Figure 2.

Representation of the latent space and classification of different macrodomains. (A) Three dimensional scatter plot of the latent space variable χ₁, χ₂ and χ₃. The data has been clustered using K-means clustering. The various color-coded clusters are representing distinct macrodomains (MDs) within the bacterial chromosome. (B) The comparison between experimentally denoted and machine learning (ML)-derived MDs. The inner and outer circles, each encoded by various color-coded regions, delineate the experimentally denoted and ML-derived macrodomains, respectively, with base pair information annotated in kilo bases (kb). (C) The bar plot of the F1-score for different MDs. The higher values of the F1-Score indicated the better classification.