Table 1.
Overview of BayesTraits results. qij design posterior transition rates, whereas P(i) correspond to posterior ancestral state probabilities. For the membrane (MBN) trait, state 0 = one MBN and state 1 = two MBN, while for the peptidoglycan (PG) trait, state 0 = no PG, state 1 = thin PG and state 2 = thick PG. “H biased” is the model where the hyperprior has been purposely biased to favour a diderm LBCA (see Materials and Methods for details).
| Node | Trait | Statistic | E | H1 | H2 | R1 | R2 | H Biased |
|---|---|---|---|---|---|---|---|---|
| LBCA | MBN | mean q01 | 2.495 | 2.352 | 2.477 | 2.288 | 2.411 | 1.431 |
| LBCA | MBN | mean q10 | 0.132 | 0.113 | 0.121 | 0.012 | 0.008 | 0.210 |
| LBCA | MBN | mean P(0) | 94.951 | 94.204 | 95.375 | 97.134 | 98.161 | 67.092 |
| LBCA | PG | mean P(0) | 22.068 | 4.022 | 38.604 | 0.397 | 0.594 | N/A |
| LBCA | PG | mean P(2) | 76.497 | 94.622 | 60.147 | 99.535 | 99.358 | N/A |
| LBCA | PG | mean q01 | 4.626 | 1.634 | 7.317 | 0.798 | 0.827 | N/A |
| LBCA | PG | mean q02 | 6.935 | 2.020 | 20.967 | 0.953 | 1.041 | N/A |
| LBCA | PG | mean q10 | 0.166 | 0.102 | 0.187 | 0.000 | 0.000 | N/A |
| LBCA | PG | mean q12 | 0.128 | 0.109 | 0.118 | 0.001 | 0.000 | N/A |
| LBCA | PG | mean q20 | 2.088 | 0.937 | 4.941 | 1.347 | 1.413 | N/A |
| LBCA | PG | mean q21 | 1.890 | 2.165 | 1.600 | 1.398 | 1.419 | N/A |
| Firmicutes | PG | mean P(0) | 17.631 | 3.936 | 30.120 | 0.611 | 0.738 | N/A |
| Firmicutes | PG | mean P(2) | 81.891 | 95.648 | 69.435 | 99.378 | 99.237 | N/A |