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

Abstract

Background‘Phylogenetic trees’ are commonly used for the analysis of chemogenomics datasets and to relate protein targets to each other, based on the (shared) bioactivities of their ligands. However, no real assessment as to the suitability of this representation has been performed yet in this area. We aimed to address this shortcoming in the current work, as exemplified by a kinase data set, given the importance of kinases in many diseases as well as the availability of large-scale datasets for analysis. In this work, we analyzed a dataset comprising 157 compounds, which have been tested at concentrations of 1 μM and 10 μM against a panel of 225 human protein kinases in full-matrix experiments, aiming to explain kinase promiscuity and selectivity against inhibitors. Compounds were described by chemical features, which were used to represent kinases (i.e. each kinase had an active set of features and an inactive set).ResultsUsing this representation, a bioactivity-based classification was made of the kinome, which partially resembles previous sequence-based classifications, where particularly kinases from the TK, CDK, CLK and AGC branches cluster together. However, we were also able to show that in approximately 57% of cases, on average 6 kinase inhibitors exhibit activity against kinases which are located at a large distance in the sequence-based classification (at a relative distance of 0.6 – 0.8 on a scale from 0 to 1), but are correctly located closer to each other in our bioactivity-based tree (distance 0 – 0.4). Despite this improvement on sequence-based classification, also the bioactivity-based classification needed further attention: for approximately 80% of all analyzed kinases, kinases classified as neighbors according to the bioactivity-based classification also show high SAR similarity (i.e. a high fraction of shared active compounds and therefore, interaction with similar inhibitors). However, in the remaining ~20% of cases a clear relationship between kinase bioactivity profile similarity and shared active compounds could not be established, which is in agreement with previously published atypical SAR (such as for LCK, FGFR1, AKT2, DAPK1, TGFR1, MK12 and AKT1).ConclusionsIn this work we were hence able to show that (1) targets (here kinases) with few shared activities are difficult to establish neighborhood relationships for, and (2) phylogenetic tree representations make implicit assumptions (i.e. that neighboring kinases exhibit similar interaction profiles with inhibitors) that are not always suitable for analyses of bioactivity space. While both points have been implicitly alluded to before, this is to the information of the authors the first study that explores both points on a comprehensive basis. Excluding kinases with few shared activities improved the situation greatly (the percentage of kinases for which no neighborhood relationship could be established dropped from 20% to only 4%). We can conclude that all of the above findings need to be taken into account when performing chemogenomics ...

What Makes a Kinase Promiscuous for Inhibitors?

“…Several screens of kinase inhibitor collections against large panels of kinases have been conducted in an effort to characterize the selectivity of kinase inhibitors and to determine which fraction of the kinome can be inhibited with existing inhibitors (Drewry et al, 2017;Fabian et al, 2005;Karaman et al, 2008;Klaeger et al, 2017). Hierarchical clustering has been performed on these datasets to identify patterns among ligands (Paricharak et al, 2013). Here, we perform hierarchical cluster analysis on a recently published dataset of 645 small-molecule inhibitors and 406 human kinases (392 wild-type kinases and 14 variants) (Drewry et al, 2017) to identify the relationships among kinases based on their binding phenotype to inhibitors.…”

Orthologue chemical space and its influence on target prediction

“…Other work mined Homo sapiens bioactivity data together with structural and historical assay space searches, to propose cross-species targets alongside respective hit compounds for the treatment of M.tuberculosis ( Martínez-Jiménez et al , 2013 ). One study comprised generation of phylogenetic and bioactivity tree representations of kinases, highlighting clustering targets in protein structure space makes incorrect assumptions about interactions in bioactivity space ( Paricharak et al , 2013 ). In fact, many factors need to be considered when performing homology-based bioactivity inference between kinase targets, illustrating the potential pitfalls of extrapolating ligand interactions between species.…”

Protein homology reveals new targets for bioactive small molecules

“…As such, information about protein orthology relationships to make predictions in less-studied species or improve predictions in model organisms by integrating data from other species has been mostly disregarded. For instance, only a handful of recent studies investigated the properties of ligands binding to related proteins in distinct organisms (Klabunde, 2007;Krü ger and Overington, 2012;Paricharak et al, 2013) or within the same organism (Schuffenhauer et al, 2003). This is in stark contrast with many areas of biology and bioinformatics where the ability to transfer results obtained in one organism to others is a central dogma.…”