Comparison of classification performance between the method presented in this paper and the classification approach of Burgos-Artizzu et al. [23].

<p>The results are the average across all the mouse pairs of Dataset A, computed employing all the metrics of <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0074557#pone-0074557-t002" target="_blank">Table 2</a> and <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0074557#pone-0074557-t003" target="_blank">3</a>. In all instances our approach outperforms <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0074557#pone.0074557-BurgosArtizzu1" target="_blank">[23]</a>. The full results are available as additional material (<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0074557#pone.0074557.s001" target="_blank">Table S1</a>).