Biology subjects

Luo, X.

Publications and source records attributed to Luo, X..

At least 37 records · Page 2Linked to original sources

bioRxiv2026

Efficient and accurate near telomere-to-telomere haplotype reconstruction of diploid genomes

Telomere-to-telomere (T2T) and haplotype-resolved assembly are crucial for understanding eukaryotic genomes. For diploid species, this resolution is critical to uncover allelic variations, inheritance patterns, and functional genomic traits. Current scaffolding methods typically employ either sequence-based or graph-based strategies. Sequence-based approaches rely on proximity signals to yield high contiguity, but underutilize assembly graph information, resulting in more structural errors and chromosomal misassignments. Graph-based methods leverage graph topology for higher accuracy but frequently struggle to achieve chromosome-scale contiguity. However, neither strategy alone can overcome its inherent limitations to simultaneously achieve high contiguity and accuracy. To address these challenges, we introduce HapFold, the first hybrid scaffolding framework that synergistically leverages the complementary strengths of both graph-based and sequence-based approaches. By integrating the topological accuracy of assembly graphs with the proximity-guided contiguity of sequence models, HapFold achieves highly accurate, chromosome-scale or near-T2T haplotype reconstructions for diploid genomes. Compared to existing methods, HapFold achieves superior assembly quality while accelerating computation by an order of magnitude. Furthermore, in the haplotype reconstruction of diploid genomes using standard Oxford Nanopore Technologies simplex reads, HapFold enables the reconstruction of a greater number of near-T2T assemblies. Our approach provides a robust and scalable solution for the high-fidelity reconstruction of haplotype-resolved diploid genomes.

bioinformatics↗

bioRxiv2026

Patient-Calibrated Dynamical Modeling and Embedded Trend-Zone Predictive Control for Closed-Loop Deep Brain Stimulation in Parkinson's Disease

Closed-loop deep brain stimulation (cDBS) for Parkinsons disease requires control strategies that tolerate noisy sensing, patient-specific stimulation responses, medication-related fluctuations, and embedded hardware constraints. We developed a patient-calibrated minute-scale dynamical model of subthalamic beta activity and an embedded explicit trend-zone predictive controller, eTZPC. The model combined a basal-ganglia mechanistic prior with stimulation-amplitude and medication-cycle recordings from five patients, and incorporated individualized stimulation-{beta}STN maps, fast- and slow-timescale stimulation responses, levodopa-related modulation, background drift, and observation noise. eTZPC was designed to maintain {beta}STN activity within a patient-specific target zone under stimulation-amplitude, step-size, and quantization constraints. Compared with dual-threshold (DT) and proportional-integral-derivative (PID) controllers across four disturbance scenarios, eTZPC achieved target-zone regulation close to PID while reducing stimulation-switching burden toward the low-switching profile of DT. Ablation analyses identified distinct contributions of smoothing, trend prediction, patient-specific action modeling, and embedded explicit implementation. Parameter-mismatch tests showed that eTZPC was relatively robust to dynamic and disturbance-parameter deviations, but remained sensitive to errors in the steady-state stimulation-{beta}STN map. Patient-in-the-loop recordings in five patients further confirmed execution consistency and compliance with stimulation-boundary and step-size constraints. These findings support patient-calibrated dynamical modeling combined with low-complexity explicit control as a feasible framework for further embedded cDBS evaluation.

neuroscience↗

bioRxiv2026

AlaRS-mediated lactylation shields PARP1 from ubiquitination to promote HNSCC progression

Abstract textMetabolic reprogramming in solid tumors causes massive lactate accumulation, yet how this drives oncogenesis remains incompletely understood. Here, we identify alanyl-tRNA synthetase (AlaRS) as a cellular lactyltransferase that promotes head and neck squamous cell carcinoma (HNSCC) progression. AlaRS directly catalyzes site-specific lactylation of the DNA repair protein PARP1 at lysines K249 and K667. Crucially, we uncover a competitive post-translational crosstalk wherein this lactylation directly antagonizes PARP1 ubiquitination. This "lactyl-shield" prevents proteasomal degradation, hyper-stabilizing the PARP1-Mortalin complex and sustaining tumor proliferation via p53/p21 dysregulation. To therapeutically exploit this mechanism, we identified chelerythrine (CHE) as a potent, selective inhibitor that directly binds the AlaRS catalytic center. CHE abrogates AlaRS lactyltransferase activity, destabilizes PARP1, and robustly suppresses HNSCC xenograft growth in vivo. These findings establish a novel metabolic-post-translational axis linking lactate accumulation to oncoprotein stabilization, providing a blueprint for targeting tRNA synthetase moonlighting functions in cancer.

