, low toxicity, and reduced propensity for resistance development, AMPs offer significant advantages over traditional antibiotic therapies. However, the discovery of novel AMPs through biological experiments remains constrained by high costs, labor-intensive workflows, and time-consuming procedures, underscoring the urgent need for in silico computational methods to design AMP sequences. Notably, shorter AMPs ([≤] 30 residues) demonstrate superior antimicrobial efficacy, improved structural stability, and minimal cytotoxicity toward human cells. To address these challenges, we present a classifier-guided diffusion framework specialized for generating AMPs shorter than 30 residues (ClsDiff-AMP30). The architecture integrates two interdependent submodels, including a noisy AMP classifier that evaluates AMP likelihood at intermediate denoising steps and a noise predictor guided by classifier-derived probability scores, dynamically adjusted via a self-optimized coefficient to modulate guidance strength. ClsDiff-AMP30 achieves a validation accuracy of 66% across 10,000 synthesized sequences by a self-developed AMP classifier. Furthermore, wet lab experiments demonstrated that all 11 selected sequences exhibited high antimicrobial activity against at least one of the three tested bacterial strains and low hemolytic activity.

bioinformatics↗

bioRxiv2025

A Population-Specific PARP1 Gene Variation Modulates PARP Trapping

Poly-(ADP-ribose) polymerase inhibitors (PARPi) block NAD+-binding pocket of PARP1, inhibiting PAR synthesis. However, they differ in their ability to retain PARP1 on damaged DNA to induce synthetic lethality in homologous recombination-deficient (HRD) cancer. Allosteric enzymatic activation requires destabilization of the helical domain (HD) of PARP1 and is indispensable for activation and chromatin retention induced by distinct PARPi. Here we report that the effect of the clinical PARPi talazoparib is robustly impacted by a common human polymorphism within the HD. PARP1V762 greatly enhanced talazoparib-driven allosteric retention on chromatin, prolonged XRCC1 recruitment, and enhanced cell killing. Talazoparib switches from Type-II PARPi behavior in PARP1A762 to allosteric, pro-retention Type-I behavior for PARP1V762. Thus, both PARPi efficacy and dose-limiting tolerability depends on PARP1 allele, motivating variant-guided cancer therapies. TEASEROne of the four FDA-approved PARPi drugs, talazoparib, is modulated by a PARP1 SNP that is widespread in the population.

cancer biology↗

bioRxiv2025

Identification of Protein Cargo in Extracellular Vesicles from Macrophages in Progressing and Regressing Tumors

Exosomes are extracellular vesicles (EVs) that carry bioactive molecules from a cell of origin and may alter gene expression of the acceptor cell via binding to cell surface receptors and/or delivering their cargo into the cell. Tumor-associated macrophages (TAMs) can increase or decrease tumor growth through various mechanisms, but the impact of TAM-EVs on tumors has been difficult to elucidate due to challenges in isolating and culturing TAMs for EV purification. In this study, we set out to uncover the protein identities within EVs from TAMs in progressing and regressing tumors, thereby uncovering EV proteins roles in TAM-tumor crosstalk. TAMs were purified from tumors using magnetic bead isolation and cultured for up to 9 days. TAMs from regressing tumors were found to be more M1-like by high expression of major histocompatibility complex (MHC) class II, while TAMs from progressing tumors are more M2-like (low expression of MHC class II). Long-term in vitro culture of TAMs resulted in reduced expression of MHC class II. EVs were harvested from plated TAMs, and EV proteins were identified by mass spectrometry and compared to EV proteins from bone marrow-derived macrophages (BMDMs). Unsupervised hierarchical clustering studies revealed that proteins from TAM-EVs converge based on their in vitro culture duration rather than their tumor origin. Signature proteins in TAM-EVs were characterized, with the identification of galectin-3-binding protein as a signature protein present in EVs from M1-like BMDMs and regressor macrophages. Finally, immunostimulatory pathways were identified in progressor-TAM-EVs through protein-protein interaction network analysis.

immunology↗

bioRxiv2025

BERT-T6: Towards High-accuracy T6SS Bacterial Toxin Identification Using Protein Language Model

