Scholarlink

Background <p>The Kazakh mare, an indigenous breed of Xinjiang, exhibits strong adaptability to arid and cold conditions while maintaining relatively stable milk production under low-input extensive farming systems. However, its genetic improvement has been constrained by traditional management practices.</p> Results <p>In this study, we monitored milk yield and milk composition over a 105-day lactation period and recorded 15 phenotypic traits. Milk yield was significantly correlated with body length, teat diameter, and teat length. The high-yield (HY) versus low-yield (LY) and high-fat (HF) versus low-fat (LF) comparisons identified 286 and 627 differentially expressed genes (DEGs), respectively. Several candidate genes were identified, including <i>PMP22</i>, <i>FAM83A</i>, <i>HSD17B3</i>, <i>AGPAT4</i>, <i>SLC50A1</i>, and <i>ERBB3</i>, which were associated with pathways including PI3K-Akt signaling, MAPK signaling, and triglyceride metabolism. High-yield mares showed metabolic differences characterized by enrichment of pathways related to the tricarboxylic acid (TCA) cycle, suggesting altered energy and intermediary metabolism involving carbohydrates, lipids, and amino acids. Metabolites associated with these differences included glycerone, α-D-glucose, D-galactose, glycerol, L-histidine, and anserine.</p> <p>In addition, machine learning analysis identified GLDC as a candidate gene potentially associated with milk fat percentage, possibly through its association with histidine. However, this relationship requires further validation.</p> Conclusion <p>Using peripheral blood samples, this study integrated differential expression analysis, mixed linear models, and machine learning approaches to identify candidate genes and metabolites associated with lactation performance in Kazakh mares. The results provide preliminary insights into molecular and metabolic features associated with lactation traits in this breed. These findings may serve as a reference for future molecular breeding and nutritional studies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Identification of candidate genes and metabolites associated with lactation performance in Kazakh mares using blood multi-omics and machine learning

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月6日

查看全文
  • Chen Meng,
  • Penghui Luo,
  • Wanlu Ren,
  • Xiaoyu Xie,
  • Yaqi Zeng,
  • Jianwen Wang,
  • Xinkui Yao,
  • Jun Meng

摘要

Background

The Kazakh mare, an indigenous breed of Xinjiang, exhibits strong adaptability to arid and cold conditions while maintaining relatively stable milk production under low-input extensive farming systems. However, its genetic improvement has been constrained by traditional management practices.

Results

In this study, we monitored milk yield and milk composition over a 105-day lactation period and recorded 15 phenotypic traits. Milk yield was significantly correlated with body length, teat diameter, and teat length. The high-yield (HY) versus low-yield (LY) and high-fat (HF) versus low-fat (LF) comparisons identified 286 and 627 differentially expressed genes (DEGs), respectively. Several candidate genes were identified, including PMP22, FAM83A, HSD17B3, AGPAT4, SLC50A1, and ERBB3, which were associated with pathways including PI3K-Akt signaling, MAPK signaling, and triglyceride metabolism. High-yield mares showed metabolic differences characterized by enrichment of pathways related to the tricarboxylic acid (TCA) cycle, suggesting altered energy and intermediary metabolism involving carbohydrates, lipids, and amino acids. Metabolites associated with these differences included glycerone, α-D-glucose, D-galactose, glycerol, L-histidine, and anserine.

In addition, machine learning analysis identified GLDC as a candidate gene potentially associated with milk fat percentage, possibly through its association with histidine. However, this relationship requires further validation.

Conclusion

Using peripheral blood samples, this study integrated differential expression analysis, mixed linear models, and machine learning approaches to identify candidate genes and metabolites associated with lactation performance in Kazakh mares. The results provide preliminary insights into molecular and metabolic features associated with lactation traits in this breed. These findings may serve as a reference for future molecular breeding and nutritional studies.