Abstract
In silico protein target deconvolution is frequently used for mechanism-of-action investigations; however existing protocols usually do not predict compound functional effects, such as activation or inhibition, upon binding to their protein counterparts. This study is hence concerned with including functional effects in target prediction. To this end, we assimilated a bioactivity training set for 332 targets, comprising 817,239 active data points with unknown functional effect (binding data) and 20,761,260 inactive compounds, along with 226,045 activating and 1,032,439 inhibiting data points from functional screens. Chemical space analysis of the data first showed some separation between compound sets (binding and inhibiting compounds were more similar to each other than both binding and activating or activating and inhibiting compounds), providing a rationale for implementing functional prediction models. We employed three different architectures to predict functional response, ranging from simplistic random forest models (‘Arch1’) to cascaded models which use separate binding and functional effect classification steps (‘Arch2’ and ‘Arch3’), differing in the way training sets were generated. Fivefold stratified cross-validation outlined cascading predictions provides superior precision and recall based on an internal test set. We next prospectively validated the architectures using a temporal set of 153,467 of in-house data points (after a 4-month interim from initial data extraction). Results outlined Arch3 performed with the highest target class averaged precision and recall scores of 71% and 53%, which we attribute to the use of inactive background sets. Distance-based applicability domain (AD) analysis outlined that Arch3 provides superior extrapolation into novel areas of chemical space, and thus based on the results presented here, propose as the most suitable architecture for the functional effect prediction of small molecules. We finally conclude including functional effects could provide vital insight in future studies, to annotate cases of unanticipated functional changeover, as outlined by our CHRM1 case study.
Highlights
Target deconvolution is an important step in the subsequent analysis of data gleaned from phenotypic screenings, to identify the modulated targets of active compounds and enable the continued dissection of the biological processes involved in a system of interest (Terstappen et al, 2007; Raida, 2011; Kotz, 2012; Lee and Bogyo, 2013)
Imposing only two functional labels may be an over-simplification, this is preferred to the complex situation resulting from the original complex BioAssay Ontology (BAO) labeling, since it reduces training data into a binary problem per protein target, ensures larger numbers of compounds are retrained within each MOA, and that generated predictions are compared between the complete spectra of functional predictions between targets
Changes in certain moieties are shown to affect binding outcome more than others; for example, one study highlighted that steric modifications near a basic nitrogen, methylation of indoles, and aniline nitrogen substitutions appeared to play important roles in determining functional activity while keeping overall structure rather similar (Dosa and Amin, 2016)
Summary
Target deconvolution is an important step in the subsequent analysis of data gleaned from phenotypic screenings, to identify the modulated targets of active compounds and enable the continued dissection of the biological processes involved in a system of interest (Terstappen et al, 2007; Raida, 2011; Kotz, 2012; Lee and Bogyo, 2013). The characterization of the functional effects of compounds is often a principle shortcoming for current in silico methods, since many protocols only provide probability for compound affinity at a target (Drakakis et al, 2013; Koutsoukas et al, 2013; Mervin et al, 2015) Existing protocols, such as the Similarity Ensemble Approach (SEA) (Keiser et al, 2007) and Prediction of Activity Spectra for Substances (PASS) (Lagunin et al, 2000), provide functional annotation by training on a compound set extracted from the MDL Drug Data Report [MDDR] (2006). The cross validation and time-split performance of the approaches has provided guidance into the choice of architecture to be deployed in-house for future triage processes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
