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

Thus, we suggest that the missense mutation in ORF1ab gene (c.11083G > T ) 310 could alters the viral load during the infection between symptomatic and asympotomatic patients [44] . Table 8 . With one extra symptomatic case the presents a missing value, giving the following sequence; TTTGTTCTTTTTTT-TNTATGAAAATGCCTTTTTACC. Figure 19 : 7 of the first 16 sequences clustered together in a 36-bps sequence, which primarily appears in symptomatic cases. The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.03.13.990242 doi: bioRxiv preprint we generate a primer set using Primer3plus [41] . This outputs 5'TTCCAAAGT-GCAGTGAAAAGAA3' as forward primer and 5'TTGCAAAAGCAGA-CATAGCAA3' as reverse primer with a total length of 175 bps. Then, we 320 run an in-silico PCR test using FastPCR 6.7 [42] with default parameters, this yields the results reported in Fig. 22 . Nevertheless, the results of the PCR Amplicons will need to be sequenced, to differentiate between the two possible sequences. Figure 22 : In-silico PCR test results from the FastPCR 6.7 software [42] using sequences TTCCAAAGTGCAGTGAAAAGAA and TTGCAAAAGCAGACATAGCAA as primers, in NC045512.2 used as a reference SARS-CoV-2 sequence.