Estrogen receptors (ERα) are a critical target for drug design as well as a potential source of toxicity when activated unintentionally. Thus, evaluating potential ERα binding agents is critical in both drug discovery and chemical toxicity areas. Using computational tools, e.g., Quantitative Structure-Activity Relationship (QSAR) models, can predict potential ERα binding agents before chemical synthesis. The purpose of this project was to develop enhanced predictive models of ERα binding agents by utilizing advanced cheminformatics tools that can integrate publicly available bioassay data. The initial ERα binding agent data set, consisting of 446 binders and 8307 non-binders, was obtained from the Tox21 Challenge project organized by the NIH Chemical Genomics Center (NCGC). After removing the duplicates and inorganic compounds, this data set was used to create a training set (259 binders and 259 non-binders). This training set was used to develop QSAR models using chemical descriptors. The resulting models were then used to predict the binding activity of 264 external compounds, which were available to us after the models were developed. The cross-validation results of training set [Correct Classification Rate (CCR) = 0.72] were much higher than the external predictivity of the unknown compounds (CCR = 0.59). To improve the conventional QSAR models, all compounds in the training set were used to search PubChem and generate a profile of their biological responses across thousands of bioassays. The most important bioassays were prioritized to generate a similarity index that was used to calculate the biosimilarity score between each two compounds. The nearest neighbors for each compound within the set were then identified and its ERα binding potential was predicted by its nearest neighbors in the training set. The hybrid model performance (CCR = 0.94 for cross validation; CCR = 0.68 for external prediction) showed significant improvement over the original QSAR models, particularly for the activity cliffs that induce prediction errors. The results of this study indicate that the response profile of chemicals from public data provides useful information for modeling and evaluation purposes. The public big data resources should be considered along with chemical structure information when predicting new compounds, such as unknown ERα binding agents.
The hybrid modeling workflow.
Estrogen receptors are cellular proteins that are activated when bound to estrogen molecules. When activated, estrogen receptors trigger the expression of gene products crucial to the endocrine system. These receptors can also be activated by certain endocrine disrupting chemicals (EDC), resulting in a disruption of normal estrogen signaling. There are two unique estrogen receptors: ERα and ERβ. These two receptors are highly similar in the DNA binding domain, but differ more significantly in other regions. While there are many EDC that interact with both receptors, the difference between these two receptors allows some ligands specifically bind to only one receptor as well. Among all known binding agents, the ERα binders are much better characterized than ERβ binders. Due to the nature of available data, this study focuses solely on ligands binding to ERα.
The performance of all resulting models
(A) Cross-validation of the 518 training set compounds; (B) external validation of 264 unknown compounds.
When estrogen receptors are activated by small molecules other than estrogens, the expression of the associated genes is deregulated leading to neurological, developmental, and reproductive toxicity. There are many small molecules with different chemical structures which exhibit interaction with the ligand binding domain of the estrogen receptor.
Considering the large number of compounds which needs to be evaluated for their estrogen receptor binding potentials, traditional experimental toxicology protocols can be costly and time-consuming. As a result, there is a strong need to effectively pre-screen and prioritize small molecules for potential endocrine disruption prior to more costly animal testing. In a 2007 publication, the U.S. National Research Council identified both high-throughput screening (HTS) and computational models as critical chemical toxicity evaluation tools in Twenty-First century toxicology. HTS has been viewed as a potential alternative to animal models due to the ability to test many molecules at a rapid pace and lower cost. The large number of HTS studies has resulted in publically available bioassay databases which are a rich source of in vitro data.Motivated by these available data, computational modeling, which costs even less than HTS, has been used as another important evaluation protocols for EDCs.
Quantitative structure-activity relationship (QSAR) modeling has been applied to develop estrogen receptor binding models in the past decade. These studies have covered a wide range of modeling approaches and data set sizes, from a descriptor-based decision tree to 3-D docking and multi-dimensional QSAR. The number of compounds used for modeling purpose in these studies range from less than 100 to more than 8000. The QSAR modeling of estrogen receptor binding agents has also been reviewed.
Ribay, K., Kim, M. T., Wang, W., Pinolini, D., & Zhu, H. (2016). Predictive Modeling of Estrogen Receptor Binding Agents Using Advanced Cheminformatics Tools and Massive Public Data. Frontiers in environmental science, 4, 12. https://doi.org/10.3389/fenvs.2016.00012