Fig. 6

COG classification of Ensifer species specific and genus conserved gene sets. The left side horizontal histogram (a) shows the percentage of predicted PEGs from conserved gene sets as classified in COG classes and processes. The COG classification of the core Ensifer proteome at the genus level (estimated using BBH with 70% identity threshold from 28 predicted proteomes from 8 species) is shown as black bars on the left side graph. Predicted proteomes conserved in E. aridi (blue), E. meliloti (pink), E. medicae (green) and E. fredii (purple) were obtained by subtracting the core Ensifer proteomes to core species proteomes (See Methods). The right side horizontal histogram (b) shows for each COG class the logarithmic base 2 ratios of the species conserved gene set fraction by that found in the core Ensifer gene set. A negative value indicates higher representation of the corresponding class in the core genus gene set as compared to the species conserved gene set and conversely, a positive value suggests that the corresponding class is more represented in the species conserved gene set. The color code used for the species enriched or core Ensifer gene sets is indicated at the bottom of the graph. COG classes, their respective descriptions and the process to which they belong to are indicated in black text in the figure. Values that are significantly different (p value < 0.05) within a COG class upon comparison of k proportions using the Marascuillo procedure as implemented in XLStat are indicated by a red “*” (a) and different red letters in (b) indicate intra-class significant differences in the log2 ratios between species conserved and genus conserved gene set fractions