XFXuepeng Fu

Xuepeng Fu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xuepeng Fu has a total of 47 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Differential Rhizosphere Bacterial Communities Are Associated with Saline–Alkali Tolerance in Soybean Genotypes

    by Hong Xue, Jiyuan Guo, Yujia Zhai, Song Liu, Weiwei Zhang, Wenyue Li, Xuepeng Fu

    Saline–alkali stress limits soybean production in the Songnen Plain of Northeast China. We screened 32 soybean genotypes at the sprout stage and evaluated 14 selected genotypes in a seedling pot experiment. We profiled rhizosphere bacteria in the three highest-ranked and three lowest-ranked genotypes from the pot screen. Under stress, the sensitive genotypes exhibited greater reductions in shoot biomass. While overall rhizosphere bacterial diversity and community composition overlapped between tolerance groups, the relative abundances of putative beneficial taxa, including Bacillus and Rhizobium, were enriched in the tolerant group. Community assembly patterns were consistent with homogeneous selection in both groups. Notably, co-occurrence network analysis revealed a clear direction of topological shifts: sensitive genotypes formed larger and more complex networks, whereas tolerant genotypes exhibited more compact networks with higher modularity. These findings suggest a microecological trade-off, where the highly modular structure of the tolerant rhizosphere may restrict the propagation of environmental disturbances and support local community stability. This study provides novel insights into the structural adaptation of rhizosphere communities in saline–alkali-tolerant soybeans.

  • Open Access

    SIMPLEX: composition-space deep learning for bio-inspired hydrogel adhesion with prospective validation on model-discovered formulations and interpretable composition rules

    by Tongfei Shen, Xueqin Lei, Hung Ka Lok, Huaicheng Li, Zhongze Pan, Xiaohui Wang, Xuepeng Fu

    Fast and accurate composition-to-property prediction is critical for accelerating the discovery of multicomponent bio-adhesive hydrogels, yet current surrogate models are predominantly evaluated under random data splits that do not reflect the distribution shift encountered in real-world iterative screening. Moreover, existing deep learning approaches for small-sample formulation data offer little systematic guidance on architecture design and often depend on laboriously engineered interaction features. This study introduces Simplex-Inspired Multimodal Predictive Learning Encoder with X-modal Fusion (SIMPLEX), a dual-modality deep encoder purpose-built for composition-driven property estimation. Its gated multimodal fusion component autonomously learns inter-component interactions without hand-crafted quadratic terms. We establish a fully reproducible benchmark covering eight models with unified hyper-parameter tuning and statistical protocols. Evaluated on 316 internal hydrogel formulations under grouped cross-validation, SIMPLEX achieves first-tier predictive performance, competitive with the strongest classical baselines. On 25 prospectively collected formulations acquired from a Bayesian-optimized closed-loop campaign, a setting that introduces a deliberate distribution shift toward high-performance compositions, SIMPLEX delivers the highest coefficient of determination ( R 2 = 0.6712 ) for adhesion-strength prediction, meeting the demand for precise quantitative estimation in iterative candidate screening. Feature dissection reveals that the native compositional representation already captures the dominant composition–property relationships, and component-wise ablation pinpoints gated multimodal fusion as the principal performance driver. This study delivers a capable deep encoder together with a rigorous, transparent evaluation paradigm for formulation-driven machine learning, providing immediately reusable baselines and design principles for the community.

  • Open Access

    Contrasting Seed Protein Phenotypes Are Associated with Nodulation and Rhizosphere Microbiome Assembly in Soybean

    by Jiyuan Guo, Lixue Wang, Song Liu, Hong Xue, Weiwei Zhang, Xuepeng Fu, Quanzhong Dong

    Soybean seed protein content varies with genotype, nitrogen acquisition, and environmental conditions. Associations between belowground microbiota and seed protein content remain insufficiently characterized. We compared three high-protein cultivars (KD91, SN148, and KD80) and three low-protein cultivars (KD122, KD96, and BD40) to examine relationships among seed composition, nodulation, and root-associated microbial communities. High-protein cultivars produced more nodules, while nodule bacterial diversity and community composition showed no significant between-group differences and were dominated by Bradyrhizobium. In contrast, rhizosphere bacterial and fungal communities differed between protein groups. Dyadobacter, Pseudoxanthomonas, and Chitinophaga had higher relative abundance in the high-protein rhizosphere. Seed protein content was positively associated with Dyadobacter (R2 = 0.234, p = 0.017) and Pseudoxanthomonas (R2 = 0.432, p < 0.001), but not with Chitinophaga. This study compared distinct cultivars grown in a single batch of peat-based greenhouse substrate; the observed microbial patterns are associated with cultivar group and cannot be attributed to seed protein phenotype alone. Dyadobacter and Pseudoxanthomonas are candidates for future functional validation.

  • Open Access

    Inhibition Mechanism of Bacillus velezensis TS-2 Against Streptomyces scabies

    by Yudong Sun, Xuanrui Zheng, Song Liu, Lichun Wang, Weiwei Zhang, Xuepeng Fu

    Potato common scab is a soil-borne disease caused by Streptomyces spp., which affects the commodity value of potato worldwide. This study evaluated the inhibitory effect and potential mechanisms of the Bacillus velezensis TS-2 strain against Streptomyces scabies, providing a basis for developing safe and effective biocontrol agents for potato common scab. Strain TS-2 was isolated from soil adhering to the surface of healthy potato tubers. Confrontation assays, scanning electron microscopy, and pot experiments showed that strain TS-2 significantly reduced the potato common scab incidence and inhibited the mycelial growth and spore formation of S. scabies. Whole genome sequencing identified strain TS-2 as Bacillus velezensis. Transcriptome analysis showed that B. velezensis TS-2 significantly downregulated key genes involved in carbohydrate, lipid, and amino acid metabolism in S. scabies, including aceB, glyA, acd, thereby disrupting the core carbon and nitrogen energy supply network essential for pathogen growth and development. B. velezensis TS-2 also induced structural damage to the cell wall (pbpA), cell membrane (ino1) and bubble sac (gvpA), reducing environmental adaptability. Noticeable downregulation of genes in the bld–whi morphological differentiation cascade and the quorum-sensing gene afsA was observed under TS-2 stress, which correlates with suppressed hyphal maturation and sporulation of S. scabies.

