Biology subjects

Zhan, J.

Publications and source records attributed to Zhan, J..

At least 19 recordsLinked to original sources

bioRxiv2026

MSMICA: computational metabolite identification in untargeted metabolomics by integrating MS, retention time, and biological evidence

Mass Spectrometry Metabolomics Identification Connection Algorithm (MSMICA) is an algorithm for automated metabolite identification in untargeted liquid chromatography-high-resolution mass spectrometry (LC-HRMS) analyses. Limitations in metabolite identification can occur due to the availability and cost of standards and prevent recognition of metabolic factors impacting human health and disease. MSMICA performs mass-to-charge-ratio matching with chemical structures and clusters of LC-HRMS features for adduct and isotope forms. A local optimization is then used to integrate retention time prediction, metabolite precursor-product and transporter correlations, and biospecimen-specific abundance information for metabolite identification. Applying MSMICA to various internal and external mammalian datasets, validation results showed a 96.2 +- 5.1% correct rate of metabolite identification. When multiple LC-HRMS datasets were used, MSMICA enabled greater metabolite identifications, expanded metabolic pathway coverage, and data harmonization. Thus, MSMICA applies multiple pieces of evidence to substantially improve metabolite identification coverage and accuracy for known metabolites.

bioinformatics↗

bioRxiv2026

Relational Graph Convolutional Networks for Glioblastoma Biomarker Discovery via ceRNA and Copy Number Variation Analysis

Glioblastoma (GBM) is a highly aggressive brain tumor with a five-year survival rate of 6.9%, attributable in substantial part to the shortage of reliable biomarkers. Competing endogenous RNA (ceRNA) and copy number variation (CNV) analyses each carry biomarker-identification potential, but existing work treats them separately and does not integrate multiple regulatory mechanisms. We therefore applied relational graph convolutional networks (RGCNs) to ceRNA and CNV knowledge graphs under a late-fusion ensemble architecture. Across 10-fold cross-validation the RGCN discriminated best among the graph architectures tested (AUCROC 0.874 {+/-} 0.070), significantly exceeding graph convolutional, graph attention and relational attention networks. Combining the ceRNA and CNV branches at the decision level gave the best overall performance (AUCROC 0.883 {+/-} 0.072; PR-AUC 0.208 {+/-} 0.152) and improved on the ceRNA-only model in precision--recall terms, although that improvement does not survive correction for multiple comparisons and we therefore report it as suggestive. Screening the late-fusion ranking against the existing glioma literature left five candidates that are absent from the curated glioblastoma biomarker set and the subject of at most one prior glioma report, among them hsa-miR-203b and hsa-miR-5683, each differentially expressed by more than fivefold on a log_2 scale. All five are computational predictions. Relational graph learning over a ceRNA network, combined with genomic dosage at the decision level, is thus a workable framework for biomarker prioritization, and the five loci give targeted experimental work a place to start.

bioinformatics↗

bioRxiv2026

Anthocyanin-associated cellular programs underlying terroir variation in Cabernet Sauvignon grape berry revealed by SEED-based deconvolution

Plant tissues consist of diverse cell populations that collectively contribute to development, metabolism, environmental responses, and phenotype formation. Although single-cell and single-nucleus RNA sequencing have greatly advanced the study of plant cellular heterogeneity, their application to large sample cohorts remains limited by cost, technical complexity, tissue dissociation constraints, and throughput. In contrast, bulk RNA-seq datasets have accumulated extensively across plant species, tissues, developmental stages, and environmental conditions, yet the celltype-level information embedded in these datasets remains difficult to resolve because plant-oriented deconvolution frameworks are still lacking. Existing deconvolution methods have largely been developed in mammalian systems and have not been systematically optimized for plant transcriptomic features, leaving their applicability under plant-specific constraints unclear. Here, we present SEED, an adaptive deconvolution framework optimized for plant transcriptomic data. SEED integrates candidate reference-template construction with seven deconvolution strategies and automatically identifies an optimal combination for a given dataset. In grapevine simulated benchmarking, SEED showed its clearest advantage under low-replication conditions and remained broadly competitive, rather than uniformly dominant, when larger pseudo-bulk sample sizes were evaluated. SEED further performed robustly in public Arabidopsis thaliana and Nicotiana tabacum datasets. Finally, we applied SEED to bulk RNA-seq data generated in this study from Vitis vinifera cv. Cabernet Sauvignon berries collected from Yinchuan and Yantai, identifying terroir-associated cell subtypes and coordinated celltype interaction patterns. Together, these results establish SEED as a practical framework for plant transcriptome deconvolution and provide a new tool for dissecting cellular heterogeneity associated with environmental adaptation and phenotype formation in plants.

