Novel Computational Approach to Predict Off-Target Interactions for Small Molecules

Abstract

Most small molecule drugs interact with unintended, often unknown, biological targets and these off-target interactions may lead to both preclinical and clinical toxic events. Undesired off-target interactions are often not detected using current drug discovery assays, such as experimental polypharmacological screens. Thus, improvement in the early identification of off-target interactions represents an opportunity to reduce safety-related attrition rates during preclinical and clinical development. In order to better identify potential off-target interactions that could be linked to predictable safety issues, a novel computational approach to predict safety-relevant interactions currently not covered was designed and evaluated. These analyses, termed Off-Target Safety Assessment (OTSA), cover more than 7,000 targets (~35% of the proteome) and > 2,46,704 preclinical and clinical alerts (as of January 20, 2019). The approach described herein exploits a highly curated training set of >1 million compounds (tracking >20 million compound-structure activity relationship/SAR data points) with known in vitro activities derived from patents, journals, and publicly available databases. This computational process was used to predict both the primary and secondary pharmacological activities for a selection of 857 diverse small molecule drugs for which extensive secondary pharmacology data are readily available (456 discontinued and 401 FDA approved). The OTSA process predicted a total of 7,990 interactions for these 857 molecules. Of these, 3,923 and 4,067 possible high-scoring interactions were predicted for the discontinued and approved drugs, respectively, translating to an average of 9.3 interactions per drug. The OTSA process correctly identified the known pharmacological targets for >70% of these drugs, but also predicted a significant number of off-targets that may provide additional insight into observed in vivo effects. About 51.5% (2,025) and 22% (900) of these predicted high-scoring interactions have not previously been reported for the discontinued and approved drugs, respectively, and these may have a potential for repurposing efforts. Moreover, for both drug categories, higher promiscuity was observed for compounds with a MW range of 300 to 500, TPSA of ~200, and clogP ≥7. This computation also revealed significantly lower promiscuity (i.e., number of confirmed off-targets) for compounds with MW > 700 and MW<200 for both categories. In addition, 15 internal small molecules with known off-target interactions were evaluated. For these compounds, the OTSA framework not only captured about 56.8% of in vitro confirmed off-target interactions, but also identified the right pharmacological targets for 14 compounds as one of the top scoring targets. In conclusion, the OTSA process demonstrates good predictive performance characteristics and represents an additional tool with utility during the lead optimization stage of the drug discovery process. Additionally, the computed physiochemical properties such as clogP (i.e., lipophilicity), molecular weight, pKa and logS (i.e., solubility) were found to be statistically different between the approved and discontinued drugs, but the internal compounds were close to the approved drugs space in most part.

Keywords: Off-targets; gene expression; machine learning; pocket search; secondary pharmacology; toxicology.

Figure 1

Outline of the OTSA computational screening of small molecule drugs using multiple computational models. By assessing drugs and metabolites with multiple tools simultaneously, safety scientists can screen in silico beyond current off-target binding and kinase screening panels. Both 2-D (six methods) and 3-D (3Decicion) methods are used to produce a list of potential off-target interactions. The predicted targets with computed scores >0.6 and above in at least 3 out of the 6 prediction methods will be moved to Step 3. These are then compared with the “body atlas” GTEX (human) and internal RNA-Seq data from untreated species (rat, monkey and mouse) to predict potential target tissues (Step 3). In addition, outcome prediction tools are used to predict the consequences of predicted interactions (Step 4). In step 5, an OTSA comprehensive off-target pharmacology report is generated. Off-target interactions considered to be of potential consequence are verified and evaluated using appropriate in vitro or in vivo models. In the current manuscript, for the approved and discontinued drugs as well as our 15 internal compounds, only steps 1 and 2 were used because of throughput limitations. Steps 3 to 5 are used to contextualize the off-target prediction and predict potential target tissues as well as biological outcomes.

Figure 2

t-SNE method derived chemical diversity distributions for 857 drugs (401 approved and 456 discontinued): 25 clusters are shown. Only the relative distance between molecules in 2-D projection (each spheres) is meaningful. X (Dim1) and Y (Dim2) are the projections in 2-dimensions from a multidimensional descriptor space. The discontinued and approved drugs are shown in filled circle and diamond shapes, respectively.

Figure 3

(A) Plot showing the clogP vs. TPSA distribution for approved (blue) and discontinued drugs (red). (B) Plot showing the clogP vs. MW distribution for approved (blue) and discontinued drugs (red). (C) Distribution of pKa values of the discontinued and approved drugs. Computed pKa values are shown on the X-axis. The Y-axis shows the number of compounds in each pKa bin. Red and blue color bars indicate discontinued and approved drugs, respectively.

Figure 4

Summary of predicted high-scoring interactions for the approved (n = 401) and discontinued (n = 457) drugs. The total predicted off-targets for each category are shown in brown. The in vitro confirmed interactions are in blue, while the novel yet to be confirmed interactions are in red.

Figure 5

(A,B) Comparison of MW and TPSA with number of predicted off-targets for the approved and discontinued drugs. The X-axis shows (A) MW and (B) TPSA. In both figures, the Y-axis shows the percentage of predicted off-targets in each of the bins.