Skip to main content


Figure 12 | Journal of Cheminformatics

Figure 12

From: Are phylogenetic trees suitable for chemogenomics analyses of bioactivity data sets: the importance of shared active compounds and choosing a suitable data embedding method, as exemplified on Kinases

Figure 12

Two different types of outlier trends, which are likely to explain the formation of 2 clusters of outliers. VEGFR3, ACK1 and LYN show consistently high SAR similarity with other kinases at both low and high distances, with lower SAR similarity against some kinases at high distances (see graphs below). CSK21, CSK22, IGF1R and WNK2 show significantly higher SAR similarity with other kinases at low distances than at higher distances, but with very high variance of the data points: in many cases, neighboring kinases show low SAR similarity or distant kinases show high SAR similarity (see graphs above). Regardless, 2 different types of outlier trends were observed, possibly explaining the grouping of the outliers in 2 different clusters in the MDS plot.

Back to article page