CORONAVIRUS Library

Description

The ChemDiv’s coronavirus library contains 21,145 small molecule compounds

The outbreak of coronavirus disease (COVID-19), first reported in Wuhan, China, on December 31, 2019, has starkly highlighted the critical need for effective drugs against coronaviruses. These viruses, belonging to the large CoV family, are capable of causing a spectrum of illnesses ranging from mild conditions like the common cold to severe, life-threatening diseases such as Middle East Respiratory Syndrome (MERS-CoV) and Severe Acute Respiratory Syndrome (SARS-CoV). The emergence of a novel coronavirus (nCoV), which had not been previously identified in humans, underscores the evolving and unpredictable nature of these pathogens.

The zoonotic origin of coronaviruses, which enables their transmission between animals and humans, poses a unique challenge to global health. Investigations have revealed that SARS-CoV was transmitted from civet cats to humans, and MERS-CoV from dromedary camels to humans. This zoonotic nature raises concerns about the potential for other coronaviruses, currently circulating in animal populations, to cross species barriers and infect humans.

Given the rapid global spread and the severe impact of COVID-19, coupled with the history of other deadly coronavirus outbreaks, the development of effective drugs against this family of viruses is of great importance. Such drugs would not only be crucial in managing current and potential future outbreaks but are also essential for global preparedness against emerging coronavirus strains. The ability to quickly develop and deploy effective treatments against novel coronaviruses is crucial for mitigating the potentially devastating health, social, and economic impacts of these outbreaks.

The ChemDiv virtual screening methodology has been specifically designed to optimize the discovery of small molecule compounds in drug discovery and implemented in the creation of this Coronavirus library. This advanced methodology encompasses several key components, each contributing to a highly effective and efficient screening process:

1. Target Identification: Utilizing Swiss-Prot Protein Targets and PDB X-Ray Structure Search, the library focuses on significant targets such as ACE2, PLpro, and 3CLP, which are crucial in various biological pathways and disease mechanisms.

1. Comprehensive Training Sets: The library leverages extensive training sets from ChEMBL 25 and PubMed databases, as well as Current Patent Literature (CAS, Integrity), ensuring a broad and deep representation of current knowledge and trends in drug discovery.

1. Machine Learning Data Curation:
1. Employs tools like KNIME/RDKit and kNN classifiers for precise data analysis and curation. It utilizes Distance in BitVector Cosine Space and FCFP12 (10,240 bit) fingerprints for in-depth compound profiling.
2. Integrates a Hybrid 2D QSAR/Fingerprint Model using Kernel Chemical Classification/Regression (kcc), enhancing the predictive accuracy of the compound's biological activity.
2. 3D Shape Similarity Virtual Screening: The library applies APF®MolSoft and references research from Lam et al. in 2018 and 2019 [1-3] and Totrov in 2008 [4] to analyze and utilize 3D shape similarities among compounds, a critical factor in predicting molecular interactions and functionality.
3. Structure-Based Docking and Screening:
1. Employs Multiple Receptor Conformation (MRC) 4D Docking using ICM-Pro MolSoft, informed by Bottegoni et al. work in 2009 [5] .
2. Utilizes Ligand-Biased Ensemble receptor Docking (LigBEnD) with ICM-Pro MolSoft, drawing on methodologies from Lam et al. in 2018 [1].
4. Quality and Safety Filters: Incorporates REOS, MedChem, and PAINS filters to eliminate compounds with reactive, toxic, promiscuous, or other undesirable motifs, ensuring the selection of high-quality, safe molecules.
5. Diversity Picking: Uses the RDKit implementation of the MaxMin algorithm [6] to ensure a diverse selection of compounds, enabling the exploration of a wide chemical space.

This sophisticated approach, combining state-of-the-art technologies and methodologies, positions the ChemDiv virtual screening methodology as a powerful tool in the search for novel and effective drug candidates. Its focus on quality, diversity, and cutting-edge analytical techniques makes it an invaluable resource for drug discovery efforts.

Publications

[1] Lam PC, Abagyan R, Totrov M. Hybrid receptor structure/ligand-based docking and activity prediction in ICM: development and evaluation in D3R Grand Challenge 3. J Comput Aided Mol Des. 2019 Jan;33(1):35-46. doi: 10.1007/s10822-018-0139-5

[2] Lam PC, Abagyan R, Totrov M. Macrocycle modeling in ICM: benchmarking and evaluation in D3R Grand Challenge 4. J Comput Aided Mol Des. 2019 Dec;33(12):1057-1069. doi: 10.1007/s10822-019-00225-9.

[3] Lam PC, Abagyan R, Totrov M. Ligand-biased ensemble receptor docking (LigBEnD): a hybrid ligand/receptor structure-based approach. J Comput Aided Mol Des. 2018 Jan;32(1):187-198. doi: 10.1007/s10822-017-0058-x.

[4] Totrov M. Atomic property fields: generalized 3D pharmacophoric potential for automated ligand superposition, pharmacophore elucidation and 3D QSAR. Chem Biol Drug Des. 2008. Jan;71(1):15-27. doi: 10.1111/j.1747-0285.2007.00605.x.

[5] Bottegoni G, Kufareva I, Totrov M, Abagyan R. Four-dimensional docking: a fast and accurate account of discrete receptor flexibility in ligand docking. J Med Chem. 2009 Jan 22;52(2):397-406. doi: 10.1021/jm8009958.

[6] Ashton, M., Barnard, J., Casset, F., Charlton, M., Downs, G., Gorse, D., Holliday, J., Lahana, R. and Willett, P. (2002), Identification of Diverse Database Subsets using Property-Based and Fragment-Based Molecular Descriptions. Quant. Struct.-Act. Relat., 21: 598-604. https://doi.org/10.1002/qsar.200290002