Figure 2.

(A) Circos showed the top 10 genera of soil bacterial communities in farmland (F) and pristine grassland (CK). The width of the line represents the size of the abundance; (B) Stacked charts represent the soil bacterial genera in croplands (F) and pristine grassland (CK). The Random Forest classification modeling visually showed the dominant genera of bacteria in croplands (F) and pristine grassland (CK); (C)The significance plot was used to assess differences in bacterial genera between treatment groups. In this plot, different colored bars indicate the various treatment groups, with red representing the CK treatment and blue representing the F treatment. The horizontal axis shows the percentage of genera in each treatment group, while the vertical axis lists the names of the different genera; (D) The importance of random forest variables, derived through a categorical algorithm, is highlighted in predicting rare bacterial genera in soils. In this visualization, green bars denote the variables that were selected using the categorical algorithm. These bars illustrate the contribution of each variable to the model, emphasizing their role in the accurate prediction of rare bacterial genera in soil samples. Significant differences of asterisk were marked with * (p < 0.05).