Accurate Identification of SARS-CoV-2 from Viral Genome Sequences using Deep Learning

Alejandro Lopez-Rincon; Alberto Tonda; Lucero Mendoza-Maldonado; Eric Claassen; Johan Garssen; Aletta D. Kraneveld

We downloaded the dataset from the NGDC repository [6] on March 15 15 2020. We removed repeated sequences and applied the whole procedure to translate the data into the sequence feature space. This leave us with a frequency 215 table of 3,827 features with 583 samples (Table 3 ). Next, we ran a state-of-theart feature selection algorithm [36] , to reduce the sequences needed to identify different virus strain to the bare minimum. Remarkably, we are then able to classify exactly all samples using only 53 of the original 3,827 sequences, obtaining a 100% accuracy in a 10-fold cross-validation with a simpler and more 220 traditional classifier, such as Logistic Regression. Table 3 : Organism, assigned label, and number of samples in the unique sequences obtained from the repository [6] . We use the NCBI organism naming convention [30] .