Extensive novel diversity and phenotypic associations in the dromedary camel microbiome are revealed through deep metagenomics and machine learning

• Published in: PloS one Vol. 20; no. 7; p. e0328194
• Main Author: Mubaraki, Fathi A.
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 17.07.2025; Public Library of Science (PLoS)
• Subjects: Analysis; Animal diseases; Animals; Biology and Life Sciences; Biomarkers; Camelids; Camelus - microbiology; Camelus dromedarius; Captivity; Comparative analysis; Computer and Information Sciences; Correlation analysis; Dairy cattle; Diet; Disease; Economic importance; Environmental conditions; Feces; Gastrointestinal Microbiome - genetics; Gut microbiota; Intestinal microflora; Learning algorithms; Livestock; Machine Learning; Medicine and Health Sciences; Metagenome; Metagenomics; Metagenomics - methods; Microbiomes; Microbiota; Microbiota (Symbiotic organisms); Microorganisms; Nutrition research; Phenotype; Physical characteristics; Physical properties; Physiology; Research and analysis methods; Research ethics; Sample size; Species diversity; Taxonomy; Saudi Arabia
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0328194

The dromedary camel, also known as one-humped camel or Arabian camel, is iconic and economically important to Arabian society. Its contemporary importance in commerce and transportation, along with the historical and modern use of its milk and meat products for dietary health and wellness, make it an ideal subject for scientific scrutiny. The gut microbiome has recently been associated with numerous aspects of health, diet, lifestyle, and disease in livestock and humans alike, as well as serving as an exploratory and diagnostic marker of many physical characteristics. Our initial pilot analysis of 55 camel gut microbiomes from the Fathi Camel Microbiome Project uses deep metagenomic shotgun sequencing to reveal substantial novel species-level microbial diversity, for which we have generated an extensive catalog of prokaryotic metagenome-assembled microorganisms (MAGs) as a foundational microbial reference database for future comparative analysis. Exploratory correlation analysis shows substantial correlation structure among the collected subject-level metadata, including physical characteristics. Machine learning using these novel microbial markers, as well as statistical testing, demonstrates strong predictive performance of microbial taxa to distinguish between multiple dietary and lifestyle characteristics of dromedary camels. We present strongly predictive machine learning models for camel age, diet (especially wheat intake), and level of captivity. These findings and resources represent substantial strides toward understanding the camel microbiome and pave the way for a deeper understanding of the nuanced factors that shape camel health.