Figure 2 | Scientific Reports

Figure 2

From: Variant-driven early warning via unsupervised machine learning analysis of spike protein mutations for COVID-19

Figure 2

Monthly ML analysis and chain variants. The clusters are linked to form chains, which are then identified with emerging variants, as shown in the top plots (A,B). In the middle (C,D) and bottom (E,F) plots we show the number of monthly infected per 100k inhabitants and percentage of occurrence for each emerging variant. The shaded regions represent the relative percentage of the variants present at each time. The left plots (A,C and E) correspond to a cutoff in the Ward distance of \(r_{W} = 200\) while the right ones (B,D and F) to \(r_{W} = 100\). Note that the chains v2 and v3 for \(r_{W} = 200\) can be associated to the Alpha and Delta VoC, respectively.

Back to article page