Faecal microbiota composition based random forest model predicts Mycobacterium Avium subsp. Paratuberculosis (MAP) shedding severity in cattle
Paratuberculosis is a devastating infectious disease caused by Mycobacterium avium subsp. paratuberculosis (MAP). The development of the paratuberculosis clinical symptoms in cattle could take up to a few years and vastly differs between individuals in severity of the symptoms and shedding of the pathogen in the environment. Identification of high shedding animals that significantly increase the burden of the pathogen in a farm environment is essential for paratuberculosis control and minimization of economic losses. Widely used methods for detection and quantification of MAP, such as culturing and PCR based techniques rely on direct presence of the pathogen in a sample and have little to no predictive value for the disease development. In the current study we investigated possibility of prediction of the shedding severity through the life of a cow based on fecal microbiota composition. Twenty calves were experimentally infected with MAP and fecal samples were collected biweekly up to four years of age. All collected samples were subjected to culturing on the selective media to obtain data about shedding severity. Faecal microbiota were profiled in a subset of samples that reflects important time points in cattle husbandry. Using faecal microbiota composition and shedding intensity data we build a random forest classifier for prediction of the animals shedding status. We found that machine learning approaches applied to microbial composition can be used to classify cows that are severely shedding MAP into the environment. However, classification accuracy strongly correlates with age of the animals and use of samples from older individuals results higher precision of classification. Classification model based on samples from the first 12 month of life showed AUC between 0.78 and 0.79, where is model based on samples from animals older than 24 month showed AUC between 0.91 and 0.92 (95% CI). . We also showed that only a relatively small number of microbial taxa are important for classification and could be considered as biomarkers. The study provides evidence for the link between microbiota composition and severity of MAP infection and shedding, as well as lays ground for development of predictive diagnostic tools based on the microbiota composition.
| Main Authors: | , , , , |
|---|---|
| Format: | Dataset biblioteca |
| Published: |
Wageningen Bioveterinary Research
|
| Subjects: | Multispecies, Mycobacterium Avium, Raw sequence reads, |
| Online Access: | https://research.wur.nl/en/datasets/faecal-microbiota-composition-based-random-forest-model-predicts- |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Paratuberculosis is a devastating infectious disease caused by Mycobacterium avium subsp. paratuberculosis (MAP). The development of the paratuberculosis clinical symptoms in cattle could take up to a few years and vastly differs between individuals in severity of the symptoms and shedding of the pathogen in the environment. Identification of high shedding animals that significantly increase the burden of the pathogen in a farm environment is essential for paratuberculosis control and minimization of economic losses. Widely used methods for detection and quantification of MAP, such as culturing and PCR based techniques rely on direct presence of the pathogen in a sample and have little to no predictive value for the disease development. In the current study we investigated possibility of prediction of the shedding severity through the life of a cow based on fecal microbiota composition. Twenty calves were experimentally infected with MAP and fecal samples were collected biweekly up to four years of age. All collected samples were subjected to culturing on the selective media to obtain data about shedding severity. Faecal microbiota were profiled in a subset of samples that reflects important time points in cattle husbandry. Using faecal microbiota composition and shedding intensity data we build a random forest classifier for prediction of the animals shedding status. We found that machine learning approaches applied to microbial composition can be used to classify cows that are severely shedding MAP into the environment. However, classification accuracy strongly correlates with age of the animals and use of samples from older individuals results higher precision of classification. Classification model based on samples from the first 12 month of life showed AUC between 0.78 and 0.79, where is model based on samples from animals older than 24 month showed AUC between 0.91 and 0.92 (95% CI). . We also showed that only a relatively small number of microbial taxa are important for classification and could be considered as biomarkers. The study provides evidence for the link between microbiota composition and severity of MAP infection and shedding, as well as lays ground for development of predictive diagnostic tools based on the microbiota composition. |
|---|