Performance of specific model features and DRMs for reverse transcriptase.

<p>Data representation in panels <b>A</b>, <b>B</b> and <b>C</b> is the same as in <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi-1002639-g003" target="_blank">Figure 3C, D and E</a>. (<b>D</b>) Recapitulation of reverse transcriptase DRMs by the HIV database and the selective model: A set of 31 literature-documented DRMs of reverse transcriptase <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi.1002639-Castro1" target="_blank">[63]</a>, <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi.1002639-Gotte1" target="_blank">[64]</a>, <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi.1002639-CeccheriniSilberstein1" target="_blank">[65]</a>, <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi.1002639-Ren1" target="_blank">[66]</a>, <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1002639#pcbi.1002639-Deval1" target="_blank">[67]</a> was categorized according to whether or not they are present in the Stanford HIV database (post-drug treatment data) and whether or not they are recapitulated by the selective model. Subscript and superscript numbers list mutation frequencies according to the HIV database and the selective model, respectively.