bioinformatics↗

bioRxiv2026

Task demands dynamically structure feature selection, routing, and integration in the human brain

AbstractFlexible visual categorization requires selecting task-relevant information from complex stimuli and transforming it through structured computations. Yet how task demands flexibly control these computations in the human brain remains unresolved. Dynamic faces provide a powerful test case, conveying stable identity through 3D shape and transient emotional expression through 4D movement. Using generative 4D face modelling, participant-specific MEG (N = 24), and information-theoretic analyses, we show that the brain implements a structured processing architecture. Task relevance first gates feature representations in occipital cortex, selectively sustaining relevant features while suppressing irrelevant ones. These features are then routed along pathway-specific channels--3D identity features to the ventral pathway and 4D emotion features to the lateral pathway. Finally, these inputs converge in temporal cortex, where identity and emotion features are synergistically integrated for learned identities. These results show how task demands dynamically structure feature selection, gating, routing, and integration, thereby constraining the computations that support flexible categorization.

neuroscience↗

bioRxiv2026

Two ancient nuclear lineages and divergent reproductive strategies drive global success of the wheat stripe rust pathogen

Understanding the mechanisms for generating and maintaining plant pathogen genetic diversity is essential for predicting disease emergence and safeguarding the food supply. Here, using population-genomic analyses of 507 global isolates of the wheat stripe rust fungus Puccinia striiformis f. sp. tritici, we uncover two nuclear lineages that diverged before modern agriculture and reinforced during wheat domestication. Distinct lineage combinations underpin three major global populations: two predominantly sympatric, homozygous lineages (nuclA-nuclA and nuclB-nuclB) and a globally dominant heterokaryotic lineage (nuclA-nuclB) of hybrid origin. This hybrid population evolves largely clonally, accompanied by recurrent somatic nuclear exchange. Identical virulence traits can arise independently via somatic nuclear exchange in clonal populations or through sexual recombination in recombining populations, revealing convergent evolutionary routes to adaptation. Together, our results provide a transformative nuclear-level perspective linking deep evolutionary history and reproductive strategy to the global success of cereal rust fungi.

genetics↗

bioRxiv2026

ATLAS: Population-Level Disease Locus Discovery via Differential Attention in Genomic Language Models

Identifying disease-associated genetic variants remains a key challenge in genomics, especially in small cohorts or for rare and complex mutation types where genome-wide association studies (GWAS) often fall short. We introduce ATLAS, a population-level framework that leverages attention signals from pretrained genomic language models (gLMs) to detect disease-associated genes and loci directly from raw sequences--without requiring explicit variant calls or supervised training. ATLAS first performs gene-level differential attention analysis to prioritize candidate genes, followed by base-level analysis to localize disease-associated regions at single-haplotype resolution. We validate ATLAS on synthetic and {beta}-thalassemia datasets, demonstrating robust performance across diverse allele frequencies (down to 10%), cohort sizes (below 200 individuals per group), and genomic scales. Compared to GWAS, ATLAS achieves higher recall of known loci and captures haplotype-specific signals missed by traditional methods. Cross-model benchmarking further shows that precise localization depends on both model size and pretraining on diverse human genomes. In summary, ATLAS offers a scalable, sequence-native alternative to traditional statistical genetics.

genetics↗

bioRxiv2025

ABA-responsive transcription factor CbNAC63 is involved in modulating biosynthesis of triterpenoid saponins in Conyza blinii

