Biology subjects

Sun, H.

Publications and source records attributed to Sun, H..

At least 19 recordsLinked to original sources

bioRxiv2026

Valine Engages a Prefrontal mGluR5 Allosteric Site to Preserve Social Behavior

Branched-chain amino acid metabolism is linked to neurodevelopmental disorders, but whether a defined amino acid directly modulates cortical receptor signaling to control behavior is unknown. Here, we show that valine is uniquely required for sociability through medial prefrontal cortex (mPFC) circuitry. Partial dietary valine restriction, but not comparable leucine or isoleucine restriction, or mPFC-restricted acceleration of BCAA catabolism through Bckdk loss, disrupts social interaction, and local valine supplementation rescues behavior. LiP-SMap, surface plasmon resonance, cryo-EM, and mutagenesis identify metabotropic glutamate receptor 5 (mGluR5) as a direct valine target and reveal a valine-selective allosteric pocket distinct from the glutamate site. Valine is signaling-biased: it drives {beta}-arrestin-linked output yet has little standalone Gq efficacy, instead potentiating glutamate-evoked mGluR5 signaling to recruit endocannabinoid/CB1 feedback, gate inhibitory transmission, and preserve mPFC-dependent social behavior. Together, these findings identify valine as a diet-derived allosteric ligand that links amino acid availability to cortical circuit control of social behavior.

neuroscience↗

bioRxiv2026

Urea is a dynamic and functional component of mammalian breast milk

Urea is a constituent of human milk, yet its regulation and interactions with the infant gut microbiome remain unclear. Combining three human cohorts, a controlled mouse model, and complementary microbiome analyses, we show that milk urea increases across lactation, is conserved across mammals, and is not explained by circulating urea levels or maternal diet. Instead, we identify lactation-dependent remodeling of mammary urea-metabolism and transport pathways, including dynamic AQP3 localization in secretory epithelium, indicative of regulated transfer of circulating urea into breast milk. Alongside urea, maternal milk and stool provide complementary reservoirs of urea-metabolizing bacteria that help establish the infant "ureabiome", whose urea-utilization capacity is supported by metagenomics. This maternal-infant axis is context-dependent: infant formulas provide substantially higher urea levels than human milk, while functional assays show that urea supports both commensal and opportunistic bacteria in a concentration-dependent manner. Together, these findings define a maternal nitrogen axis through which lactation can shape the assembly and function of the infant gut microbiome.

microbiology↗

bioRxiv2026

Biologically grounded cell profiling across microscopy modalities

Microscopy-based cell profiling has broad applications in biological discovery, disease characterization, and phenotypic drug screening. Modern microscopy continues to push the limits of resolution, speed, depth and throughput, but better imaging does not automatically lead to better biomedical discovery and translation. A key bottleneck is feature representation: existing features are either handcrafted or learned as black-box embeddings and often lack explicit biological meaning. Here we propose biological grounding as a first principle for cell profiling and implement it through MorphAgent, an AI agent framework in which each quantitative feature originates from a biological hypothesis and is anchored to a cellular structure or process. By integrating biological knowledge, multimodal reasoning and automated validation, MorphAgent makes biologically grounded feature design systematic and scalable. We show biological grounding fundamentally changes the properties of cell profiling features. Across three microscopy modalities, MorphAgent produces compact, expressive and transferable features for drug screening, cell state characterization and disease morphology analysis. In wide-field Cell Painting, a common assay for phenotypic drug screening, it improves perturbation-retrieval mean average precision by 48% over CellProfiler and 20% over DeepProfiler while using substantially fewer dimensions. In confocal mitochondrial imaging, biologically grounded features transfer across independently acquired datasets and imaging resolutions, supporting accurate aging-state classification. In structured illumination microscopy (SIM) imaging of Tau-labeled samples, biologically grounded super resolution features reveal nanoscale morphologies inaccessible at conventional resolution, enhancing the discrimination of disease-associated mutations. Moreover, biologically grounded features provide a hierarchical organization of cellular morphology, enabling multilevel alignment with transcriptomics data and facilitating mechanistic understanding. Biologically grounded cell profiling thus brings AI reasoning closer to biological mechanisms, advancing biomedical discovery and translation.

