Biology subjects

Si, Z.

Publications and source records attributed to Si, Z..

6 recordsLinked to original sources

bioRxiv2026

Pollution and Anthropogenic Stressors Are Associate with Cetacean Vulnerability in Coastal Waters: Fine-Scale Diagnostics from eDNA and Multispecies Modeling

Indo-Pacific humpback dolphin (Sousa chinensis) and finless porpoise (Neophocaena phocaenoides) are increasingly threatened across their native range, yet the relative influence of multiple stressors in shaping their population dynamics remains unclear. Current conservation strategies for both species are limited by incomplete data and limited assessment of affecting factors. Here, we integrated eDNA metabarcoding with Joint Species Distribution Modeling (JSDM) to assess how environmental gradients, pollution, and trophic associations interactively influence cetacean distributions in Hong Kong waters. We show that degraded water quality and intensified human activity negatively associated with cetacean occurrence, with clear species-specific responses to different stressors. S. chinensis covaried most strongly with Secchi disc depth, and presence of vessels, while N. phocaenoides was negatively associated with nitrate nitrogen and microbial pollution (sewage). The JSDM variance partitioning analysis highlighted that the occurrence of S. chinensis was primarily associated with anthropogenic disturbances (30.04%), followed by water physical properties (26.63%), whereas N. phocaenoides was more strongly associated with physical (40.9%) and anthropogenic disturbances (35.2%). By integrating eDNA and JSDM, our approach provides fine-scale diagnostics of species-specific vulnerabilities, supporting adaptive conservation strategies and guiding the realignment of protected areas to mitigate biodiversity loss in urbanized marine ecosystems. Environmental ImplicationOur study demonstrates that hazardous water pollutants, including microbial contamination, nutrient enrichment, and chemical stressors, vessel pressure, show strong, species-specific impacts on resident cetaceans in Hong Kong. By integrating eDNA metabarcoding with joint species distribution models, we provide a diagnostic framework that directly links pollutant profiles to ecological risk. These findings highlight that conventional conservation strategies overlooking pollution drivers are insufficient for marine megafauna persistence. Our approach offers an early-warning system for monitoring hazardous pollutants in coastal ecosystems and supports adaptive management strategies to mitigate biodiversity loss in urbanized seascapes.

ecology↗

bioRxiv2025

Developing a versatile technology for agroinfiltration in multiple plants

Agrobacterium-mediated transient transformation methods are widely used in plant science and molecular pharming. However, plant immune response and post-transcriptional gene silencing (PTGS) can compromise gene expression, limiting the adaptability of these methods across plant species. The lack of a robust reporter system further hinders advancements in agroinfiltration technology. To address these challenges, we developed a fungal bioluminescence pathway (FBP)-based reporter system that utilizes caffeic acid to generate self-sustained luminescence. By incorporating NahG and P19, our FBP assay enhances transient transformation efficiency in over 20 plant families, including major crops, vegetables, ornamental plants, and trees. Furthermore, we demonstrate its utility for assessing protein localization, interactions, and transcriptional regulation in multiple plant species, thereby broadening our understanding of gene function throughout the plant kingdom.

plant biology↗

bioRxiv2023

EfNST: A composite scaling network of EfficientNet for improving spatial domain identification performance

Spatial Transcriptomics (ST) leverages Gene Expression Profiling while preserving Spatial Location and Histological Images, enabling it to provide new insights into tissue structure, tumor microenvironment, and biological development. The identification of spatial domains serves as not only the foundation for ST research but also a crucial step in various downstream analyses. However, accurately identifying spatial domains using computational methods remains a tremendous challenge due to the poor computational performance of many existing algorithms. Here, we propose EfNST, a deep learning algorithm based on a composite scaling network of the EfficientNet Network, designed specifically for the analysis of 10X Visium spatial transcriptomics data. We applied EfNST to three different datasets: human Dorsolateral Prefrontal Cortex, human breast cancer and mouse brain anterior. EfNST outperforms five advanced competing algorithms, achieving the best Adjusted Rand Index (ARI) scores of 0.554, 0.607, and 0.466, respectively. Notably, EfNST demonstrated high accuracy in identifying fine tissue structure and discovering corresponding marker genes with an improved running speed. In conclusion, EfNST offers a novel approach for inferring spatial organization of cells from discrete datapoints, facilitating the exploration of new insights in this field.

