Table 1.
Parameter estimates and likelihood scores of the nested branch and branch-sites models for detecting selective constrains between GCYC1A and GCYC1B.
| Clade | Model | n.p. | Log Likelihood (lnL) | 2ΔLa | P-valueb | Estimates of ω value |
|---|---|---|---|---|---|---|
| BRANCH MODEL | ||||||
| Null | One-ratio | 39 | −2898.3994 | ω0 = 0.3690 | ||
| GCYC1A as foreground | Two-ratio | 40 | −2895.4538 | 5.8911 | 0.015* | ω0 = 0.4168, ωGCYC1A = 0.2083 |
| GCYC1B as foreground | Two-ratio | 40 | −2898.3696 | 0.05958 | 0.8 | ω0 = 0.3640, ωGCYC1B = 0.3884 |
| GCYC1A + GCYC1B as foreground (post-duplication) | Two-ratio | 40 | −2896.9931 | 2.8125 | 0.094 | ω0 = 0.4262, ωpost−duplication = 0.2943 |
| GCYC1A, GCYC1B as foreground (lineage-specific) | Three-ratio | 41 | −2895.4048 | 3.1769 | 0.075 | ω0 = 0.4260, ωGCYC1A = 0.2085, ωGCYC1B = 0.3909 |
| BRANCH-SITE MODEL (CLADE MODEL C) | ||||||
| M2a_rel (null) | One-ratio | 42 | −2883.5782 | P1 = 0.3177, ωP1 = 0.0182; P2 = 0.1128, ωP2 = 1.0000; P3 = 0.4789, ωP3 = 0.4789 | ||
| GCYC1A as foreground | Two-ratio | 43 | −2879.2689 | 8.6185 | 0.0033*** | P1 = 0.5190, ω0 = 0.1030, ωGCYC1A = 0.1030; P2 = 0.0473, ω0 = 1.0000, ωGCYC1A = 1.0000; P3 = 0.4338, ω0 = 0.8369, ωGCYC1A = 0.2075 |
| GCYC1A + GCYC1B as foreground (post-duplication) | Two-ratio | 43 | −2880.7007 | 5.755 | 0.016* | P1 = 0.5148, ω0 = 0.1108, ωpost−duplication = 0.1108; P2 = 0.2042, ω0 = 1.0000, ωpost−duplication = 1.0000; P3 = 0.2811, ω0 = 0.9178, ωpost−duplication = 0.1472 |
| GCYC1A, GCYC1B as foreground (lineage-specific) | Three-ratio | 44 | −2879.1924 | 8.7715 | 0.012* | P1 = 0.5081, ω0 = 0.0990, ωGCYC1A = 0.0990, ωGCYC1B = 0.0990; P2 = 0.5081, ω0 = 1.0000, ωGCYC1A = 1.0000, ωGCYC1B = 1.0000; P3 = 0.4241, ω0 = 0.8520, ωGCYC1A = 0.1784, ωGCYC1B = 0.7077 |
a
Models were compared by the double value of the difference in likelihood (2ΔL). Each likelihood ratio was calculated in comparison of alternative models and the suitable null model described in the main text.
b
The statistical significance of models were tested by Likelihood Ratio Test (LRT). P-values were calculated from the 2ΔL of each comparison in the Chi-Square distribution with the degree of freedom equals to difference of parameters (n.p.) between compared models. Significant p-value is highlighted in bold.
*
< 0.05;
**< 0.01;
***
< 0.001.