bioinformatics↗

bioRxiv2026

Brain age prediction from structural and functional connectivity across the lifespan using connectome-based predictive modeling

Brain maturation and aging are characterized by widespread changes in structural and functional brain organization. The extent to which these modalities capture shared versus distinct signatures of lifespan development and aging remains unclear. In this study, we applied connectome-based predictive modeling (CPM) to connectomes derived from diffusion MRI and resting-state functional MRI in Human Connectome Project Development, Young Adult, Aging, and combined Lifespan cohorts to predict chronological age. Both structural and functional connectomes significantly predicted age across cohorts, demonstrating that age-related information is distributed throughout whole-brain connectivity patterns. Prediction performance varied across lifespan stages, with generally stronger performance in the Development, Aging, and Lifespan cohorts than in Young Adulthood. Structural and functional models produced significantly correlated predicted ages in several cohorts, indicating shared age-related information across modalities. Structural-functional convergence analyses revealed limited and variable correspondence between modality-specific predictive features, suggesting that structural and functional connectomes capture both shared and complementary aspects of brain maturation and aging. Network-level analyses demonstrated that age-predictive information was distributed across canonical brain networks, including cerebellar, frontoparietal, salience, somatomotor, and subcortical networks, with patterns varying across cohorts and modalities. Multimodal CPM improved prediction relative to unimodal models in several cohorts, supporting the complementary contribution of structural and functional connectivity to age prediction. Cross-cohort, cross-sex, and cross-modality analyses further characterized the generalizability and shared information of predictive models across lifespan stages, sexes, and imaging modalities. This work demonstrates the utility of CPM for characterizing distributed structural and functional connectivity patterns associated with chronological age across the lifespan and provides a foundation for future studies of individual variability in brain maturation and aging.

neuroscience↗

bioRxiv2026

Transcription factors read a second regulatory code in chromatin

Transcription factors (TFs) decode gene regulatory information written in DNA, yet how this vocabulary is interpreted within chromatin remains largely unexplored. Here, using an upgraded NCAP-SELEX platform, we systematically map the nucleosomal DNA recognition landscapes of 269 human TFs. We uncover a widespread, chromatin-dependent mode of sequence recognition: many TFs recognize motifs on nucleosomal DNA that are distinct from their canonical naked-DNA binding sites, revealing that nucleosome architecture encodes a second gene regulatory code in chromatin. Cryo-electron microscopy structures of TF-nucleosome complexes demonstrate that this second code is read through a combination of nucleosome-induced DNA deformation and direct protein-histone contacts. Functional analyses show that these chromatin-encoded motifs actively promote chromatin accessibility and drive cell-type-specific cis-regulatory activity in vivo. Strikingly, a single TF can deploy its canonical and nucleosome-derived motif repertoires to partition and govern entirely distinct physiological programs. Together, our findings establish that TFs interpret two complementary layers of genomic information: the primary DNA sequence and a second code written into the nucleosome architecture. This chromatin-encoded layer of regulatory information fundamentally expands our understanding of how TF specificity is achieved and how gene regulatory networks are wired in multicellular organisms.

systems biology↗

bioRxiv2026

Multiomics Characterization Identifies S1P-secreting USSHigh Skeletal Muscle Stem Cells as Essential Drivers of Niche Remodeling and Muscle Regeneration

