By integrating corrected sequences of EV family with their serotype around 48,382 records, we are the first team who have applied deep learning approach (CNN) to classify these 308 genotypes of EV family with out-performance. Moreover, we have constructed the first web application in deep learning that is fully automatic to provide preciserapid prediagnosis on EV genotype.

We are the first team to composited the pipeline by the filter for homology searchCNN model to ensure the submitted sequences belonged to EV family then classify them into suitable genotype. Meanwhile, we implement the web application that is fully automatic to provide preciserapid prediagnosis on EV genotype by submissions, especially for EV-71D-68 as 99.397.8, respectively.

The web application of EV-CNN can perform the genotyping immediately on those long reads from biopsies in third-generation sequencers like oxford nanopore.  That means we can identifyprevent those enteroviruses caused severe syndromes rapidly to reach the goal of prediagnosis. This integrated platform will be applied to clinical laboratories, research community for disease surveillance.