Non-parametric synergy modeling of chemical compounds with Gaussian processes

Abstract

Background Understanding the synergetic and antagonistic effects of combinations of drugs and toxins is vital for many applications, including treatment of multifactorial diseases and ecotoxicological monitoring. Synergy is usually assessed by comparing the response of drug combinations to a predicted non-interactive response from reference (null) models. Possible choices of null models are Loewe additivity, Bliss independence and the recently rediscovered Hand model. A different approach is taken by the MuSyC model, which directly fits a generalization of the Hill model to the data. All of these models, however, fit the dose–response relationship with a parametric model. Results We propose the Hand-GP model, a non-parametric model based on the combination of the Hand model with Gaussian processes. We introduce a new logarithmic squared exponential kernel for the Gaussian process which captures the logarithmic dependence of response on dose. From the monotherapeutic response and the Hand principle, we construct a null reference response and synergy is assessed from the difference between this null reference and the Gaussian process fitted response. Statistical significance of the difference is assessed from the confidence intervals of the Gaussian process fits. We evaluate performance of our model on a simulated data set from Greco, two simulated data sets of our own design and two benchmark data sets from Chou and Talalay. We compare the Hand-GP model to standard synergy models and show that our model performs better on these data sets. We also compare our model to the MuSyC model as an example of a recent method on these five data sets and on two-drug combination screens: Mott et al. anti-malarial screen and O’Neil et al. anti-cancer screen. We identify cases in which the HandGP model is preferred and cases in which the MuSyC model is preferred. Conclusion The Hand-GP model is a flexible model to capture synergy. Its non-parametric and probabilistic nature allows it to model a wide variety of response patterns.

Toxicogenomic Fingerprint Identification in Springtails to Assess Pesticide-Contaminated Soils

“…We used the HandGP model (Shapovalova et al, 2022) for the analysis of interaction effects between imidacloprid and clothianidin, or imidacloprid and cyproconazole.…”

“…For consistency check, we also analyze the data using the MuSyC model, which is based on a parametric model over the whole dose-response surface and thus the closest parametric alternative to the HandGP approach (Shapovalova et al, 2022;Wooten et al, 2021) 3…”

“…Therefore, we here tested the effects of two binary pesticide mixtures on the reproduction of F. candida: (1) the neonicotinoid pesticides imidacloprid and clothianidin, and (2) imidacloprid and the fungicide cyproconazole. We analyzed the phenotypic data with recently developed Hand GP models (Shapovalova et al, 2022) to assess mixture interaction effects, such as synergism or antagonism, on springtail reproduction. (Wooten et al, 2021), which gave comparative results (see Table S3.1).…”

“…GP models, therefore, address all three challenges associated with DGEA of complex mixture exposure data. Previously, GP models have been applied to DGEA (Kalaitzis & Lawrence, 2011), drug-drug interaction modeling (Shapovalova et al, 2022), and pharmacological responses and biomarkers development (Wang et al, 2020), but to date not yet for mixture toxicity data. Besides the advantages mentioned above, our approach can consider that various experiment can have the same chemical compound and treat the single experimental data of the same chemical compound jointly across all the experiments.…”

“…We selected two binary mixtures of either two neonicotinoids, imidacloprid and clothianidin, or imidacloprid and cyproconazole (an azole fungicide). First, we quantified the interaction toxicity effects of the mixtures on F. candida reproduction by Hand GP models (Shapovalova et al, 2022). By studying the interaction effects on mixture toxicity to reproduction, we demonstrate that the pesticide exposure caused genetic interaction effects as non-additive toxicity stems from molecular interaction effects and, hence, gene expression.…”