Type VI secretion system effectors target the cell wall, membranes and nucleic acids, leading to the killing of bacteria or impairment of host cell defense mechanisms. Accurate identification of T6SEs will be beneficial to understand the virulence of these bacteria via type VI secretion systems as well as bacterial pathogenesis. Although some traditional machine learning-based and deep learning-based tools have been developed to distinguish T6SEs from non-T6SEs, we believe there is still room for further improvement. To obtain the robust feature for model construction, we successively investigate various classic sequence-based features and embeddings from pre-trained transformer-based protein language models. Building upon the model incorporating ProtBert embeddings, we employed a transfer learning approach to fine-tune the ProtBert protein language model with a downstream T6SE classification task. The resulting BERT-T6 model demonstrates performance significantly superior to baseline models. More importantly, with an accuracy of 0.959, a sensitivity of 0.909, a specificity of 0.973, a precision of 0.905, a F1-score of 0.907, MCC of 0.881, our model achieves performance competitive with state-of-the-art binary and multi-class predictors. This work highlights the effectiveness of utilizing BERT with transfer learning for T6SE prediction. BERT-T6 provides a robust and precise approach for identifying T6SEs, offering promise for enhancing studies of bacterial virulence mechanisms.

bioinformatics↗

bioRxiv2025

SpaDiff: Denoising for Sequence-based Spatial Transcriptomics via Diffusion Process

Spatial transcriptomics enables transcriptome-scale analysis with spatial resolution but suffers from spot-swapping, where RNA molecules drift from their true locations, introducing noise and reducing spatial specificity. We introduce SpaDiff, a denoising method that treats spot-swapping as a diffusion process. SpaDiff simulates the displacement of RNA molecules and reverses it to restore their original spatial distribution while preserving molecular counts. Evaluations on simulated and real data demonstrate that SpaDiff enhances spatial specificity of gene expression, improves data integrity, and supports more accurate downstream analyses such as clustering and spatial domain identification.

bioinformatics↗

bioRxiv2025

Development of a High-Throughput Screening Method for Anti-NNV Drugs

Nervous necrosis virus (NNV) is an infectious pathogen, characterized by rapid infection and high mortality, commonly encountered in the aquaculture industry. Therefore, there is an urgent need to develop effective antiviral drugs against NNV, which is crucial for targeted treatment strategies, pathogen control, and loss reduction. Screening of effective antiviral active compounds, namely drug precursors, is key to developing highly efficient drugs against NNV. Reverse transcription-quantitative real-time PCR (RT-qPCR) is widely applied to screen and evaluate various substances to prevent and control virus infection. However, RT-qPCR is a cumbersome procedure and not suitable for high-throughput rapid screening. The aptamer TNA1c, which specifically binds to NNV-infected cells, was used to construct a target-driven activatable aptamer probe (TAA). Then, the TAA probe was applied to establish a high-throughput screening (TAA-HTS) method for efficient evaluation of substances against NNV infection. TAA-HTS technology achieved rapid, sensitive, and specific screening of bioactive substances with significant anti-NNV effects. As compared to commonly used analytical methods, such as RT-qPCR, TAA-HTS has the advantages of easy operation and high sensitivity and specificity. The findings of this study provide data support and a theoretical basis for the development of effective antivirus preparations. IMPORTANCEIn this study, a target-driven activatable aptamer probe was employed to establish a high-throughput screening method for anti-NNV drugs. This method enables rapid, sensitive, and specific screening of anti-NNV compounds while reducing background interference. Although RT-qPCR is currently the most widely used technique for antiviral drug screening and validation, the screening approach developed in this study significantly shortens screening time while achieving results consistent with those of RT-qPCR. This demonstrates that the screening method presented here has the potential to become a universal strategy for future drug screening against viral diseases. Moreover, this study not only provides a novel technical solution for NNV control but also demonstrates the potential of aptamer-based probes in high-throughput antiviral screening, serving as a model for combating viral pathogens in aquaculture and other fields.

pathology↗

bioRxiv2025

A mouse model for cerebral/cortical visual impairment (CVI) impairs vision and disrupts the spatial frequency tuning of neurons in visual cortex

Cerebral/cortical visual impairment (CVI) is a visual disorder associated with perinatal hypoxic injury. The pathophysiology of CVI is poorly understood in part because of the lack of an animal model. Here we developed a murine model of CVI from existing rodent early postnatal hypoxia models for periventricular leukomalacia. Exposure to hypoxia during the equivalent to the human third trimester did not perturb motor function but caused severe impairments in binocular depth perception and visual acuity. Impaired vision was associated with normal retinal function, but reduced size of the visual thalamus, and aberrant tuning for spatial frequency by populations of excitatory neurons in primary visual cortex. This murine model of CVI provides a framework for triangulating circuit deficits with specific visual impairments and testing potential therapeutic interventions.