molecular biology↗

bioRxiv2026

H2O: A Foundation Model Bridging Histopathology to Spatial Multi-Omics Profiling

Spatial omics technologies have revolutionized the molecular profiling of tissues but remain constrained by high costs and limited scalability. While hematoxylin and eosin (H&E) staining is ubiquitous, it lacks molecular specificity. Here, we present H2O (Histopathology to Omics), a generalist AI framework that bridges the modality gap between histopathology and spatial multi-omics, enabling the direct inference of spatial transcriptomics (ST) and proteomics (SP) landscapes from routine H&E images. H2O integrates Vision Transformers (ViT) with Large Language Models (LLM) via contrastive learning to align histological morphology with semantic molecular knowledge. This cross-modal approach allows the model to incorporate spatial expression profiles into histological pattern recognition, effectively decoding the molecular heterogeneity underlying tissue morphology. Trained on a pan-tissue dataset of 1.3 million paired H&E-spatial patches across 25 organs and cancer types, H2O predicts spatial omics expression from histology with high concordance to sequenced measurements and consistently outperforms state-of-the-art models across three cancer benchmarks. Notably, H2O recovers the MIF-CD74/CD44 signaling axis directly from H&E images, highlighting its capacity to infer biologically meaningful cell-cell communication without molecular profiling. Applying on three additional public cohorts covering fetal and paediatric thymus tissues, human metastatic lymph node, and breast cancer, encompassing human development, 3D spatial frameworks, and integrative multi-omics, H2O yields biologically concordant insights, demonstrating superior accuracy, robustness, and generalizability across real-world applications in diverse scenarios. H2O converts routine histopathology into a portal for spatially resolved multi-omics profiling by computationally generating transcriptomic and proteomic landscapes, thereby enhancing tissue phenotyping and enabling scalable, integrative tissue-atlas construction.

bioinformatics↗

bioRxiv2026

Abnormal ventricular wall patterning precedes and drives MYBPC3 hypertrophic cardiomyopathy

BACKGROUNDExcessive trabeculations and myocardial crypts are recurrent features across cardiomyopathies, yet their developmental origins and clinical significance remain poorly defined. To reveal the link between cardiac morphogenesis and disease, we generated humanized mouse models carrying patient-derived MYBPC3 frameshift mutations associated with overlapping hypertrophic cardiomyopathy (HCM) and left ventricular non-compaction (LVNC). METHODSWe applied CRISPR-Cas9 to introduce distinct MYBPC3 frameshift alleles into the mouse genome and performed comprehensive phenotypic and transcriptomic profiling from fetal life through adulthood. RESULTSAdult homozygous Mybpc3 frameshift mutant mice like humans displayed hallmark HCM; however, without LVNC. Fetal and neonatal mutant hearts exhibited markedly enlarged ventricular trabeculae and crypts that progressed postnatally into the observed adult hypertrophy. Transcriptomic analysis revealed stage-specific dysregulation of oxidative metabolism, nonsense-mediated decay (NMD), and cell cycle pathways, peaking at postnatal days 1 and 7, indicating that these stages represent critical time points in disease onset. The persistent NMD signature, also observed in phenotype-negative heterozygotes, suggests a compensatory stress response. Enlarged trabeculae exhibited 2-fold increased trabecular cardiomyocyte proliferation, reversing the normal compact-trabecular proliferative gradient and leading to impaired ventricular compaction in neonates. Hey2CreERT2 lineage tracing demonstrated invasion of Hey2+ compact cardiomyocytes into the trabeculae and ectopic trabecular expression of the Prdm16 transcription factor, indicating defective ventricular wall patterning and maturation. Postnatally, Hey2+-derived cardiomyocytes became restricted to the outer/compact myocardium in mutants, while the inner/trabecular myocardium underwent accelerated hypertrophy concurrent with Prdm16 downregulation. Mice with a Mybpc3 missense variant also exhibited Hey2+ myocardial lineage expansion into trabeculae but no increased proliferation, implicating additional mechanisms beyond Hey2 regulation. Postnatal Prdm16 restoration, via transgenic expression in Mybpc3-null mice effectively attenuated hypertrophy, establishing a causal link between Mybpc3 loss, Prdm16 decline, and pathological remodeling. CONCLUSIONSMybpc3 governs ventricular wall maturation by regulating cardiomyocyte proliferation, patterning, and maturation, partly via Prdm16. Disruption of these developmental programs precedes and drives adult HCM, highlighting a developmental role for sarcomeric proteins, and revealing postnatal Prdm16 modulation as an antihypertrophic therapeutic strategy.

