ALLERDET
Description
A web app that predicts food allergens based on (1) pairwise sequence alignment, which identifies similar regions between biological sequences, and (2) use of Restricted Boltzmann Machines and Decision Trees. Researchers achieved an accuracy of 97.26%. The app is located at http://allerdet.frangam.com/.
Creator
University of Granada, University of Sevilla
Information
Pediatrics or Adult
Pediatrics/Adult
Speciality
Immunology
Modality
Amino Acid Sequence (FASTA Format)
Training
4670 total allergenic sequences
Github
Publication
FDA