  • Open Access

    Predicting Root Colonization by Plant Growth-Promoting Rhizobacteria: Opportunities and Limits of Deep Learning

    by Xiaohui Wang, Zhixuan Zhu, Linhang Li, Jie Liu, Ka Lok Hung, Tahir Iqbal, Jan Neil Shih, Zhongze Pan, Zihan Miao, Mengtong Chen, et al.

    Plant growth-promoting rhizobacteria (PGPR) can support crop growth and protection, but their performance varies across strains, hosts, soils, and management conditions. Colonization measurements can help explain this variation, although persistence, microbial function, and plant benefit are distinct outcomes. This critical review examines how genome-informed models could prioritize strains for defined colonization tasks. We connect the ecology of dispersal, attachment, competition, and persistence with phenotype measurements, sequence representations, and validation design. Protein language models, DNA-based encoders, pan-genome features, and multi-task learning provide candidate tools, supported mainly by applications to related prediction problems. Their value for strain-level persistence must be assessed against phenotype-based and simpler genomic baselines under matched evaluation conditions. We organize the evidence around three application directions, examine assumptions about laboratory-to-field transfer, functional annotation, and trait independence, and propose a metadata schema and benchmark framework. Independent strain and soil holdouts, endpoint-specific measurements, and prospective field studies are central to assessing added predictive value. The resulting framework is a proposal for testing when genomic information and model complexity improve strain prioritization, with scope determined by the available labels and validation setting.

  • Open Access

    Alfalfa with Forage Crop Rotation Alleviates Continuous Alfalfa Obstacles through Regulating Soil Enzymes and Bacterial Community Structures

    by Yanxia Xu, Zhuxiu Liu, Zhongbao Shen, Zhao Yang, Xuepeng Fu, Xiaolong Wang, Shasha Li, Hua Chai, Ruoding Wang, Xiaobing Liu, et al.

    Alfalfa is a perennial herbaceous forage legume that is significantly and adversely affected by monocropping. Crop rotation is the most effective measure to overcome continuous cropping obstacles. However, the mechanisms of how bacterial communities are affected and the potential links between these effects and cropping systems remain poorly understood. Based on a long-term field experiments with continuous alfalfa and forage crops with alfalfa rotation in the black soil region of the western Songnen Plain in Northeast China, the alterations in soil bacterial community structure using high-throughput sequencing of the 16S rRNA gene and soil chemical properties and enzyme activities were analyzed. The alfalfa–forage oats–silage maize–alfalfa and alfalfa–silage maize–forage oats–alfalfa system significantly increase the levels of total phosphorus and available phosphorus, and promote the activities of acid phosphatase, β-glucosidase, leucine aminopeptidase, and N-acetyl-β-glucosaminidase in comparison to continuous alfalfa. While alfalfa crop rotation did not affect the α-diversity of soil bacteria, it significantly altered the bacterial community composition and structure. Some key taxa are significantly enriched in the crop rotation system soils, including Bacillus, Sphingobium, Paenibacillus, Hydrogenispora, Rubrobacter, Haliangium, and Rubellimicrobium. Additionally, crop rotation with alfalfa increased the stability and complexity of the soil bacterial co-occurrence network. Based on our findings, we recommend promoting the alfalfa–forage oats–silage maize–alfalfa and alfalfa–silage maize–forage oats–alfalfa rotation systems as ideal practices for overcoming the challenges associated with continuous cropping of alfalfa. These systems not only enhance soil nutrient content and enzyme activities but also foster a beneficial microbial community, ultimately improving soil functionality and crop performance.


  • Open Access

    Removal of Dominant Species Impairs Nitrogen Utilization in Co-Existing Ledum palustre and Vaccinium uliginosum Communities Subjected to Five-Year Continuous Interruptions

    by Yadong Duan, Xuepeng Fu, Xingang Zhou, Danmei Gao, Lei Zhang, Fengzhi Wu

    Unguided exploitation has impaired the sustainability of natural resources of agronomic non-wood forest plants from understory boglands in boreal forests. The extreme consequences of plant–soil interplay on medicinal plant communities under continuous interruptions need to be understood to implement strategies which can cope with possible ecological degradation. In this study, co-existing Ledum palustre and Vaccinium uliginosum communities were investigated after a four-year interruption of continuous removal of dominant species in stands at Xing’an Mountain. Nitrogen (N) availability was assessed by above-ground biomass and N content in nondominant plants and the biophysiological properties of rhizosphere soil. The removal treatment promoted soil mineral and organic N contents, but also reduced abundances of the soil communities of Rozellomycota phylum (by 82.76%), ericoid mycorrhiza of Meliniomyces varia (by 81.60%) and Phialocephala fort (by 69.54%). Vaccinium uliginosum overcame L. palustre through higher N utilization (biomass/%N) although the latter had higher abundances of soil Odiodendron maius and P. fort. The microbial community attributes accounted for a large proportion of N availability following the removal of dominance. In conclusion, our study demonstrates that understory agronomic plants in northern boglands should no longer be under continuous exploitation. Strategies should be considered to improve the promotion of N uptake by managing local soil microbial communities.