Hit Expansion from Screening Data Based upon Conditional Probabilities of Activity Derived from SAR Matrices

Abstract

A new methodology for activity prediction of compounds from SAR matrices is introduced that is based upon conditional probabilities of activity. The approach has low computational complexity, is primarily designed for hit expansion from biological screening data, and accurately predicts both active and inactive compounds. Its performance is comparable to state-of-the-art machine learning methods such as support vector machines or Bayesian classification. Matrix-based activity prediction of virtual compounds further extends the spectrum of computational methods for compound design.

Follow-up: Prospective compound design using the ‘SAR Matrix’ method and matrix-derived conditional probabilities of activity

“…Thus, given the low probability of activity, this VC is predicted to be inactive. In benchmark calculations on sets of known active and inactive compounds, conditional probability calculations yielded reasonably accurate predictions of activity, at least comparable to current state-of the-art machine learning approaches 5 .…”

“…SARM calculations generate many virtual compounds (VCs) that populate chemical space around structurally related series. In order to prioritize virtual candidate compounds from SARMs in a target/assay-specific manner, activity prediction methods have been developed including local Quantitative SAR (QSAR) models utilizing compound neighborhood information in SARMs 4 and an approach that derives conditional probabilities of activity from SARMs 5 .…”

The SAR Matrix Method and an Artificially Intelligent Variant for the Identification and Structural Organization of Analog Series, SAR Analysis, and Compound Design

“…Binary activity assignments are typically obtained for compounds from single‐concentration biological screens when a threshold of inhibition or residual activity is applied. The second predictive approach relies on the derivation of conditional probabilities of activity based upon the distribution of individual cores and substituents in pre‐classified active and inactive compounds across SARM ensembles [8] . Conditional probabilities of cores and substituents from screening hits are combined to predict the activity of virtual analogs from SARMs and prioritize candidate compounds for hit expansion [8] .…”

“…[8] Conditional probabilities of cores and substituents from screening hits are combined to predict the activity of virtual analogs from SARMs and prioritize candidate compounds for hit expansion. [8] Given the analog series content of SARMs, this predictive approach is primarily applicable to target-focused screening libraries comprising subsets of structurally related compounds. In test calculations, activity predictions on screening data based on SARM-derived conditional probabilities met or exceeding the performance of machine learning models.…”

“…In test calculations, activity predictions on screening data based on SARM-derived conditional probabilities met or exceeding the performance of machine learning models. [8] Furthermore, the approach was successfully applied prospectively to predict novel active compounds for cancer targets on the basis of cell-based screens. [9] In this case, focused libraries consisting of secondary structure mimetic compounds were screened in cell-based anticancer assays and subsequently analyzed in SARMs, which yielded virtual analogs.…”