Skip to main content

Table 2 Log-likelihood, AIC and parameter estimates under random-site models for flaA sequences

From: Divergent evolution and purifying selection of the flaA gene sequences in Aeromonas

Model

p a

ωb

c

AICd

κe

Estimates of parameters

Positively selected sites

M0: one ratio

1

0.062 ± 0.000

-6830.350

13662.700

2.093

ω = 0.062

None

M1a: nearly neutral

2

0.114 ± 0.015

-6689.544

13383.089

2.192

p0 = 0.92 (p1 = 0.08)

ω = 0.034, ω = 1

Not allowed

M2a: positive selection

4

0.114 ± 0.015

-6689.544

13387.092

2.192

p0 = 0.918, p1 = 0.082,

(p2 = 0.000), ω0 = 0.034,

ω1 = 1; ω2 = 58.267

179 I (at P = 0.747)

M3: discrete (K = 3)

5

0.071 ± 0.008

-6601.883

13213.766

2.085

p0 = 0.675, p1 = 0.254

(p2 = 0.071), ω0 = 0.003

ω1 = 0.122, ω2 = 0.546

None

M7: beta

2

0.071 ± 0.009

-6603.940

13211.879

2.087

p = 0.152, q = 1.683

Not allowed

M8: beta & ω > 1

4

0.072 ± 0.008

-6600.818

13209.635

2.082

p0 = 0.996, (p1 = 0.004)

p = 0.160, q = 1.976

ω = 1.92

179 I (at P = 0.967)

M8a: beta & ωs = 1

4

0.072 ± 0.008

-6601.490

13210.978

2.086

p0 = 0.991, (p1 = 0.009)

p = 0.165, q = 2.190

ω = 1.0

Not allowed

  1. a p, number of free parameters for each model
  2. b ω, averaged ω ± standard error of dN/dS ratio over all sites in the flaA gene alignment
  3. c ℓ, log-likelihood value for each model
  4. d AIC, Akaike's Information Criterion value
  5. e κ, transition/transversion rate [27]
  6. Values of ω > 1 and sites inferred to be under positive selection are in bold