AI Science Tracker
EN

Leveraging machine learning to uncover multi-pathogen infection dynamics across co-distributed frog families

Museum of Southwestern Biology, Department of Biology, University of New Mexico, Albuquerque, New Mexico, United States; Department of Biology, University of Central Florida, Orlando, Florida, United States · 29 Jan 2025 · biology

Discovery

Leveraging machine learning to uncover multi-pathogen infection dynamics across co-distributed frog families

Source check: Source checked. Not independent scientific confirmation.

Finding: Among 1,281 sampled frogs, about 20% had at least one pathogen detected. Ranidae accounted for 74.3% of infections and had higher Bd intensity than the other two families. RF models identified host species and temperature-linked variables as key predictors; Bd/Rv infections were associated with higher latitudes and cooler, more stable temperatures, while Perkinsea trends differed. Three pathogens were screened/quantified by qPCR. RF models used overfitting checks, cross-validation and repeated 70/30 train/test splits (100 iterations); infection-status models covered up to 1,281 samples, with a 216-sample Bd intensity model. Results are observational and non-causal; the paper cautions about small positive groups/overfitting and does not sample across seasons.

Technology: Balanced random-forest classification/regression models predicted infection status and intensity from host taxonomy, age, sex, geography and environmental variables; variable importance plus one-way analyses identified pathogen-specific predictors.

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Balanced random-forest classification/regression models predicted infection status and intensity from host taxonomy, age, sex, geography and environmental variables; variable importance plus one-way analyses identified pathogen-specific predictors.
Test and result
Among 1,281 sampled frogs, about 20% had at least one pathogen detected. Ranidae accounted for 74.3% of infections and had higher Bd intensity than the other two families. RF models identified host species and temperature-linked variables as key predictors; Bd/Rv infections were associated with higher latitudes and cooler, more stable temperatures, while Perkinsea trends differed.
Limitation / still unknown
Three pathogens were screened/quantified by qPCR. RF models used overfitting checks, cross-validation and repeated 70/30 train/test splits (100 iterations); infection-status models covered up to 1,281 samples, with a 216-sample Bd intensity model. Results are observational and non-causal; the paper cautions about small positive groups/overfitting and does not sample across seasons.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://peerj.com/articles/18901/ · https://pmc.ncbi.nlm.nih.gov/articles/PMC11786709/

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary source full text Methods and Results; version/date checked against bibliographic record..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse biology discoveries.