bioinformatics↗

bioRxiv2023

UDP-glucosyltransferase 71C4 regulates seed development by redistributing phenylpropanoid metabolism in cotton

Seeds are important for plant reproduction, and hence it is important to identify genes regulating seed development. Here, we focused on a member of the glycosyltransferase (GT) family, UDP-glucosyltransferase 71C4 (UGT71C4), which influences the seed width, seed length, and yield traits (lint percentage and seed index) by regulating the lignin and flavonoid pathways of phenylpropane metabolism, affects the oil and protein contents of mature seeds, and controls the seed size by regulating cell proliferation. Overexpression of UGT71C4 leads to seed enlargement by activating expression of peroxiredoxins and the flavonoid metabolism pathway; induces accumulation of ROS, which promotes cell proliferation; and significantly improves yield traits, with the seed index increasing from 10.66 to 11.91 and protein increasing by 5.34%, but oil content decreasing 8.98%. Conversely, knockout of UGT71C4 causes seeds to become smaller; the lignin metabolism pathway to be activated, especially enzymes relating to lignin synthesis leading to increased ectopic deposition of lignin in the ovule and constrained ovule growth and development; and significant improvement of yield traits, with lint percentage increasing from 39.62% to 41.94%, seed index decreasing from 10.66 to 8.60, protein content decreasing 4.28%, and no significant change in oil content. Our research provides new insights into seed size regulation through UDP-glucosyltransferase, providing potential methods for improving plant yield.

plant biology↗

bioRxiv2022

Polyimidazolium protects against an invasive clinical isolate of Salmonella Typhimurium

Frequent outbreaks of Salmonella Typhimurium infection in both the animal and human population with potential for zoonotic transmission pose a significant threat to the public health sector. The rapid emergence and spread of more invasive multidrug-resistant clinical isolates of Salmonella further highlight the need for the development of new drugs with effective broad-spectrum bactericidal activities. Synthesis and evaluation of main-chain cationic polyimidazolium 1 (PIM1) against several gram-positive and gram-negative bacteria have previously demonstrated the efficacy profile of PIM1. The present study focuses on antibacterial and anti-biofilm activities of PIM1 against Salmonella both in vitro and in ovo setting. In vitro, PIM1 exhibited bactericidal activity against all tested three strains of Salmonella at a low dosage of 8 g/ml. Anti-biofilm activity of PIM1 was evident with complete inhibition for the initial attachment of biofilms at 16 g/ml and degradation of pre-formed biofilms in a dose-dependent manner. During the host cell infection process, PIM1 reduces extracellular bacterial adhesion and invasion rates to limit the establishment of infection. Once intracellular, the drug-resistant strain was tolerant and protected from PIM1 treatment. In a chicken egg infection model, PIM1 exhibited therapeutic activity for both Salmonella strains with stationary-phase and exponential-phase inocula. Moreover, PIM1 showed a remarkable efficacy against the stationary phase inocula of drug-resistant Salmonella by eliminating the bacteria burden in >50% of infected chicken egg embryos. Collectively, PIM1 has demonstrated its potential as a drug candidate for treatment of Salmonella infections, as well as a solution to tackle egg contamination issues on poultry farms.

microbiology↗

bioRxiv2022

Atomic model of Vesicular Stomatitis Virus and Mechanism of Assembly

Like other negative-strand RNA viruses (NSVs) such as influenza and rabies, vesicular stomatitis virus (VSV) has a three-layered organization: a layer of matrix protein (M) resides between the membrane envelope, studded by glycoprotein (G), and the nucleocapsid, composed of the nucleocapsid protein (N) and the encapsidated genomic RNA. Lack of in situ atomic structures of these viral components has limited our understanding of the virion assembly mechanism. Here, by cryoEM and sub-particle reconstruction, we have determined the in situ structures of M and N inside VSV at 3.47 [A] resolution. In the virion, N and M have a stoichiometry of 1:2. The in situ structures of both N and M differ from their crystal structures in their N-terminal segments and oligomerization loops. N-RNA, N-N, and N-M-M interactions govern the formation of the capsid. A double layer of M contributes to packaging of the helical nucleocapsid: the inner M (IM) joins neighboring turns of the N helix, while the outer M (OM) contacts G and the membrane envelope. The pseudo-crystalline organization of G is further mapped by cryoET. The mechanism of VSV assembly is delineated by the network interactions of these viral components.

biochemistry↗