Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Using the nucleotide substitution rate matrix to detect horizontal gene transfer

Figure 3

Average root mean square (RMS) error per matrix element when estimating the rate matrix Q from simulated sequence data. Curves are shown for three different sequence lengths, N = 100, 1000, and 10000. Note that the increase in the error for higher percent sequence identity is a sampling effect: more similar sequences have fewer sites of change, and therefore the P ^ MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaacuWGqbaugaqcaaaa@2DE5@ determined from the data have higher sampling error.

Back to article page