developmental biology↗

bioRxiv2026

Benchmarking and Experimental Validation of Machine Learning Strategies for Enzyme Engineering

Enzyme-directed evolution increasingly relies on computational tools to prioritize mutations, yet their practical value is difficult to assess because kinetic data are often aggregated across heterogeneous assay conditions, inflating apparent generalization. Here we introduce EnzyArena, a curated benchmark in which kinetic parameters (kcat, Km and kcat/Km) are grouped by experimental campaign and measured under matched conditions, thereby minimizing confounders such as pH, temperature, reaction, and batch. Using EnzyArena, we systematically evaluate 20 models spanning four widely used strategies for guiding enzyme evolution: protein-ligand binding affinity prediction, protein stability prediction, zero-shot fitness prediction and enzyme kinetic parameter prediction. Across subsets derived from public databases and 25 independent mutagenesis datasets, most models exhibit weak, fragile or inconsistent correlation with catalytic activity. Kinetic-parameter predictors perform strongly on database-derived subsets but lose their advantage on independent datasets, whereas zero-shot predictors show more consistent generalization. A simple consensus of multiple zero-shot models further improves the precision of identifying beneficial mutants. We prospectively validated these findings in a wet-lab campaign (150 mutants) comparing random mutants, UniKP-prioritized mutants and ESM-1v-prioritized mutants (representing supervised kinetic-parameter prediction and zero-shot fitness prediction, respectively), where ESM-1v achieved the highest utility and UniKP underperformed the random baseline. Together, this study establishes realistic baselines for computational mutant prioritization and highlights consensus zero-shot strategies as a practical starting point for enzyme engineering. Zero-shot fitness predictors are the most competitive and generalizable class, but still achieve only modest correlations on independent datasets. Notably, a simple consensus of multiple zero-shot models substantially improves the precision of identifying superior mutants and currently offers the most practical computational benefit for enzyme engineering. Our work establishes EnzyArena as a fair benchmark for enzyme activity prediction, defines realistic performance baselines for existing tools, and highlights consensus zero-shot strategies as the most useful computational tool in practice for enzyme-directed evolution.

bioengineering↗

bioRxiv2026

Synthetic Immunological Niche Reveals Early Immune Dysregulationand Stratifies Therapeutic Response in Type 1 Diabetes

In Type 1 Diabetes (T1D), disease onset and response to immunotherapy vary widely among individuals, reflecting heterogeneous stage-specific immune dysregulation that remains undefined. To investigate this heterogeneity, we engineered a microporous polycaprolactone scaffold that forms a synthetic immunological niche (IN) upon subcutaneous implantation, enabling in vivo capture of systemic immune dysregulation. In non-obese diabetic (NOD) mice, longitudinal transcriptomic profiling of IN-infiltrating cells identified early-stage genes relatively enriched for myeloid cells, followed by progressive increases in T cell-associated dysregulation at later stages that distinguished T1D progressors from non-progressors. We derived an early-stage IN-based T1D gene signature capturing immune alterations. The signature stratified NOD progressors from non-progressors as early as 6 weeks of age and was conserved across human T1D datasets, distinguishing T1D from non-diabetic individuals in spleen and pancreatic lymph node samples, but not peripheral blood. Signature-based stratification further revealed enrichment of macrophage-associated TNF- pathways in NOD progressors, validated in human T1D islets. Given heterogeneous response to anti-TNF- therapy, IN profiling identified resistance-associated mechanisms and enabled derivation of a pathway score that prospectively distinguished treatment-sensitive from resistant mice prior to therapy, establishing the IN as a minimally invasive platform for detecting stage-wise immune dysregulation and stratifying immunotherapy response in T1D.

