Biology subjects

Zhan, Z.

Publications and source records attributed to Zhan, Z..

7 recordsLinked to original sources

bioRxiv2026

MorphOTU: A universal image-based framework for delineating biodiversity discovery

Biodiversity can be quantified only after organisms are assigned to reproducible units, yet most individuals encountered in nature lack reliable species-level identifications. Molecular operational taxonomic units can organize unnamed diversity, whereas image-based approaches generally depend on predefined species levels. Here, we show that operational biodiversity units can be derived directly from phenotypes. We developed morphOTU, a framework combining self-supervised representation learning, operational metric supervision, and adaptive hierarchical clustering to organize specimen images in continuous phenotypic space. Across five benchmark datasets comprising flowers, wood anatomy, and beetle habitus, morphOTUs recovered coherent species-level structure and produced -diversity estimates close to those obtained from expert identifications. This structure remained informative when species were excluded from representation learning, when labeled data were sparse, and per-species sampling was limited. In a heterogeneous field-survey dataset of 4,717 insects representing 269 species across 12 orders, fine-tuning on only 28 common species produced diversity estimates close to expert labels (Shannon index, 3.75 versus 3.53). Visual explanations localized variation to biologically meaningful structures, including body outliers, surface sculpture, floral symmetry, and wood vessels. Morphological units therefore provide an operational layer for organizing and quantifying biodiversity before, alongside, or in the absence of formal species names.

bioinformatics↗

bioRxiv2026

Functional In-Context Learning in Genomic Language Models with Nucleotide-Level Supervision and Genome Compression

AO_SCPLOWBSTRACTC_SCPLOWGenomic foundation models aim to learn general-purpose representations directly from DNA sequence, enabling sequence understanding, generation, and probabilistic reasoning across a wide range of biological tasks. Scaling such models to genomic lengths, however, remains challenging due to the tension between long-range context, nucleotide-level resolution, and practical computational efficiency. Architectural innovations have enabled increasingly long nominal inputs, but often struggle to translate additional context into meaningful performance gains, particularly in the presence of sparse functional signal along eukaryotic genomes. In this work, we revisit the design of long-context genomic foundation models from the perspective of training objective and data construction. We introduce Factorized Nucleotide Supervision (FNS), which reconciles efficient k-mer tokenization with single-nucleotide likelihoods through probability marginalization, and Genome Compression Pretraining (GCP), which reshapes the training distribution by concentrating on gene-centric and regulatory regions. Together, these techniques enable standard transformer-based models to perform functional in-context learning without sacrificing nucleotide-level fidelity or computational efficiency. Building on these ideas, we present a family of autoregressive genomic foundation models supporting contexts of up to 98k base pairs across eukaryotic and prokaryotic genomes. Across training-free evaluations and downstream fine-tuning benchmarks, our models consistently improve over prior approaches and match or exceed state-of-the-art baselines while enabling substantially more efficient inference. Together, these results demonstrate that aligning supervision and data regimes with the biological structure of genomic sequence provides a principled and effective path toward scalable and biologically faithful genomic language modeling. Models, data, and scripts for downstream analyses are publicly available at https://huggingface.co/GenerTeam.

genomics↗

bioRxiv2025

ER-to-Golgi transport machinery promotes the excessive cargo-triggered unfolded protein response

Disruptions to ER homeostasis trigger the unfolded protein response (UPR) to restore proteostatic balance. While defects in the secretory machinery are known to induce ER stress, it remains unclear whether specific trafficking components directly modulate UPR signaling. Here, we demonstrate that neuronal overexpression of the gap junction protein UNC-9 activates the IRE1-XBP1 arm of the UPR in C. elegans. Genetic deletion of ERGIC2 or ERGIC3--genes encoding COPII-associated proteins required for UNC-9 transport--suppresses this UPR activation, revealing an unexpected role for these factors beyond cargo trafficking. Mechanistically, ERGIC2 and ERGIC3 interact with the ER chaperone BiP, facilitating its release from IRE1 to enable UPR and alleviate cargo aggregation. Our findings redefine the UPR as a process dynamically regulated by early secretory components and provide novel insights into how cells integrate trafficking demand with stress adaptation, with implications for ER stress-associated diseases such as neurodegeneration.

cell biology↗

bioRxiv2025

Unexpected diversity of Isidoides (Anthozoa: Octocorallia: Isidoidae) revealed by morphology and phylogenomics analysis with descriptions of four new species

Isidoidae Heestand Saucier, France & Watling, 2021 is a rare octocoral family currently represented by a single genus and species, Isidoides armata Nutting, 1910 recorded in the western Pacific Ocean. The taxonomic status and diversity of Isidoides is unclear, due to the lack of diagnostic taxonomic features and limited taxon sampling. Based on 23 Isidoides specimens obtained from the northwestern to southwestern Pacific, we carried out morphological and phylogenetic analyses to reveal the taxonomic status of new species and develop reliable features for species identification. The 23 specimens could be classified into four well-supported clades by the phylogenomic analysis of ultraconserved elements (UCEs), four groups by 28S rDNA, and two groups by mtMutS-cox1. Integrating morphology and molecular data, we uncovered unexpected diversity of Isidoides composed of the known species Isidoides armata and three new species, viz., I. elegans sp. nov., I. gracilis sp. nov. and I. pseudarmata sp. nov. The morphological analysis showed high intraspecific morphological variation in colony color and the size, shape and arrangement of polyps. By contrast, sclerite forms with their surface sculpturing are more diagnostic features for species identification. Our phylogenetic and species delimitation analyses indicate that UCEs have higher resolution than the nuclear 28S rDNA and the mitochondrial genes mtMutS and cox1 for species discrimination within Isidoidae. ZooBank: urn:lsid:zoobank.org:pub:392485F5-502E-4383-B153-45B167571190

zoology↗

bioRxiv2025

Beyond Natural Antibodies: Scaffold-Based Generation of Novel Anti-3CLpro Nanobody Nb01

