Exploring the mutation tendency of a specific region in the virus taxonomic groups using data mining techniques

The GDNQ motif is mainly found in nonsegmented (-)ssRNA viruses and plays a role in the catalytic and polymerase activity pathways. However, in some recently reported dsRNA mycoviruses, the GDNQ motif of the (-)ssRNA virus group is found instead of the GDD motif commonly found in the dsRNA virus group. Therefore, in this study, data mining techniques were used to explore the mutation tendency of a specific region in the virus taxonomic groups. Viral protein sequences were obtained from the NCBI Virus data repository, and the performance of several known string-searching algorithms (BF, KMP, and BM) was evaluated to establish an optimal strategy for extracting useful information from sequence big data. Then, the pattern-matching performance of the algorithm was tested to detect whether the conserved region sequence was a common motif type or a mutated motif type. As a result of the analysis, it was confirmed that the BF method produced the fastest results. However, the BM method recorded higher accuracy. Through the virus mutation tendency derived from the analysis results, it was confirmed that 7.1% of dsRNA viruses had a mutated motif type based on the total data analyzed and that it was relatively recent that ssRNA virus motif variants began to be discovered in dsRNA viruses. It is expected that additional in-depth analysis of the host range and infection route of viruses with the mutation will help understand the evolution of dsRNA and ssRNA mycoviruses.