Acutely induced cellular senescence is increasingly recognized as vital for tissue repair but the precise roles of senescent cells in skeletal muscle regeneration remain paradoxical. Here, using single-cell RNA sequencing and a Unified Senescence Score (USS) methodology, we identify a distinct USSHigh (USSHi) muscle stem cell (MuSC) subpopulation emerging in early phase of acute injury induced muscle regeneration. Isolation via the surface marker MHCII enables characterization of these cells as exhibiting canonical senescent-like features. Further characterization through integrated transcriptomics, proteomics, and secretomics reveals their highly pro-inflammatory secretory capacity. Moreover, lipidomic profiling uncovers that the USSHi MuSCs undergo profound sphingolipid metabolic remodeling to generate upregulated bioactive lipid Sphingosine-1-phosphate (S1P). In vivo specific ablation of this subpopulation using a Pax7-CreERT; Ccl2-LSL-DTA genetic strategy significantly impaired muscle regeneration. Furthermore, spatial transcriptomics and cell-cell communication analyses revealed that USSHi MuSCs act as critical niche modulators, directing crosstalk with multiple cell types. Specifically, S1P is essential for mediating direct USSHi interaction with endothelial cells and USSLo cells to promote endothelial angiogenesis and MuSC expansion in the regenerating muscle. Crucially, exogenous S1P rescues the regenerative deficits caused by USSHi MuSC ablation. Altogether our findings define a senescent-like MuSC population emerging in acute injury induced muscle regeneration and provide definitive evidence to demonstrate the positive role of these cells in enhancing muscle repair via their robust secretory phenotypes and inter-cellular communicative capacity.

cell biology↗

bioRxiv2026

A Cre-responsive reporter quail for flexible genetic lineage tracing

Genetic lineage tracing is a powerful approach for investigating cell fate and tissue development, but versatile Cre/loxP-based tools remain limited in avian models. Here, we generated and characterised a Cre-responsive transgenic quail line that enables permanent genetic labelling of embryonic cell populations. The reporter line ubiquitously expresses nuclear-localised mCherry, which is excised following Cre-mediated recombination, enabling expression of nuclear-localised EGFP. We demonstrate that this reporter can be coupled to multiple Cre delivery strategies, providing flexible spatial and temporal control of lineage labelling. Ubiquitous or cell-type-specific Cre expression induces recombination and stable EGFP labelling that is maintained during clonal expansion. Optogenetic activation of a split Cre system enables highly spatially and temporally restricted recombination. Furthermore, lipid nanoparticle-mediated delivery of Cre mRNA enables efficient recombination in distinct embryonic cell populations, including mesodermal, neuroepithelial, neural crest, endothelial and circulating cells. This transgenic reporter quail provides a flexible platform for permanent lineage labelling without the need to generate tissue-specific Cre transgenic lines and can be readily combined with diverse Cre delivery strategies to investigate cell fate, lineage and morphogenesis in vivo.

developmental biology↗

bioRxiv2026

Multigenerational machine learning-based genomic prediction for dermo resistance in eastern oyster Crassostrea virginica

Dermo disease caused by the protist Perkinsus marinus poses a major threat to Eastern oyster aquaculture. We previously conducted genomic selection for dermo resistance and found increased effectiveness over phenotypic selection. Here, we report improved genomic predictions using combined data from three successive generations. We evaluated nine different genomic selection models, including three supervised machine learning architectures: gradient boosting, logistic regression, and random forest. Combining data across multiple generations did not, by itself, substantially improve the accuracy of most models, but the increased sample size provided genotyping confidence of loci with rare alleles. Correlation accuracy of all models significantly increased with the inclusion of low-frequency variants and strong-effect markers identified through a genome-wide association study. Gradient boosting machine learning models outperformed other genomic selection models across all training scenarios, suggesting enhanced capacity to learn generalizable genomic signals associated with dermo resistance. The best gradient boosting model achieved a peak correlation accuracy of 0.410, a substantial improvement over the previous peak accuracy of 0.274 from traditional models. Together, our results highlight the potential of machine learning for genomic selection and the significance of rare variants in determining dermo resistance.

