Advances in exploring activity cliffs

Abstract

The activity cliff (AC) concept is of comparable relevance for medicinal chemistry and chemoinformatics. An AC is defined as a pair of structurally similar compounds with a large potency difference against a given target. In medicinal chemistry, ACs are of interest because they reveal small chemical changes with large potency effects, a concept referred to as structure-activity relationship (SAR) discontinuity. Computationally, ACs can be systematically identified, going far beyond individual compound series considered during lead optimization. Large-scale analysis of ACs has revealed characteristic features across many different compound activity classes. The way in which the molecular similarity and potency difference criteria have been addressed for defining ACs distinguishes between different generations of ACs and mirrors the evolution of the AC concept. We discuss different stages of this evolutionary path and highlight recent advances in AC research.

Keywords: Activity cliff concept; Activity data analysis; Cliff categories; Compound potency differences; Molecular similarity; Structure–activity relationships.

Fig. 1

Exemplary activity cliffs. On the left, a MACCS-based AC is shown (Tc 0.86). Fingerprint-based ACs are first generation ACs. In the center, an MMP-cliff and a retrosynthetic version (RMMP-cliff) applying an activity class-dependent potency difference threshold are depicted (exemplary second generation ACs). On the right, an analog series-based AC with class-dependent potency difference threshold is shown (third generation AC). Further details are provided in the text. For all compounds, potency (pK_i) values are reported and structural differences are highlighted in red. From the left to the right, AC targets were the histamine H4, adenosine A1, and adenosine A2a receptor, respectively

Fig. 2

Activity class-dependent potency difference thresholds. The compound potency distribution for neurokinin 1 receptor ligands (top left) is represented in a boxplot (center) and the interquartile range (IQR) is determined (right). On the basis of the IQR, activity classes are assigned to different categories (IQR < 1; CAT 2: 1 ≤ IQR < 2; CAT 3: IQR ≥ 2) and only classes of CAT 2 or 3 are subjected to AC analysis. At the bottom, the corresponding potency difference distribution of RMMPs is displayed (left). From the mean and standard deviation (σ) of the distribution (center), the activity class-dependent potency difference threshold for AC formation is calculated as the mean plus two σ (ΔpK_i = 1.4) (right)

Fig. 3

Three-dimensional activity cliffs. Shown are three exemplary 3D-cliffs where ligands are distinguished by different types of interactions. Bound conformations of highly and weakly potent cliff compounds are colored green and red, respectively. In addition, an exemplary off-pocket cliff according to reference 28 is shown at the bottom

Fig. 4

Activity cliff networks. For two activity classes, RMMP-cliff networks are shown. Nodes represent compounds and edges pairwise ACs. Highly and weakly potent cliff partners are colored green and red, respectively. Networks on the left and right were generated applying are constant potency difference threshold (ΔpK_i ≥ 2) and an activity class-dependent threshold (as specified), respectively. In each case, the number of ACs, participating compounds, and AC clusters are reported. Clusters containing coordinated ACs are distinguished from clusters formed by isolated ACs

Fig. 5

Dual-site activity cliffs and single-site analogs. a Shown is an exemplary ds-AC together with both single-site analogs displaying a compensatory potency effect on AC formation. Structural modifications at different sites are colored orange and blue, respectively, and pK_i values are reported in circles. b Shown is an exemplary ds-AC with two structural isomers (connected with the highly and weakly potent ds-AC compound through dashed green and red arrows, respectively). Structural modifications are highlighted in orange and pK_i are reported in circles

Fig. 6

Isomer/MMP-cliffs. Shown is an exemplary isomer/MMP-cliff with activity against the prostanoid EP4 receptor. A structural isomer of the weakly potent MMP-cliff compound was identified forming an additional AC with the highly potent cliff compound

Fig. 7

Activity cliffs with privileged substructures. Different ACs containing PSs are shown. Substituents and the PS are colored in orange and pink, respectively. For each AC, the target name and potency values are provided