Time-series trend of pandemic SARS-CoV-2 variants visualized using batch-learning self-organizing map for oligonucleotide compositions

Abstract

To confront the global threat of coronavirus disease 2019, a massive number of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genome sequences have been decoded, with the results promptly released through the GISAID database. Based on variant types, eight clades have already been defined in GISAID, but the diversity can be far greater. Owing to the explosive increase in available sequences, it is important to develop new technologies that can easily grasp the whole picture of the big-sequence data and support efficient knowledge discovery. An ability to efficiently clarify the detailed time-series changes in genome-wide mutation patterns will enable us to promptly identify and characterize dangerous variants that rapidly increase their population frequency. Here, we collectively analyzed over 150,000 SARS-CoV-2 genomes to understand their overall features and time-dependent changes using a batch-learning self-organizing map (BLSOM) for oligonucleotide composition, which is an unsupervised machine learning method. BLSOM can separate clades defined by GISAID with high precision, and each clade is subdivided into clusters, which shows a differential increase/decrease pattern based on geographic region and time. This allowed us to identify prevalent strains in each region and to show the commonality and diversity of the prevalent strains. Comprehensive characterization of the oligonucleotide composition of SARS-CoV-2 and elucidation of time-series trends of the population frequency of variants can clarify the viral adaptation processes after invasion into the human population and the time-dependent trend of prevalent epidemic strains across various regions, such as continents.

Unsupervised explainable AI for the collective analysis of a massive number of genome sequences: various examples from the small genome of pandemic SARS-CoV-2 to the human genome

“…The comparison of March 2020, when almost all clades first appeared, with January 2021 showed that their sequences presented differential locations, and August 2020 showed an intermediate pattern. The existence of such time-series changes shows that the separation on the BLSOM has biological significance, as previously reported by Abe et al (2021).…”

“…When studying the 2009 swine-derived flu pandemic (H1N1/2009), we detected directional time-series changes in the oligonucleotide composition due to possible adaptations to the new host, namely, humans (Iwasaki et al, 2011a), and these findings have shown that near-future prediction was possible, albeit partially (Iwasaki et al, 2013a). We found similar time-series changes in the oligonucleotide composition for ebolavirus, MERS coronavirus (Wada et al, 2016(Wada et al, , 2017 and SARS-CoV-2 (Wada et al, 2020b;Iwasaki et al, 2021;Abe et al, 2021). In the present paper, SARS-CoV-2 genomes, which are of great interest to society, were used as an operative example to illustrate how this unsupervised explainable AI supports efficient knowledge discovery from big sequence data.…”

“…We have recently analyzed time-series changes in short and long oligonucleotide compositions in a large number of SARS-CoV-2 genomes and found many oligonucleotides that are expanding rapidly in the virus population, which allowed us to predict candidate advantageous mutations for growth in human cells (Wada et al, 2020b;Ikemura et al, 2020). Furthermore, the oligonucleotide BLSOM can classify the virus sequences into not only the known clades but also their subgroups (Abe et al, 2021). After the above-mentioned publications, a large number of sequences of this virus have continued to accumulate and will exceed one million.…”

AI for the collective analysis of a massive number of genome sequences: various examples from the small genome of pandemic SARS-CoV-2 to the human genome

“…When studying the 2009 swine-derived flu pandemic (H1N1/2009), we detected directional timeseries changes in the oligonucleotide composition due to possible adaptations to a new host, namely humans (Iwasaki et al, 2011a), and these findings have shown that near-future prediction was possible, albeit partially (Iwasaki et al, 2013b). We found similar time-series changes in the oligonucleotide composition for Zaire ebolavirus, MERS coronavirus (Wada et al, 2016(Wada et al, , 2017 and SARS-CoV-2 (Ikemura et al, 2020;Wada et al, 2020b;Abe et al, 2021;Iwasaki et al, 2021).…”

“…We have recently analyzed time-series changes in short and long oligonucleotide compositions in a large number of SARS-CoV-2 genomes and found many oligonucleotides that are expanding rapidly in the virus population, which allowed us to predict candidate advantageous mutations for growth in human cells (Ikemura et al, 2020;Wada et al, 2020b;Iwasaki et al, 2021). Furthermore, the oligonucleotide BLSOM can classify the virus sequences into not only the known clades but also their subgroups (Abe et al, 2021). After the above-mentioned publications, a large number of sequences of this virus have continued to accumulate and will exceed one million.…”