Interpretable models for species identification and origin authentication

Datharsis, a University of Granada spin-off specialised in explainable omics data analysis, is looking to join a consortium as the partner responsible for data analysis, either leading the data analysis work package or contributing specific analysis tasks.

Fighting wildlife trafficking often depends on identifying the species and geographical origin of seized samples from their chemical, spectral or omics fingerprint. That evidence has to hold up before inspectors and courts, so a black-box prediction is not enough. The model must show which variables support each classification and how reliable it is.

Our team has tackled the same problem in food fraud, verifying the declared origin of avocado samples and comparing classification models by how well their conclusions hold.