bioengineering↗

bioRxiv2026

Molecular Glue-Induced Homodimerization Drives Targeted CRBN Autodegradation

Molecular glue degraders (MGDs) offer a sophisticated, proximity-based approach to protein modulation. In this study, we introduce LJY-3-60, a novel proximity-inducing agent that unexpectedly triggers the potent and selective autodegradation of CRBN. Evidence from CRISPR-Cas9 screening and IP-MS reveals that this degradation process is strictly governed by the intrinsic CRL4CRBN machinery, independent of any extrinsic E3 recruitment. Through a combination of cellular and biophysical characterizations, we demonstrate that LJY-3-60 acts as a molecular bridge to template CRBN homodimerization. This mechanism is unequivocally elucidated by the atomic-resolution co-crystal structure of the CRBNMidi-LJY-3-60 complex. The structure explicitly delineates the homodimerization interface, revealing how the ligand reorganizes the protein surface to stabilize a non-canonical architecture that drives trans-autoubiquitination and subsequent proteasomal degradation. Furthermore, LJY-3-60 serves as a highly effective, controllable off-switch to mitigate PROTAC-induced toxicity. Ultimately, this work delivers a robust chemical tool for modulating CRBN stability. By demonstrating how a small molecule can functionally mimic an endogenous E3 substrates degron to catalyse targeted autodegradation, this study establishes a rational structural framework for designing the next generation of self-destructive modulators in targeted protein degradation (TPD) therapeutics.

biochemistry↗

bioRxiv2026

Telomere-to-telomere assembly and haplotype analysis of tetraploid Dendrobium officinale illuminate Orchidaceae polyploid evolution and mycorrhizal symbiosis genes

Dendrobium officinale is a typical epiphytic orchid. We report the telomere-to-telomere (T2T) genome assembly for D. officinale, representing the first T2T reference genome within the Orchidaceae family. The assembly is anchored to 19 chromosomes and contains 38 complete telomeres and 15 characterized centromeres. We further generated haplotype-resolved assemblies of the autotetraploid genome, identifying 12,761 sets of tetra-allelic genes. Based on synonymous substitution analysis, we inferred that the autotetraploidization event occurred approximately 0.86 million years ago. A systematic analysis of the SWEET gene family across the genus Dendrobium revealed that the gene family size is shaped primarily by epiphytic types and environmental factors. In D. officinale from Langshan, eight SWEET genes were specifically expressed in roots, suggesting they may play specialized roles in the root mycorrhizal system, potentially contributing to the D. officinales ability to recruit and maintain fungal partners. Together, these resources provide valuable foundations for studies of orchid evolution, functional genomics, and molecular breeding.

bioinformatics↗

bioRxiv2026

DNA Adenine Methylation Clock in Brain Aging and Alzheimer's Disease Progression

N6-methyldeoxyadenosine (N6medA) is a recently identified endogenous DNA modification widely found in bacteria, plants, and eukaryotes. In mammals, N6medA has been implicated in brain function, immunity, and response to environmental stress, but its relevance to gene regulation and mammalian aging remains controversial due to its extremely low abundance (< 1 per 10 million adenines) and an uncertainty regarding its genomic origin. We have developed and validated an ultrasensitive isotope dilution nano liquid chromatography-nanospray ionization Orbitrap mass spectrometry methodology to quantify N6medA in genomic DNA. Applying this approach to human prefrontal cortex tissues, we found that genomic N6medA levels increase linearly with chronological age (Pearson correlation coefficient, 0.95). Individuals with mild cognitive impairment (MCI) and Alzheimers disease (AD) exhibited a trend toward elevated cortical N6medA levels relative to age-matched controls. Genome-wide profiling of N6medA in human prefrontal cortex was conducted using two independent methods: NAME-Seq and MeDIP-Seq, which revealed age associated adenine methylation changes reminiscent of established epigenetic aging signatures such as the 5-methylcytosine clocks. N6medA mapping experiments identified a subset of genomic loci that were altered in MCI and AD. Pathway analysis of cross-validated adenine methylation sites revealed an enrichment of genes involved in neuronal function and age-related neurological processes, including glutamatergic synapse, axon guidance, and long-term depression. Finally, mass-spectrometry-based photoaffinity proteomics with synthetic DNA representing a region of the APP gene identified N6medA reader proteins with known roles in DNA repair, replication and transcription. Together, these findings identify N6medA as an age-associated DNA modification in the human brain and suggest that its accumulation and recognition by specific protein readers may contribute to molecular processes underlying brain aging and age-related neurodegeneration.

