Classification and prediction of Mycobacterium Avium subsp. Paratuberculosis (MAP) shedding severity in cattle based on young stock heifer faecal microbiota composition using random forest algorithms

Comparison of Machine Learning Tree-Based Algorithms to Predict Future Paratuberculosis ELISA Results Using Repeat Milk Tests

Machine learning algorithms have been applied to various animal husbandry and veterinary-related problems; however, its use in Johne's disease diagnosis and control is still in its infancy. The following proof-of-concept study explores the application of tree-based (decision trees and random forest) algorithms to analyze repeat milk testing data from 1197 Canadian dairy cows and the algorithms' ability to predict future Johne's test results. The random forest models using milk component testing results alongside past Johne's results demonstrated a good predictive performance for a future Johne's ELISA result with a dichotomous outcome (positive vs. negative). The final random forest model yielded a kappa of 0.626, a roc AUC of 0.915, a sensitivity of 72%, and a specificity of 98%. The positive predictive and negative predictive values were 0.81 and 0.97, respectively. The decision tree models provided an interpretable alternative to the random forest algorithms with a slight decrease in model sensitivity. The results of this research suggest a promising avenue for future targeted Johne's testing schemes. Further research is needed to validate these techniques in real-world settings and explore their incorporation in prevention and control programs.

Pathological, microscopic, and molecular diagnosis of paratuberculosis/John's disease in naturally infected dromedary camel (Camelus dromedarius)

Background and Aim

Paratuberculosis (PTB) or John's disease is a chronic disease of ruminants impeding the reproduction and productivity of the livestock sector worldwide. Since there is a lack of pathological studies explaining the nature and development of the disease in camels, this study aimed to highlight the anatomopathological changes of PTB in camels, which may help in verifying and validating some diagnostic tests used to detect the etiology of the disease in camel tissues.

Materials and Methods

In August 2017, at Alselaa border's Veterinary Clinic of Al Dhafra Region, Western Abu Dhabi, UAE, one imported culled she-camel of 2 years old was subjected to clinical, microscopic, and anatomopathological investigations along with real-time quantitative polymerase chain reaction (q-PCR) to confirm the infection and correlate between clinical signs and pathological lesions of the PTB in dromedary camels.

Results

Clinically, typical clinical signs compliant with the pathognomonic gross and histologic lesions of PTB were seen in naturally infected dromedary camel. As presumptive diagnosis microscopically, acid-fast coccobacillus bacterium clumps were demonstrated in direct fecal smears as well as in scraped mucosal and crushed mesenteric lymph node films, and in histopathological sections prepared from a necropsied animal and stained by Ziehl-Neelsen stain. Free and intracellular acid-fast clump phagosomes were further confirmed as Mycobacterium avium subsp. paratuberculosis by q-PCR.

Conclusion

Clinical signs and pathological lesions of paratuberculosis in a dromedary camel were found to be similar to those of the other susceptible hosts.

Metabolomic changes in polyunsaturated fatty acids and eicosanoids as diagnostic biomarkers in Mycobacterium avium ssp. paratuberculosis (MAP)-inoculated Holstein-Friesian heifers

Mycobacterium avium subspecies paratuberculosis (MAP) is the causative organism of Johne's disease, a chronic granulomatous enteritis of ruminants. We have previously used naturally MAP-infected heifer calves to document metabolomic changes occurring in MAP infections. Herein, we used experimentally MAP-inoculated heifer calves to identify biomarkers for MAP infections. At 2-weeks of age, 20 Holstein-Friesian (HF) calves were experimentally inoculated with MAP. These calves, along with 20 control calves, were sampled biweekly up to 13-months of age and then monthly up to 19-months of age. Sera were assessed using flow infusion electrospray high-resolution mass spectrometry (FIE-HRMS) on a Q Exactive hybrid quadrupole-Orbitrap mass spectrometer for high throughput, sensitive, non-targeted metabolite fingerprinting. Partial least squares-discriminate analysis (PLS-DA) and hierarchical cluster analysis (HCA) discriminated between MAP-inoculated and control heifer calves. Out of 34 identified metabolites, six fatty acyls were able to differentiate between experimental groups throughout the study, including 8, 11, 14-eicosatrienoic acid and cis-8, 11, 14, 17-eicosatetraenoic acid which were also detected in our previous study and so further suggested their value as biomarkers for MAP infection. Pathway analysis highlighted the role of the alpha-linoleic acid and linoleic acid metabolism. Within these pathways, two broad types of response, with a rapid increase in some saturated fatty acids and some n-3 polyunsaturated fatty acids (PUFAs) and later n-6 PUFAs, became predominant. This could indicate an initial anti-inflammatory colonisation phase, followed by an inflammatory phase. This study demonstrates the validity of the metabolomic approach in studying MAP infections. Nevertheless, further work is required to define further key events, particularly at a cell-specific level.