genomics↗

bioRxiv2026

Improved two-stage GWAS identifies rare genetic variants associated with dermo susceptibility in the eastern oyster

Perkinsus marinus is a protist that causes the wasting disease dermo in the Eastern oyster, Crassostrea virginica. The disease causes high mortality of both wild and cultured oysters, impacting ecosystem health and aquaculture. Despite decades of research, the genetic mechanisms of dermo resistance remain poorly understood. Traditional selective breeding based on phenotypes has yielded improvements in dermo resistance, but progress has been slow, likely due to the trait's complex genetic architecture. Understanding the genetic mechanisms underlying dermo resistance may facilitate advanced breeding to accelerate improvement. This study conducted genome-wide association studies (GWAS) of dermo resistance, leveraging a multi-generational dataset comprising 2,423 oysters phenotyped for survival following a dermo challenge and genotyped using a high-density 66K SNP array. A two-stage GWAS with more inclusive quality control identified 48 dermo-resistance markers, including many associated with genes for innate immune response and energy metabolism. Most dermo-resistance markers showed rare minor alleles occurring at higher frequencies in oysters that died after challenge, suggesting that these rare alleles are deleterious and associated with dermo susceptibility. The strongest association was with a polymorphism in the mucin-5AC-like gene, explaining 8.1% of phenotypic variation and suggesting that mucus production plays a crucial role in dermo resistance. The inclusion of these markers has improved genomic prediction accuracy, and their identification provides new insights into genomic variation and the architecture of dermo resistance, informing future strategies for genetic improvement.

genomics↗

bioRxiv2026

The wild-type hERG cryo-EM structure fails to support conduction and evolve toward an inactivated-like selectivity filter conformation in molecular dynamics simulations

The atomic structure of the human Ether-a-go-go-Related Gene (hERG) K+ channel has recently been resolved by cryo-electron microscopy (cryo-EM) under both high- and low-K+ conditions, in order to obtain information on the mechanism of the K+-sensitive, very rapid C-type inactivation typical of this channel. Although the currently available high-K+ structures have been widely interpreted as representing the conductive, active state, whether they correspond to a dynamically stable conductive conformation remains unresolved. Here, we used extensive all-atom molecular dynamics (MD) simulations, with and without Electronic Continuum Correction (ECC), to investigate selectivity filter (SF) dynamics and ion permeation in wild-type (WT) hERG and the non-inactivating N629D mutant. Across all membrane potentials tested, WT hERG failed to support K+ permeation and instead spontaneously evolved towards a non-conductive SF conformation characterized by extracellular dilation, localized inner constriction, depletion of the outer ion-binding sites, and persistent trapping of K+ ions within the central binding sites of the filter, closely resembling the inactivated SF of Shaker channels recently resolved by cryo-EM. By contrast, N629D maintained a stable conductive SF architecture, analogous to the conductive filters of canonical K+ channels such as KcsA, while exhibiting robust voltage-dependent K+ permeation. ECC enhanced ion permeation in N629D mutant, but failed to support conduction in WT hERG, which remained structurally and functionally non-conductive. Together, these findings challenge the prevailing interpretation of the high-K+ WT cryo-EM structure as a stable conductive state and identify SF remodeling as the structural mechanism underlying hERG C-type inactivation.

biophysics↗

bioRxiv2026

Regulation of a Classical Allosteric Molecular Machine by an Intrinsically Disordered Domain: the C-termini of GroEL