In the post-pandemic era, continuous mutations and persistent infections of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) pose a significant threat to global health, making the normalization of preventive measures imperative. Due to its high conservation, the 3CLpro is relatively stable across all these variants without significant changes, suggesting that drugs targeting this enzyme could be effective against all variant viruses. And compared to traditional antibodies, nanobodies show significant advantages against SARS-CoV-2 and its variants. Traditional antibodies often lose their inhibitory activity due to viral mutation. Nanobodies are characterised by their small size, high stability, high affinity for antigen binding, and high water solubility, which enable them to capture viruses quickly and effectively address the continuous mutations and persistent infections of SARS-CoV-2. In this research, we devised a strategy to produce nanobodies by utilizing a fragment-generating large language model, resulting in the identification of a nanobody named Nb01. Nb01 exhibits potent and broad-spectrum neutralizing activity against numerous SARS-CoV-2 variants. The nanobody Nb01, directed against the SARS-CoV-2 3CLpro, demonstrated efficacy against the majority of prevalent SARS-CoV-2 variants. Notably, it has a higher affinity than the current best performing nanobodies (S43, bn03, R14, and 3-2A2-4) for most of the variants, including the Alpha (27.5%), Gamma (29.7%), Omicron BA.2 (32.2%), BA.4/5 (81.2%), BF.7 (64.2%) and XBB (5.5%) variants. For other variants, Nb01 displayed affinities that were on par with these benchmark nanobodies. In summary, the exceptional specificity, low toxicity, robust stability, and extensive spectrum of Nb01 indicate its potential to be developed as a nanobody therapeutic for the management of SARS-CoV-2 infections and its diverse variants.

bioinformatics↗

bioRxiv2021

The REF6-dependent H3K27 demethylation establishes transcriptional competence to promote germination in Arabidopsis

Seed germination is a critical developmental switch from a dormant state to active growth, which involves extensive changes in metabolism, gene expression and cellular identity. However, our understanding of epigenetic and transcriptional reprogramming during this process is limited. The histone H3 lysine 27 trimethylation (H3K27me3) plays a key role in regulating gene repression and cell fate specification. Here, we profile H3K27me3 dynamics and dissect the function of H3K27 demethylation during germination. Our temporal genome-wide profiling of H3K27me3 and transcription reveal delayed H3K27me3 reprogramming compared with transcriptomic changes during germination, with H3K27me3 changes mainly occurring when the embryo is entering into vegetative development. REF6-mediated H3K27 demethylation promotes germination but does not significantly contribute to H3K27me3 dynamics during germination, but rather stably establishes an H3K27me3-depleted state permissive to transcription. By analyzing REF6 genomic binding, we show that it is absent from mature embryo chromatin and gradually establishes occupancy during the course of germination to counteract increased PRC2 activity. Our study provides key insights into the dynamics of gene expression and H3K27me3 during seed germination and suggests the function of H3K27me3 in facilitating cell fate switch. Furthermore, we reveal the importance of H3K27 demethylation-established transcriptional competence in germination and likely other developmental processes.

plant biology↗

bioRxiv2021

Population genomics of Salmonella enterica serovar Weltevreden ST365, an emerging predominant causative agent of diarrheal disease

Salmonella enterica serovar Weltevreden is a recently emerged pathogen, and as such we lack a comprehensive knowledge of its microbiology, genomics, epidemiology and biogeography. In this study, we analyzed 174 novel S. Weltevreden isolates including 111 isolates recovered from diarrheal patients in China between 2006 and 2017. Our results demonstrate that the ST365 clone was the predominant causative agent of the diarrhea-outbreak during this period, as vast majority of the isolates recovered from diarrheal patients belonged to this sequence type (97.37%, 74/76). We also determined the ST365 clone as the predominant sequence type of S. Weltevreden from diarrheal patients globally from previously published sequences (97.51%, 196/201). In order to determine the possible antimicrobial genes and virulence factors associated with S. Weltevreden, we performed whole genome sequencing on our novel isolates. We were able to identify a range of key virulence factors associated with S. Weltevreden that are likely to be beneficial to their fitness and pathogenesis. Furthermore, we were able to isolate a novel 100.03-kb IncFII(S) type virulence plasmid that used the same replicon as pSPCV virulence plasmid. Importantly, we demonstrated through plasmid elimination a functional role for this plasmid in bacterial virulence. These findings are critical to further our knowledge of this high consequence pathogen. ImportanceSalmonella Weltevreden is a newly emerged foodborne pathogen and has caused several outbreaks of diarrheal diseases in some regions in the world. However, comprehensive knowledge of microbiology, genomics, epidemiology and biogeography of this newly emerged pathogen is still lack. In this study, we made an unexpected discovery that S. Weltevreden sequence type (ST) 365 is the causative agent in the diarrhea-outbreak in China and many other regions of the world. We also shown that this sequence type was widely recovered from animal, food, and environmental samples collected in different regions in the world. Importantly, we discovered a novel IncFII(S) type virulence plasmid commonly carried by S. Weltevreden strains of both human, animal, and food origins. These data facilitate future studies investigating the emergence of S. Weltevreden involved in diarrheal outbreaks and the global spread of S. Weltevreden strains.

microbiology↗