biochemistry↗

bioRxiv2026

Macrophage SLC9A1 Links Endocytic Trafficking to Innate Immune Activation in Myocardial Injury

Excessive innate immune activation drives adverse remodeling after myocardial infarction (MI), yet the upstream mechanisms by which macrophages sense ischemic danger signals remain poorly defined. Here we tested whether macropinocytosis functions as a mediator of post-ischemic inflammation and whether the Na/H exchanger SLC9A1 links membrane ion transport to innate immune activation in the injured heart. Macropinocytosis was robustly activated in infarct-associated macrophages, which are the predominant cell type with the macropinocytotic activity in the injured heart. Pharmacologic inhibition of macropinocytosis with 5-(N-ethyl-N-isopropyl)amiloride (EIPA) improved cardiac function and attenuated post-MI remodeling. EIPA also attenuated cardiac inflammatory responses induced by systemic lipopolysaccharide and Poly(I:C). To define macrophage-intrinsic mechanisms, we generated monocyte- and monocyte-derived macrophage-specific Slc9a1 knockout mice. Genetic deletion of Slc9a1 recapitulated the cardioprotective effects of EIPA and markedly suppressed interferon-stimulated gene programs in infarct-associated macrophages, as revealed by single-cell RNA sequencing. Mechanistically, SLC9A1 promoted endocytic uptake of Poly(I:C) acid and enhanced endosome-dependent inflammatory signaling. Together, these findings identify macrophage macropinocytosis as a regulator of innate immune activation after MI and reveal SLC9A1 as a previously unrecognized link between membrane ion transport and inflammatory signaling in the injured heart. Targeting SLC9A1-dependent membrane trafficking pathways may therefore represent a strategy to limit maladaptive inflammation in ischemic heart disease.

immunology↗

bioRxiv2026

Mitotic bookmarking by Prox1 preserves mammalian neuronal lineage identity memory via promoting timely H3K27me3 restoration

Mitosis poses a daunting challenge to transgenerational inheritance of cell identity memory. How neuronal lineage identity is faithfully transmitted across mitosis remains largely unexplored. Here we report that, in mouse hippocampus, the transcription factor Prox1 acts as a mitotic bookmark for safeguarding neuronal lineage identity across cell divisions. Prox1 exhibits mitotic retention in dentate gyrus (DG) neural stem cells and defines DG lineage identity by suppressing the alternative cornu ammonis (CA) identity. Mitotic bookmarking by Prox1 at key bivalent CA identity genes promotes timely and precise restoration of Polycomb repressive complex 2 (PRC2)-mediated H3K27me3 deposition to avoid ectopic expression of CA identity genes and lineage identity crisis. Remarkably, specific mitotic retention-deficient Prox1 conditional knock-in mice produce severe DG developmental defects. Thus, mitotic bookmarking imprints neuronal lineage identity in mouse hippocampus, which is likely to represent a fundamental principle underlying the preservation of lineage identity memory in mammalian brain development.

developmental biology↗

bioRxiv2026

UBE3C retrofits the proteasome to enforce degradation of ultra-stable folds

Neurodegeneration and proteinopathies arise from mutated or misfolded proteins that form ultra-stable, proteotoxic aggregates that evade proteasomal degradation despite ubiquitylation1-8. These conditions are exacerbated by the deubiquitylase USP14 that can prematurely rescue substrates9-11. To counter these obstacles to effective proteolysis, eukaryotes evolved the highly conserved HECT-type E3 ligase UBE3C--dysregulated in neurodegeneration12,13 and overexpressed in cancers14-17--to reprogram the 26S proteasome toward heightened activity via unknown mechanisms18-25. Here we visualized functional dynamics of the human UBE3C-retrofitted proteasome enforcing degradation of a re-engineered superfolder GFP that normally escapes proteolysis, using time-resolved cryo-electron microscopy. A continuum of non-equilibrium conformations, comprising fourteen proteasome conformers orthogonally combined with four UBE3C states or three USP14 states, reveals a cryptic UBE3C-receptor site in the proteasomal lid and captures key intermediates of ubiquitin-chain elongation and branching at four linkage-specific ubiquitin-binding sites. Remarkably, UBE3C creates an extreme shortcut for ubiquitin shuttling that bypasses USP14, promotes USP14 recycling and allosterically strengthens substrate-unfolding forces of the AAA-ATPase unfoldase to ultimately surmount the unfolding energy barrier of ultra-stable proteins. These findings define the complete functional cycle of the UBE3C-reprogrammed proteasome, illuminating how UBE3C simultaneously controls USP14 and the proteasome to enforce clearance of ultra-stable folds, and establish crucial mechanistic foundations for proteolysis-targeting therapeutic discovery.

biochemistry↗

bioRxiv2026

Tirzepatide improves pancreatic β-cell function in mice and patients with type 2 diabetes.