The bacterial chaperonin GroEL is a canonical example of an ATP-dependent molecular machine that must couple ligand binding to productive conformational work. GroEL passes through a series of distinct structural shifts, driven by ATP binding and hydrolysis, which power a facilitated protein folding reaction. How the complex allostery of the GroEL oligomer creates a folding cycle that is both efficient and directional remains incompletely understood. Here, we combine variable-temperature native ion mass spectrometry with single-molecule FRET to examine how the intrinsically disordered, highly conserved GroEL C-terminal tails impact the allosteric behavior of a single GroEL ring. Our observations show that the C-terminal tails restrain the conformational dynamics of the GroEL ring, most likely through direct interactions with the upper apical domains of the GroEL subunits, a constraint that is progressively released as ATP binds. These results support a model in which the C-terminal tails act as an entropic regulator of the GroEL reaction cycle: transient interactions between the tails and GroEL apical domains restrain premature ring opening and tune the energetic threshold for productive engagement by the smaller GroES co-chaperonin. By linking disordered tail dynamics to the classically cooperative reorganization of the GroEL ring, this mechanism enforces an ordered allosteric cascade that minimizes wasteful formation of empty GroEL-GroES cavities. These findings reveal how the conformational properties of an intrinsically disordered element can be exploited to optimize the energetic efficiency and functional timing of a large allosteric machine.

biophysics↗

bioRxiv2026

Automatic bioinformatic software named entity recognition from literature

Bioinformatics software and databases are essential components of modern life science research, yet their mentions in the scientific literature are often inconsistent and difficult to systematically identify at scale. The lack of a comprehensive and up-to-date catalog of bioinformatics resources hinders efforts toward automated biomedical knowledge extraction and streamlined data analysis. Here we present SNAIL, a hybrid named entity recognition framework designed to automatically identify bioinformatics software and database (SW/DB) names from biomedical texts. SNAIL integrates complementary lexical and semantic modeling strategies. The lexical component captures orthographic patterns and contextual cues characteristic of SW/DB names, while the semantic component leverages contextual embeddings generated by transformer-based language models such as SciBERT, combined with an explicit token-masking strategy to enhance entity-focused representations. A large training corpus was constructed automatically through a hybrid pipeline that integrates citation-hinted extraction with large language model-assisted distillation. Evaluation on two independent benchmark datasets and real-world research articles demonstrates that SNAIL substantially outperforms existing approaches, including domain-specific methods such as bioNerDS2 and general-purpose large language models such as ChatGPT, Gemini, Grok and Claude. Applying SNAIL to large-scale literature analysis further reveals distinct journal-level preferences across bioinformatics subfields. These results demonstrate that SNAIL provides an accurate and scalable solution for identifying bioinformatics resources in scientific texts and enables systematic meta-analysis of tool usage and research trends.

bioinformatics↗

bioRxiv2026

Near-infrared optoacoustic modulation of the blood-brain barrier permeability using size-tuned hyperbranched gold nanoconstructs

The blood-brain barrier (BBB) constitutes a major bottleneck for the systemic delivery of most therapeutic agents to the central nervous system. Here, we report near-infrared reversible optoacoustic modulation of the BBB permeability (NIR-ROAMBBB), leveraging endothelial tight junction targeting hyperbranched gold nanoconstructs (HBGNCs) to amplify localized optoacoustic transduction under femtosecond laser excitation. We first synthesized HBGNCs with tunable particle sizes (62-150 nm) and consistent branch morphologies via a seed-mediated growth approach, and uncovered a non-monotonic relationship between particle dimension and optoacoustic output, where the 62 nm HBGNCs generated nearly twofold stronger optoacoustic signal than gold nanorods and gold nanostars under matched excitations. Conjugation with BV11 antibodies against junctional adhesion molecule A increased HBGNC endothelial association and cerebral accumulation, enabling focal and fluence-dependent transient BBB opening (3-6 h) under 800 nm femtosecond pulsed laser excitation, as validated by in vitro trans-endothelial electrical resistance measurements, ex vivo Evans blue extravasation staining, and in vivo NIR imaging. Featuring deep tissue penetration of NIR light, robust optoacoustic conversion of HBGNCs, and negligible femtosecond laser-induced photothermal damage, this non-invasive strategy enables precise focal modulation of BBB permeability and potential drug delivery.

