Aryl hydrocarbon receptor (AHR) modulators library

The library of Aryl Hydrocarbon Receptor (AhR) modulators represents a sophisticated collection of compounds designed to interact with a critical regulatory protein involved in various physiological and pathological processes. The AhR is a ligand-activated transcription factor that plays a significant role in the body's response to environmental toxins, aiding in the metabolism of xenobiotics. Beyond its role in detoxification, AhR is increasingly recognized for its involvement in immune system regulation, cell proliferation, and differentiation, making it a promising target for therapeutic intervention in cancer, immune disorders, and other diseases.
This unique library was assembled using standard computational techniques, including filtration by physicochemical properties, selection by fingerprint similarity, molecular docking, and machine learning algorithms. Our AhR Library comprises 16,846 compounds and three different HTVS strategies:
 

1. Physicochemical Filters with Tanimoto Similarity Selection: The initial step involves applying physicochemical filters based on the distribution of known AhR modulators from the Tox21 dataset. This filtering process selects compounds that share desirable physicochemical properties with known modulators, ensuring that only the most relevant molecules are considered for further analysis. Subsequently, the Tanimoto Similarity measure, using Extended Connectivity Fingerprints 6 (ECFP6), is employed to select compounds. This similarity metric compares the structural fingerprints of the compounds, prioritizing those that are most similar to known AhR modulators. This method ensures that selected compounds have a higher likelihood of exhibiting AhR modulatory activity due to their structural resemblance to known active compounds.
2. Physicochemical Filters with HTVS Docking: This strategy also begins with the application of the same physicochemical filters as in strategy 1 to narrow down the pool of potential candidates based on their similarity to known AhR modulators in the Tox21 dataset. Following this initial selection, High-Throughput Virtual Screening (HTVS) docking is performed. This computational docking process simulates the interaction between each compound and the AhR protein, predicting their binding affinity. Compounds with higher predicted affinity are considered more promising as potential modulators. This approach not only considers the structural similarity but also the potential for effective interaction with the AhR, providing a more targeted selection of candidates.
3. Machine Learning Model on Tox21 AhR Modulators Data: This strategy leverages the power of machine learning to analyze the dataset of AhR modulators from Tox21. A predictive model is built using this dataset, trained to identify patterns and features that are indicative of successful AhR modulation. Once the model is trained, it is applied across the entire Screening Compounds library from ChemDiv to predict potential AhR modulators. This approach can uncover novel compounds that may not have been identified through traditional similarity and docking methods, as the machine learning model can recognize complex patterns in the data that are less apparent to other methods.