Conyza blinii, a traditional Chinese medicinal herb, accumulates triterpenoid saponins. NAC transcription factors (TFs) are pivotal for plant stress responses and hormone-mediated secondary metabolism, while abscisic acid (ABA) regulates terpenoid biosynthesis in medicinal plants. However, how NAC TFs mediate ABA-regulated triterpenoid saponin synthesis in C. blinii remains unclear. In order to address this gap, the focus was directed towards CbNAC63--a NAC TF identified from the nocturnal low-temperature transcriptome of C. blinii. Phylogenetic and motif analyses showed it belongs to the stress-responsive NAC subclade with conserved domains. Y2H assays revealed its self-interaction, enhancing transcriptional activity; subcellular localization confirmed nuclear localization. RT-qPCR results indicated CbNAC63 exhibits tissue-specific expression under nocturnal low-temperature stress, accompanied by increased triterpenoid saponin accumulation. ABA signaling primarily regulates this process: exogenous ABA enhanced CbNAC63 expression and saponin content, while fluridone suppressed both. Overexpression (OE) and virus-induced gene silencing (VIGS) demonstrated CbNAC63 positively regulates saponin biosynthesis. Y1H and dual-luciferase reporter assays confirmed CbNAC63 directly binds to the promoters of Cb{beta}AS and CbHMGR and activates their transcription. In summary, the study shows CbNAC63 integrates ABA and low-temperature signals, activating MVA pathway genes to regulate saponin synthesis. This unveils a novel "ABA-CbNAC63-MVA" module, enriches the classic regulatory network, and provides a target for improving medicinal plant quality. HighlightsIdentifies CbNAC63 as a novel regulator that integrates ABA and cold signals to boost saponin production in medicinal plants.

plant biology↗

bioRxiv2025

Deep learning reveals a microprotein atlas in maize and uncovers novel regulators of seed amino acid metabolism

Microproteins represent a vast and functionally important class of genes that remain largely unexplored in plant genomes. Here, we developed DeepMp, a deep learning framework that integrated multi-omics evidence and built the most comprehensive plant microprotein atlas to date, identifying 18,338 high-confidence candidates in maize. The majority of these appear to have originated de novo from regions previously annotated as noncoding, and they show hallmarks of rapid, lineage-specific evolution and pronounced tissue specificity. These novel microproteins were found integrated into core regulatory networks, particularly in organ development and nutrient storage. Focusing on the maize kernel, population-scale analyses linked microprotein expression to natural variation in amino acid content. We functionally validated three grain-filling-specific candidates originating from noncoding regions by CRISPR-Cas9 knockouts, which confirmed their roles as precise modulators of arginine, aspartate, and methionine levels, without pleiotropic effects on kernel morphology. Our study provides a foundational resource and analytical framework, establishes microproteins as a new and functionally important coding layer in maize, and uncovers a previously untapped source of targets for crop improvement.

genomics↗

bioRxiv2025

ESM-LoRA-Gly: Improved prediction of N- and O-linked glycosylation sites by tuning protein language models with low-rank adaptation (LoRA)

Glycosylation associates with many diseases ranging from cancer to neurodegeneration and understanding these disease mechanisms requires the precise identification of glycosylation sites. Computational prediction of glycosylation sites has been useful to complement laborious experimental methods, while existing tools lack sufficient accuracy and scalability. Here, we introduce ESM-LoRA-Gly, a method that employs Low-Rank Adaptation (LoRA) to fine-tune the ESM2-3B protein language model for predicting both N- and O-linked glycosylation sites. According to the evaluation on the benchmark datasets, ESM-LoRA-Gly outperforms existing state-of-the-art techniques. The improvement is particularly significant (>100% in Matthews correlation coefficient) for the O-linked dataset. By substantially reducing trainable parameters while maintaining predictive power, ESM-LoRA-Gly enables computationally efficient proteome-scale predictions. This approach should be instrumental for advancing glycoproteomic research and accelerating therapeutic discovery for glycosylation-related diseases.

bioinformatics↗

bioRxiv2025

RESM: Capturing sequence and structure encoding of RNAs by mapped transfer learning from ESM (evolutionary scale modeling) protein language model