bioengineering↗

bioRxiv2026

Public-Data Reanalysis Links MS4A4A to M2-like Human Myeloid States and Supports a Predicted Four-Pass Transmembrane Fold

Background/Objectives: MS4A4A is associated with M2-like macrophage states, while the MS4A gene cluster modifies soluble TREM2 levels and Alzheimer's disease risk. We asked whether MS4A4A consistently marks the M2 side of human myeloid activation and whether AlphaFold supports a proposed MS4A4A-MS4A6A interaction. Methods: We reanalysed four public human datasets: bulk RNA-seq and ATAC-seq of primary monocyte-derived macrophages from independent three-donor cohorts, and single-cell RNA-seq atlases of healthy liver and severe COVID-19 blood. MS4A4A and an MS4A4A-MS4A6A complex were modelled with AlphaFold 3 and evaluated using pLDDT, predicted aligned error, and ipTM. Results: MS4A4A was higher in M2 (IL-4) than in M1 (IFN-gamma; + LPS) macrophages in all three donors (log2 fold change +2.68, adjusted P = 0.0013). Its promoter showed the highest mean accessibility in M2. MS4A4A was macrophage-enriched in liver and monocyte-enriched in blood, and was detected in 76.7% of M2-like versus 39.3% of M1-like liver macrophages, with the difference driven mainly by the proportion of positive cells. AlphaFold confidently modelled the four transmembrane helices (mean pLDDT 83.1), but the predicted MS4A4A-MS4A6A interface was not supported (ipTM 0.59). Conclusions: MS4A4A is consistently associated with the M2 side of human myeloid activation across independent transcriptomic, chromatin, and single-cell datasets. The findings are associative, and the proposed MS4A4A-MS4A6A interface remains an untested structural hypothesis.

genomics↗

bioRxiv2026

Brain-Language Alignment During Naturalistic Reading and Its Disruption by Mind-Wandering

Encoding models offer a principled framework for linking computational representations of language to neural activity, but most electroencephalography (EEG) evidence for brain-language alignment comes from tightly controlled, word-by-word reading paradigms. Whether such alignment is detectable during naturalistic reading, and how it is affected by lapses in attention, remains unclear. We addressed these questions using ROAMM, a multimodal dataset containing simultaneous EEG and eye-tracking recordings with time-resolved mind-wandering (MW) annotations from 44 participants reading naturalistic texts. Ridge regression encoding models were trained to predict fixation-aligned EEG spectral power and fixation-related potentials (FRPs) from five word-embedding models (GloVe, word2vec, BERT, GPT-2, and Llama 3). Using permutation testing with false discovery rate correction, we found statistically reliable brain-language alignment across both feature types, with contextual embeddings outperforming static embeddings. Spectral alignment was strongest in the alpha and low-beta bands over parietal electrodes, while FRP-based alignment peaked 200-300 ms after fixation onset over central and parietal-occipital regions. Leveraging ROAMMs span-level MW annotations, we further show that brain-language alignment is systematically reduced during MW, an effect that was substantially larger for oscillatory (PSD) than for event-related (FRP) features. These findings demonstrate that modern language-model representations are reflected in EEG activity during naturalistic reading despite the modalitys inherent noise, and that fluctuations in attention constitute an underappreciated source of variability in brain-language encoding studies.