neuroscience↗

bioRxiv2025

Metabolomics-Guided Machine Learning Reveals Diagnostic and Mechanistic Biomarkers in CHB with MASLD

BackgroundMetabolic dysfunction-associated steatotic liver disease (MASLD) often coexists with chronic hepatitis B (CHB), yet early diagnosis remains challenging, particularly in non-obese patients or those with subclinical features. This study aimed to identify metabolic signatures of CHB-related MASLD and construct a predictive model using untargeted metabolomics integrated with machine learning. MethodsSerum metabolomics was performed on 160 subjects (80 CHB + MASLD and 80 healthy controls). Differential metabolites were identified and analyzed using KEGG enrichment and 4 machine learning algorithms (Random Forest, XGBoost, SVM, and LASSO). Metabolite-clinical correlations and diagnostic model performance were evaluated. ResultsA total of 924 differential metabolites were identified, with significant enrichment in pathways related to the TCA cycle, sphingolipid metabolism, and amino acid turnover. Machine learning prioritized six robust and biologically relevant metabolites: L-aspartic acid, succinic acid, caproic acid, sebacic acid, monomenthyl succinate, and glycolaldehyde, which consistently distinguished CHB + MASLD patients from controls (AUC > 0.75). These metabolites reflect key disruptions in mitochondrial function, lipid oxidation, and redox homeostasis. Integrated models combining metabolomics with clinical indices achieved perfect classification (AUC = 1.000). ConclusionCHB-associated MASLD is driven by systemic metabolic remodeling centered on mitochondrial overload, oxidative stress, and impaired amino acid metabolism. The identified metabolites provide mechanistic insights and hold promise for non-invasive MASLD screening in CHB patients. This study underscores the potential of multi-algorithmic metabolomics in advancing early diagnosis and personalized management of complex liver comorbidities.

bioinformatics↗

bioRxiv2025

Rethinking Kalman Filters for Motor Brain-Machine Interface: The Fundamental Limitations and A Perspective Shift

The Kalman filter has been introduced to the motor brain-machine interface (BMI) field for over 20 years and remains one of the most widely used models due to its simple and intuitive nature. However, this paper demonstrates that the application of Kalman filters in the BMI field violates the models own assumptions and numerous neuroscience principles, resulting in six critical limitations: (1) the observation matrix mapping from kinematics to neural activity causes individual neurons to be modeled independently, ignoring neural population activity; aligned kinematic and neural activity sequences fail to capture preparatory information; (3) the presence of behaviorally irrelevant neural activity causes observation weights to be excessively underestimated; (4) noise in the observation matrix renders posterior covariance estimates biased; (5) Kalman gain computation in observation space leads to repeated noise accumulation and (6) computational inefficiency. All these problems can be resolved through a simple perspective shift: separating the decoder from the Kalman filter and treating its predicted kinematics as the Kalman filters observation rather than neural activity. Experiments conducted on CRCNS-recorded monkey dorsal premotor area and primary motor signals performing computer cursor control tasks validate the existence of the six limitations and their violations of neuroscience principles, while demonstrating the superiority of the improved Kalman filter across all evaluated metrics.

neuroscience↗

bioRxiv2025

Automodification of N-terminal serine residues in PARP2 impacts PARP2 release from DNA damage sites

ADP-ribosyltransferases PARP1 and PARP2 are involved in DNA repair mechanisms and play a major role in detecting DNA damage. PARP1/2 enzymes transfer the ADP-ribosyl moiety from NAD+ to DNA damage proteins, histones and to itself, activating the DNA repair cascade. Recent studies have shown that Histone PARylation Factor (HPF1) forms a joint active site with the catalytic domain of PARP1/2 and alters their target modification site from glutamate/aspartate towards serine. In this work, we have identified key serine residues within the N-terminus of full-length PARP2 that are the main targets of PARP1/2 automodification. We demonstrated this using site-directed mutagenesis, gel-based PARylation assays of automodification reactions, and by measuring the release of PARP2 from the DNA damage site using a fluorescence polarization assay. We show that in the presence of HPF1, PARP2 serine 8 and serine 73 are predominantly ADP-ribosylated and serine 8, which is present in both human PARP2 isoforms, is the major site for PARP2 automodification. Our results provide insight into the mechanistic role of the N-terminus of PARP2 in the PARylation-dependent release of PARP2 from DNA damage sites.

biochemistry↗