RNA sequences exhibit lower evolutionary conservation than proteins due to their informationally constrained four-letter alphabet, compared to the 20-letter code of proteins. More limited information makes unsupervised learning of structural and functional evolutionary patterns more challenging from single RNA sequences. We overcame this limitation by mapping RNA sequences to pseudo-protein sequences to allow effective transfer training from a protein language model (protein Evolution-Scale Model 2, protESM-2). The resulting RNA ESM (RESM) outperforms 12 existing RNA language models in zero-shot prediction, not only in sequence classification but also in RNA secondary structure and RNA-RNA interaction prediction. Further supervised fine-tuning demonstrates RESMs generalizability and superior performance over the existing models compared across multiple downstream tasks, including mRNA ribosome loading efficiency and gene expression prediction, despite RESM being trained exclusively on noncoding RNAs. Moreover, RESM can generalize to unseen sequences beyond its 1,024-nucleotide training limit, achieving 81.3% improvement over state-of-the-art methods in supervised secondary structure prediction for RNAs up to 4,000 nucleotides, limited only by the available GPU memory, while providing >1000-fold speedup compared to MSA-based approaches. RESM provides a robust foundation for deciphering RNA sequence-structure-function relationships, with broad implications for RNA biology.

bioinformatics↗

bioRxiv2025

Hedyotis Diffusae Herba Mitigates Rotenone-Induced Neurotoxicity via Akt Pathway in SH-SY5Y Cells

Hedyotis Diffusae Herba (HDH), known in traditional Chinese medicine for its heat-clearing and detoxifying properties, has been widely used to manage inflammation-related conditions. However, its neuroprotective potential and mechanism in the context of Parkinsons disease (PD) remain unclear.Network pharmacology and molecular docking were applied to predict the interactions between HDH components and PD-associated targets. Molecular dynamics simulations assessed the binding stability of major compounds to core targets. Rotenone-induced SH-SY5Y cell injury served as an in vitro PD model to evaluate HDHs biological effects. Cell viability, reactive oxygen species (ROS), IL-6 secretion, and Akt phosphorylation were assessed by CCK-8 assay, ROS detection kit, ELISA, and western blot, respectively. MK2206, an Akt inhibitor, was used to validate pathway involvement.Seven bioactive compounds were identified in HDH, among which stigmasterol and quercetin exhibited strong binding affinities with AKT1, IL-6, and JUN. Simulations confirmed stable interactions. HDH significantly increased cell viability and decreased intracellular ROS and IL-6 levels in rotenone-treated SH-SY5Y cells. Akt phosphorylation was restored by HDH, and the effect was blocked by MK2206, indicating Akt pathway participation.HDH exerts neuroprotective effects against rotenone-induced toxicity via antioxidant and anti-inflammatory activities and Akt pathway activation. These findings provide mechanistic insight and pharmacological support for HDH as a potential candidate in complementary therapy for PD.

cell biology↗

bioRxiv2025

Genes govern metabolism-Enzymes define pathways and metabolic relationships

Gene-centric pathway mapping tools, widely used to interpret untargeted liquid chromatography mass spectrometry (LC-MS) metabolomics data, may underperform because a single metabolite can generate multiple spectral features, inflating false positive rates. Classic enzymology, which established metabolite flow before gene sequencing, offers experimentally validated precursor-product relationships that could overcome these ambiguities. We evaluated whether enzymology-defined precursor-product correlations are consistently detectable in human plasma LC-MS data. We detected amino acids, carnitine-related, TCA cycle, and pentose phosphate pathway metabolites in one individual sampled eight times over five years and in 50 adults sampled 6 to 8 times each. In the single participant repeated measures, strong positive correlations were observed for most direct precursor-product pairs. The longitudinal and cross-sectional analyses reproduced these patterns. Precursor-product proportionality, a fundamental principle of enzymology, is detectable in LC-MS datasets and remains consistent across studies. Applying these correlations to metabolomics workflows can improve pathway analysis, help metabolite identification, and reveal how genetic variations, diets, therapeutic drugs, and environmental exposures jointly impact metabolic pathways.

systems biology↗

bioRxiv2025

Designing a broad-spectrum four-helix bundle targeting different strains of SARS-CoV-2 Spike Receptor Binding Domains with ACE2-like binding interface

