Table 4.

Accuracy of data sets generated along a 24 taxa tree (JC and HKY models).

Accuracy of ML inferred trees from data sets of 1,000 bp generated along a 24 taxa tree and under a growing percentage of editing positions. Background sites evolved according both JC and HKY models [17,18]. Results from multiple alignments without editing sites are also shown as control.

24 taxa tree - JC model
% editing	0	1	3	5	7	10
1,000 nucleotides
Genomic	1	0.92	0.86	0.88*	0.80*	0.79*
cDNA	1	0.88	0.82	0.77	0.64	0.58
Control	1	0.95	0.88	0.91	0.90	0.86
24 taxa tree – HKY model
% editing	0	1	3	5	7	10
1,000 nucleotides
Genomic	1	0.94	0.94	0.91*	0.90*	0.90*
cDNA	1	0.93	0.92	0.83	0.78	0.73
Control	1	0.94	0.90	0.87	0.81	0.70

* indicates that the comparison between genomic and cDNA accuracy is significant (Pχ²₁< 0.05)