EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predicting dairy cattle PL via longitudinal rumen microbiome dynamics using machine learning approaches

Panping Yang, Zhuoxuan Wu iD, Sixi Zhang, Lei Yan, Yuefei Yan, Feng Zhou, Xiaoli Ren, Yujia Li, Shuhua Gu, Shanshan Gao, Zhiyang Zhang, Yang Zhang und 3 weitere

Microbiology Spectrum · 05.05.2026 · 14 (5)

DOI: 10.1128/spectrum.02969-25 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Microbiology Spectrum, erschienen am 05.05.2026. Verfasst von Panping Yang und 14 weiteren Autor:innen. Beteiligte Einrichtung: College of Animal Science and Technology, Nanjing Agricultural University, Nanjing, Jiangsu, China. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aging is a spontaneous biological process involving intricate regulatory mechanisms over time. Studies in mice and humans indicate that the gut microbiota is closely linked to the aging process and plays an important role in it. However, the relationship between the rumen microbiota and aging in dairy cows remains unclear. In this study, we characterized rumen microbial differences across different parities in 341 dairy cows using 16S rRNA amplicon sequencing and identified microbial markers associated with productive lifespan (PL) and farm profitability through machine learning analysis. Our findings reveal that as parity increases, the rumen microbiota undergoes systematic succession: alpha diversity indices significantly decrease, microbial interactions weaken, the abundance of Proteobacteria increases, while the abundance of Bacteroidetes decreases in higher-parity cows. By integrating machine learning with 16S sequencing, we identified characteristic microbial markers predictive of PL and farm profitability. Specifically, the support vector regression model achieved a predictive performance with an area under the curve (AUC) of 0.788 and identified eight key genera associated with the PL of dairy cows. Meanwhile, the random forest (RFTEST) model attained an AUC of 0.763 and selected eight key microorganisms linked to the economic benefits of the farm, with fivefold cross-validation confirming the reliability of RFTEST. Combined with SHapley Additive exPlanations (SHAP) analysis, the genus-level taxa Eubacterium_hallii_group and Prevotella_7 can serve as indicator strains for PL and farm profitability in dairy cows. Therefore, alterations in the rumen microbiota may serve as a key driver of aging in dairy cows. This study aims to provide insights for improving PL and farm profitability through the modulation of rumen microbiota. IMPORTANCE In the dairy industry, longevity is a critical economic trait that directly impacts overall farm profitability. Although dairy cows have a natural lifespan of approximately 20 years—with optimal productivity often extending beyond the fifth parity—their average PL is only about 2.7 parities. Identifying factors influencing PL is therefore crucial. Given the vital role of the rumen microbiota in regulating dairy performance, milk fat/protein synthesis, and other key physiological processes, elucidating its correlation with PL is essential for developing probiotic interventions to enhance longevity. Furthermore, early detection of aging-associated microbial signatures could facilitate proactive adjustments to feeding strategies. Notably, this is the first study to link parity-driven microbiome succession with PL prediction in dairy cattle. Consequently, by identifying microbial molecular markers linked to PL and potential probiotic targets, this study highlights promising opportunities to improve dairy cow health and advance sustainable dairy farming practices.

Schlagwörter

Stichwörter der Autor:innen

biomarkerdairy cowmachine learningparityrumen microbiota

Autor:innen und Institutionen

  • College of Animal Science and Technology, Nanjing Agricultural University, Nanjing, Jiangsu, China.
  • Henan Dairy Herd Improvement Center, Zhengzhou, Henan, China.
  • Runan County Product Quality Inspection and Testing Center, Zhumadian, Henan, China.
  • Henan Provincial International Joint Laboratory for Dairy Health Farming, Zhengzhou, Henan, China.
  • Henan Seed Industry Development Center, Zhengzhou, Hennan, China.
  • Henan Dairy Production Performance Measurement Co., Ltd., Zhengzhou, Henan, China.

ORCID-Profile: Zhuoxuan Wu, Huixia Li, Zhen Zhang

Zitieren

APA

Yang, P., Wu, Z., Zhang, S., Yan, L., Yan, Y., Zhou, F., Ren, X., Li, Y., Gu, S., Gao, S., Zhang, Z., Zhang, Y., Zhu, H., Li, H., & Zhang, Z. (2026). Predicting dairy cattle PL via longitudinal rumen microbiome dynamics using machine learning approaches. Microbiology Spectrum, 14(5). https://doi.org/10.1128/spectrum.02969-25

Vancouver

Yang P, Wu Z, Zhang S, Yan L, Yan Y, Zhou F, et al. Predicting dairy cattle PL via longitudinal rumen microbiome dynamics using machine learning approaches. Microbiology Spectrum. 2026;14(5). doi:10.1128/spectrum.02969-25

BibTeX
@article{Yang2026,
  author = {Yang, Panping and Wu, Zhuoxuan and Zhang, Sixi and Yan, Lei and Yan, Yuefei and Zhou, Feng and Ren, Xiaoli and Li, Yujia and Gu, Shuhua and Gao, Shanshan and Zhang, Zhiyang and Zhang, Yang and Zhu, Huiyuan and Li, Huixia and Zhang, Zhen},
  title = {Predicting dairy cattle PL via longitudinal rumen microbiome dynamics using machine learning approaches},
  journal = {Microbiology Spectrum},
  year = {2026},
  volume = {14},
  number = {5},
  doi = {10.1128/spectrum.02969-25}
}
Kostenlos starten →