One major strategy for the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) to evade antibody drugs or preventive vaccines is high mutation rate of the spike receptor binding domain (RBD). Because variable RBDs of different SARS-CoV-2 strains must bind to the same human receptor angiotensin-converting enzyme 2 (hACE2) for viral cell entry and infection, we hypothesize that designing a protein with the same or very similar hACE2 binding interface might have a broad-spectrum effect against various SARS-CoV-2 strains. The designed protein binds specifically to the WT-RBD (with micromolar affinity) but not to RBDs from other SARS-CoV-2 strains. However, two rounds of the E. coli display and Magnetic Cell Sorting (MACS) selection are sufficient to yield a protein named CYN1 with nanomolar binding affinities not only to the WT-RBD but also to those of Omicron BA.1, XBB.1.16, and JN.1. Molecular dynamics simulations and free-energy hotspot analysis revealed that CYN1s broader spectrum capability stems from its engagement of essentially all ACE2 hotspot residues critical for WT-RBD binding, unlike the designed protein. The discovery of CYN1, differing by only four mutations from the designed protein, confirms that targeting the small interface of human viral receptors--rather than the entire receptor--offers a viable strategy for developing broad-spectrum inhibitors. This approach minimizes potential off-target effects arising from receptor multifunctionality.

molecular biology↗

bioRxiv2025

RPcontact: Improved prediction of RNA-protein contacts using RNA and protein language models

Determining key contacts between RNA-protein interactions is essential for understanding the molecular mechanisms of numerous biological processes, including transcription, splicing, and translation. However, progress in this area has been impeded by the scarcity of RNA-protein complex structures in the Protein Data Bank (PDB) and the challenges posed by traditional structural determination techniques. Recent computational advancements, including deep learning methods like AlphaFold 3 and RoseTTAFoldNA, have improved contact prediction but are still limited by the availability of homologous sequences and templates. Here, we introduce RPcontact, a novel computational method designed to predict RNA-protein contacts using large language models tailored for RNA (ERNIE-RNA) and proteins (ESM-2). Despite being trained entirely on ribosomal RNA-protein (rRNA-protein) complexes, RPcontact demonstrates robust and generalized performance in predicting contacts for both dimeric and multimeric non-rRNA-protein complexes. The performance of RPcontact on contact predictions significantly improves over the binary contacts inferred from RNA-protein complex structures predicted by AlphaFold 3 and RoseTTAFoldNA, highlighting its potential in RNA-protein complex structure and function prediction.

bioinformatics↗

bioRxiv2025

Functional Implications of the Conformational Landscape of a Multidrug Transporter Revealed by Structures of Zebrafish Abcb4

The hallmark of multidrug resistance conferred by human ABC transporter ABCB1 (hP-gp) is the recognition and efflux of diverse range of drugs, though the precise mechanism of polyspecificity remains unresolved. In aquatic animals such as zebrafish, Abcb4, a functional homolog to hP-gp, plays a vital role in surviving environmental toxicants. Here, we show that Abcb4 exhibits comparable basal and drug-stimulated ATPase activity to hP-gp. Using cryo-EM, we captured five inward-facing Abcb4 conformations with varying separations between its two lobes, illustrating its open-and- close motion. The range of separation exceeds that seen in published P-gp structures that appear to be conformationally restricted. This global open-and-close motion is coupled with individual helix movement, resulting in a highly fluid substrate-binding pocket. These dynamic changes, likely underlying the polyspecificity of substrate recognition, predict unconventional protein-ligand interactions that are supported by structures of Abcb4 bound to the P-gp inhibitors tariquidar and elacridar, and the substrate vincristine.

biochemistry↗

bioRxiv2025

The TONSL-MMS22L complex and FANCM form an interdependent complex on chromatin to counter replication stress