The dual incretin receptor agonist tirzepatide improves {beta}-cell function in T2D patients, but the underlying mechanism remains unclear. This study aimed to elucidate the molecular pathway through which tirzepatide restores {beta}-cell functional improvement. High-fat diet (HFD)-fed C57BL/6J mice were treated with vehicle, a GIP analogue, semaglutide or tirzepatide. Tirzepatide significantly reduced body weight and improved glucose tolerance in HFD-fed mice without altering {beta}-cell mass, proliferation, or apoptosis. Instead, tirzepatide reversed {beta}-cell dedifferentiation, as indicated by reduced ALDH1A3 expression and restored levels of the identity transcription factors PDX1 and MAFA. Single-cell RNA sequencing (scRNA seq) and in vitro studies revealed that tirzepatide up-regulated FOXO1, reactivating the FOXO1-PDX1/MAFA axis. In T2D patients, tirzepatide improved glycemic control, reduced insulin demand, increased HOMA-{beta}, and decreased HOMA-IR. Improvement in HOMA-{beta} correlated positively with baseline insulin resistance. Hence, our study suggested that tirzepatide restores {beta}-cell function in T2D by reprogramming stressed {beta} cells and re-establishing {beta}-cell identity through FOXO1-dependent transcriptional reactivation. These findings provide a mechanistic basis for the superior efficacy of dual incretin receptor agonism in T2D management. ARTICLE HIGHLIGHTSO_LITirzepatide restores {beta} cell identity and function without altering {beta} cell mass in HFD induced diabetic mice. C_LIO_LITirzepatide reverses {beta}-cell dedifferentiation and restores key {beta}-cell transcription factors (PDX1, MAFA) through reactivation of the AKT-FOXO1 signaling pathway. C_LIO_LITirzepatide increases HOMA-{beta} and decreases HOMA-IR in T2D patients, and improvements in HOMA-{beta} positively correlate with baseline insulin resistance. C_LIO_LIThese results demonstrate that tirzepatides therapeutic benefits are not only metabolic but also involve direct restoration of {beta}-cell identity and function. This highlights {beta}-cell reprogramming as a novel therapeutic avenue, thus supporting the broader clinical adoption of dual incretin receptor agonists. C_LI

physiology↗

bioRxiv2026

Differential higher-order superassembly of HECT-type UBE3 ligases controlled by calcium signals

The yeast E4 ligase Hul5 transiently associates with the proteasome during heat shock and proteotoxic stress to maintain cytosolic protein quality control. Two of its human orthologs, UBE3B and UBE3C, have specialized to safeguard mitochondria and inherit Hul5 function, respectively. We determined cryo-electron microscopy structures of full-length human UBE3B and UBE3C in complex with calmodulin at 2.9-3.5 [A] in calcium-free and calcium-saturated conditions. Calmodulin acts as an inter-protomer molecular glue clamping UBE3B into ring-shaped anti-parallel homodimers or asymmetric trimers, and remodels UBE3C into a conformation competent for proteasome association. Calcium binding to calmodulin promotes disassembly of these higher-order complexes, toggling UBE3B/C from E4 to E3 activity. These findings reveal how calmodulin regulates higher-order architectures of E3/E4 ligases to rewire the ubiquitin-proteasome system via calcium signaling.