neuroscience↗

bioRxiv2026

An AI System for Autonomous Algorithm Evolution in Drug Development

Artificial intelligence (AI) is increasingly permeating the drug development pipeline. Numerous algorithms for accelerating this multi-stage and multi-task process have been constructed, which depends heavily on expert design and labor-intensive task-specific optimization. Given that AI-driven acceleration of drug development is recognized as a cumulative, often synergistic, effect across multiple stages, the autonomous evolution of existing algorithms across the entire pipeline is demanded to achieve a holistic advancement. Here, we present DrugEvolve, a multi-role large language model system for systematic and autonomous algorithm evolution in drug development. DrugEvolve realizes a closed-loop evolution process by incorporating Researcher, Engineer, and Analyst domains, and enables an iterative design, implementation, evaluation, and refinement of algorithm by leveraging scientific knowledge and accumulated evolutionary experience. Across eleven representative tasks spanning target identification, drug discovery, preclinical study, and clinical trial, DrugEvolve autonomously evolved the corresponding task-specific algorithms and achieved substantial performance enhancement on 120 benchmark test sets. Moreover, it showed robust generalizabilities across heterogeneous data modalities (ranging from biological sequence and graph to molecular topology and textual language), and realized gains in both predictive and generative tasks. Collectively, this AI system can serve not only as an algorithmic infrastructure for drug development, but also as a transferable paradigm for broader scientific domains.

pharmacology and toxicology↗

bioRxiv2026

PlantOmicsGWAS: An end-to-end, reproducible framework for plant genome-wide association and genomic prediction using linear and pan-genome references

Genome-wide association studies (GWAS) play a crucial role in unraveling the genetic foundations of complex traits in plants but are also hampered by the application of heterogeneous tools, incompatible file formats and disparate computational environments. Existing GWAS frameworks are often restricted to a single linear reference genome, limiting the capacity for the analysis of structural variations and presence/absence variations (PAV) within plant populations. These issues pose obstacles to reproducibility, scalability, and comprehensive investigations. Here, we present PlantOmicsGWAS, an open-source Python framework for reproducible plant genome-wide association analysis and genomic prediction. It integrates reference indexing, FASTQ quality control, alignment, variant calling, VCF normalization, PLINK conversion, linkage disequilibrium analysis, population-structure estimation, association testing, marker scoring, genomic prediction, and visualization within a unified Linux and HPC workflow. The framework supports conventional linear-reference analyses and includes an optional pangenome-oriented module for working with multiple assemblies and graph-derived variation. Using a Vitis benchmark dataset containing 120 accessions and 118,247 graph-derived variants, PlantOmicsGWAS reduced manual workflow fragmentation and generated standardized association outputs. This tool provides a modular and extensible platform for plant GWAS and pan-GWAS workflows while retaining compatibility with established command-line tools and common genotype formats. The GWAS workflow described herein is adaptable to a range of sequencing methods and plant genomes, bridging research on crop related issues across various biological levels, from the individual organism to entire populations. PlantOmicsGWAS implements Bayesian sparse linear mixed modeling (BSLMM) through GEMMA for multi-trait association discovery, while also supporting FaST-LMM, regression-based approaches, and machine-learning algorithms (Random Forest, XGBoost) as benchmarking alternatives. The PlantOmicsGWAS, a versatile toolkit is available at GitHub https://github.com/plantomicsgwas1-boop/PlantOmicsGwas_V1 and on Linux and HPC platform (https://pypi.org/project/PlantOmicsGwas/1.0.2/).

bioinformatics↗

bioRxiv2026

A cross-modal generative model for incomplete and degradedprostate MRI with multicentre clinical validation

Missing or degraded sequences can limit prostate multiparametric MRI. We developed MSCNet, a sequence-conditioned cross-modal generative framework for reconstructing unavailable contrasts and restoring degraded acquisitions. Across ten completion tasks, task-specific MSCNet achieved mean structural similarity of 0.818 versus 0.798 for the strongest task-matched comparators; matched-capacity analyses showed larger differences in lesion fidelity and boundary preservation. In a blinded 1,000-case reader study, overall image quality met the prespecified non-inferiority criterion for DWI, ADC and T2W completion, but not T1W. In a separate 200-case diagnostic assessment, AUCs for clinically significant cancer were 0.860 with acquired images, 0.841 with MSCNet and 0.797 with baseline-generated images. A locked 186-case three-hospital cohort supported multicentre transportability. These retrospective results support quality-controlled cross-modal reconstruction as an adjunct to acquired prostate MRI.

bioengineering↗