FANCM is branchpoint DNA translocase essential for cellular response to replication stress. Here, we show that replication stress stimulates FANCM and the TONSL-MMS22L heterodimer bound to histones H3-H4 to form an interdependent complex on chromatin. TONSL-MMS22L recruits FANCM and Fanconi anemia (FA) core complex to stalled and collapsed forks, maintains FANCM on replication-stressed chromatin, promotes FANCD2 monoubiquitination, facilitates both repair and replication traverse of DNA interstrand crosslinks (ICLs), and suppresses sister chromatid exchanges, through its interactions with FANCM and H3-H4. Reciprocally, both DNA translocase activity and phosphorylation of FANCM facilitate recruitment of TONSL-MMS22L and RAD51 to perturbed forks. Moreover, TONSL-MMS22L and FANCM function together to promote activation of the FA pathway, ICL repair, homologous recombination and replication traverse. Cancer patients with tumors with wildtype FANCM and low expression of TONSL-MMS22L have a more favorable prognosis than those with high expression. Thus, FANCM-TONSL-MMS22L acts coordinately as a complex on chromatin that resolves replication stress, and this complex may present a therapeutic target for wildtype FANCM-linked cancer.

cell biology↗

bioRxiv2024

Molecular evolution of a reproductive barrier in maize and related species

Three cross-incompatibility loci each control a distinct reproductive barrier in both domesticated maize (Zea mays ssp. mays) and its wild teosinte relatives. These three loci, Teosinte crossing barrier1 (Tcb1), Gametophytic factor1 (Ga1), and Ga2, each play a key role in preventing hybridization between incompatible populations and are proposed to maintain the barrier between domesticated and wild subspecies. Each locus encodes both a silk-active and a matching pollen-active pectin methylesterase (PMEs). To investigate the diversity and molecular evolution of these gametophytic factor loci, we identified existing and improved models of the responsible genes in a new genome assembly of maize line P8860 that contains active versions of all three loci. We then examined fifty-two assembled genomes from seventeen species to classify haplotype diversity and identify sites under diversifying selection during the evolution of these genes. We show that Ga2, the oldest of these three loci, was duplicated to form Ga1 at least 12 million years ago. Tcb1, the youngest locus, arose as a duplicate of Ga1 before or around the time of diversification of the Zea genus. We find evidence of positive selection during evolution of the functional genes at an active site in the pollen-expressed PME and predicted surface sites in both the silk- and pollen-expressed PMEs. The most common allele at the Ga1 locus is a conserved ga1 allele (ga1-Off), which is specific haplotype containing three full-length PME gene copies, all of which are non-coding due to conserved stop codons and are between 610 thousand and 1.5 million years old. We show that the ga1-Off allele is associated with and likely generates 24-nt siRNAs in developing pollen-producing tissue, and these siRNAs map to functional Ga1 alleles. In previously-published crosses, the ga1-Off allele was associated with reduced function of the typically dominant functional alleles for the Ga1 and Tcb1 barriers. Taken together, this seems to be an example of a type of epigenetic trans-homolog silencing known as paramutation functioning at a locus controlling a reproductive barrier.

genetics↗

bioRxiv2024

Loss function of tumor suppressor FRMD8 confers resistance to tamoxifen therapy via a dual mechanism

Approximately 40% ER-positive breast cancer patients suffer from therapeutic resistance to tamoxifen. Although reduced ER level is the major cause of tamoxifen resistance, the underlying mechanisms remain elusive. Here, we report that FRMD8 raises the level of ER at both transcriptional and post-translational layers. FRMD8 deficiency in MMTV-Cre+; Frmd8fl/fl; PyMT mice accelerates mammary tumor growth and loss of luminal phenotype, and confers tamoxifen resistance. Single-cell RNA profiling reveals that Frmd8 loss decreases the proportion of hormone-sensing differentiated epithelial cells and downregulates the levels of ER. Mechanically, on one hand, loss of FRMD8 inhibits ESR1 transcription via suppressing the expression of FOXO3A, a transcription factor of ESR1. On the other hand, FRMD8 interacts both with ER and UBE3A, and disrupts the interaction of UBE3A with ER, thereby blocking UBE3A-mediated ER degradation. In breast cancer patients, FRMD8 gene promoter is found hypermethylated and low level of FRMD8 predicts poor prognosis. Therefore, FRMD8 is an important regulator of ER and may control therapeutic sensitivity to tamoxifen in ER-positive breast cancer patients. Impact StatementTumor suppressive protein FRMD8 inhibits breast cancer progression by regulating the level of ER and has the possibility to be a potential target for overcoming tamoxifen treatment resistance.

cancer biology↗