biochemistry↗

bioRxiv2026

PlotGDP: an AI Agent for Bioinformatics Plotting

High-quality bioinformatics plotting is important for biology research, especially when preparing for publications. However, the long learning curve and complex coding environment configuration often appear as inevitable costs towards the creation of publication-ready plots. Here, we present PlotGDP (https://plotgdp.biogdp.com/), an AI agent-based web server for bioinformatics plotting. Built on large language models (LLMs), the intelligent plotting agent is designed to accommodate various types of bioinformatics plots, while offering easy usage with simple natural language commands from users. No coding experience or environment deployment is required, since all the user-uploaded data is processed by LLM-generated codes on our remote high-performance server. Additionally, all plotting sessions are based on curated template scripts to minimize the risk of hallucinations from the LLM. Aided by PlotGDP, we hope to contribute to the global biology research community by constructing an online platform for fast and high-quality bioinformatics visualization. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=78 SRC="FIGDIR/small/702995v3_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@73c77dorg.highwire.dtl.DTLVardef@12ddfc6org.highwire.dtl.DTLVardef@be5963org.highwire.dtl.DTLVardef@de90b6_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphical abstract.C_FLOATNO PlotGDP: an AI Agent for Bioinformatics Plotting C_FIG

bioinformatics↗

bioRxiv2026

HistoSweep enables cellular-resolution tissue quality control for gigapixel images in digital pathology and spatial omics

High-resolution histology images are indispensable for pathology and increasingly serve as the structural backbone for spatial omics. Yet whole-slide images (WSIs) frequently contain artifacts, acellular voids, and background regions that, when included in computational workflows, introduce noise, degrade model accuracy, and compromise biological interpretation. Existing tools provide only coarse foreground-background separation, leaving a gap in fine-grained quality control (QC). Here we present HistoSweep, a scalable framework that generates morphology-aware tissue masks at cellular resolution. By integrating density filtering, texture descriptors, and adaptive thresholding, HistoSweep systematically removes non-informative tissue regions while preserving biologically meaningful microstructures. It processes billion-pixel WSIs in minutes on standard CPUs, requiring no GPU acceleration, and is deployable across research and clinical settings. Across 25 WSIs spanning distinct tissues, disease states, and spatial omics platforms, HistoSweep consistently outperformed existing methods. It enhanced visualization and segmentation, improved virtual cell type predictions, and safeguarded spatial transcriptomics integrity by detecting transcript leakage and transcript-histology misalignment. By enabling fine-grained, scalable QC, HistoSweep provides a foundational preprocessing step for reliable and reproducible digital pathology and spatial omics analyses.

bioinformatics↗

bioRxiv2026

Dual-channel graph learning reveals similarity and complementarity in protein-protein interaction networks

Protein-protein interactions (PPIs) are governed by two fundamental interfacial mechanisms: similarity-driven, often involving symmetric structural motifs, and complementarity-driven, arising from geometric and physicochemical matching between binding surfaces. Despite their biological significance, computational models have largely overlooked the coexistence and interplay of these twofold interaction modes. Here, we introduce DMG-PPI, a dual-channel graph neural network framework that jointly models similarity and complementarity in PPI networks, extending prior heterophilous GNN concepts to explicitly disentangle these dual interaction modes. The model consists of two parallel message-passing pathways: Alignment Message Passing (AMP), which aggregates signals from similar neighbors to capture symmetric interfaces, and Divergence Message Passing (DMP), which contrasts node features to extract complementary binding patterns. The signals captured by AMP and DMP are integrated via an adaptive fusion strategy within each block, and the outputs of blocks are aggregated using the MixHop framework to encode higher-order interaction patterns. DMG-PPI substantially outperforms state-of-the-art methods on classical benchmark datasets, achieving a 7.19% improvement in Micro-F1 over the second-best method. Additionally, the dual-channel framework provides interpretable insights into key binding residues by identifying interfacial mechanisms. Overall, DMG-PPI serves as a powerful tool that reveals the mechanisms behind accurate PPI predictions and facilitates downstream biological analysis. Author summaryProteins interact through multiple interfacial principles rather than a single uniform mechanism, reflecting diverse modes of molecular recognition between binding partners. Accurately modeling this heterogeneity remains a central challenge in protein-protein interaction (PPI) prediction. Existing computational approaches often overlook such complexity and treat all interactions as arising from a dominant pattern. To address this limitation, we present DMG-PPI, a dual-channel graph neural network framework designed to account for distinct interaction principles within a unified model. By modeling both similarity-based and complementarity-based interaction signals, DMG-PPI enables a more comprehensive representation of interaction patterns in PPI networks. Evaluations on widely used benchmark datasets demonstrate that our approach achieves robust and consistent improvements over existing methods. Beyond prediction accuracy, DMG-PPI provides biologically meaningful interpretability by highlighting key residues and revealing distinct interfacial interaction mechanisms, offering valuable insights for downstream structural and functional studies.

bioinformatics↗