Skip to main content

Table 2 Summary of generalized linear model results. Each row corresponds to a single model using multiple explanatory variables. Significant values are indicated in bold. Quasibinomial and negative-binomial error distributions were used and results are given after calculating type-II analysis-of-variance using a F test (Quasibinomial family) or a likelihood ratio test (negative-binomial family).

From: Amplified fragment length homoplasy: in silico analysis for model and non-model species

   Explanatory variables
   Upstream parameters AFLP parameters AFLP diagnostics
   Species No. of sba CG cont No. of peaks per profile Peak length Homoplasy per peak
Response variables Distribution (Df = 2) (Df = 1) (Df = 2) (Df = 1) (Df = 1) (Df = 1)
In silico analyses        
   Peak length distributionb Quasibinomial F1,1349 = 2.09e-12,
P = 1
    F1,1349 = 967.20,
P < 0.001
 
   No. of peaks per profile Negative binomial χ2 = 1574.87,
P < 0.001
χ2 = 1869.47,
P < 0.001
χ2 = 275.24,
P < 0.001
   
   Homoplasy rate H Quasibinomial F2,278 = 0.98,
P = 0.38
  F2,278 = 0.72,
P = 0.49
F1,278 = 576.26,
P < 0.001
  
   No. of CFc Negative binomial χ2 = 19.90,
P < 0.001
  χ2 = 2.99,
P = 0.22
χ2 = 25.32,
P < 0.001
χ2 = 37.92,
P < 0.001
 
   No. of CFc in peaks with more than 10 fragments Negative binomial χ2 = 6.96,
P = 0.03
  χ2 = 0.90,
P = 0.64
χ2 = 0.87,
P = 0.35
χ2 = 2.74,
P = 0.10
 
Empirical analyses        
   Fluorescence intensity Negative binomial      χ2 = 22.33,
P < 0.001
χ2 = 2.39,
P = 0.12
  1. aTotal number of selective bases added for the 2 primers.
  2. bPeak length distribution is expressed in relative frequencies.
  3. c"CF" corresponds to comigrating fragments within a peak.