A multi-conformational virtual screening approach based on machine learning targeting PI3Kγ

Zhu, Jingyu; Jiang, Yingmin; Jia, Lei; Xu, Lei; Cai, Yanfei; Chen, Yun; Zhu, Nannan; Li, Huazhong; Jin, Jian

doi:10.1007/s11030-021-10243-1

Cited by 12 publications

(6 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Methodsmentioning

confidence: 99%

“…Molecular Docking Setup: Four mainstream molecular docking programs were employed in this study: two commercial software programs, namely CDOCKER from Discovery Studio 2017 (DS2017) and Glide from Schrödinger (version 2018); two academic software programs, namely AutoDock Vina (version 1.2.5) and LeDock. [53] During docking, the positions of all docking pockets were set to the location where the eutectic ligand was situated. Moreover, considering the practical virtual screening process, semi-flexible docking was used for each docking program (rigid for the receptor and flexible for the ligand).…”

Section: Methodsmentioning

confidence: 99%

“…In order to evaluate the overall performance of docking enrichment, the docking scores of the dataset were used to generate a receiver operating characteristic (ROC) curve, and the corresponding area under the curve (AUC) value was determined. [53] Generation and Validation of Pharmacophore Models: All 22 crystal structures were prepared by the Prepare Protein module of DS2017, including adding polar hydrogens, removing water molecules, repairing broken chains, and incorporating CHARMm fields with default parameters. Then, the Receptor-Ligand Pharmacophore Generation (RLPG) protocol in DS2017 was employed to directly generate pharmacophore models from the receptor-ligand interactions as revealed in the 3D structures.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Integration of Multicomplex‐Based Pharmacophore Modeling and Molecular Docking in Machine Learning‐Based Virtual Screening: Toward the Discovery of Novel PI3K Inhibitors

Qiu,

Jia,

Yuan

et al. 2024

Advcd Theory and Sims

Self Cite

View full text Add to dashboard Cite

The phosphatidylinositol‐3 kinase (PI3K) pathway is a crucial intracellular signaling pathway within living cells. The hyperactivation of PI3K signaling cascades is a common occurrence in human cancers, rendering PI3K a promising therapeutic target. Although several PI3K inhibitors are already available on the market, the adverse side effects of current therapies continue to highlight the necessity for the development of novel PI3K inhibitors. In this study, a virtual screening strategy employing naïve Bayesian classification (NBC) models, based on multicomplex‐based molecular docking and pharmacophore modeling, is developed. First, the docking accuracy and scoring reliability of four docking software are assessed, and Glide demonstrated higher predictability for PI3K inhibitors. Second, pharmacophore models are generated based on the current reported PI3K‐inhibitor interactions, and five pharmacophore hypotheses displayed significant capability in discriminating active PI3K molecules from inactive ones. Subsequently, three NBC models are constructed based on molecular docking and/or pharmacophore models, and the validation results showed that the NBC model, combining multicomplex‐based molecular docking and pharmacophore, significantly improved the hit rate of virtual screening against PI3K. Finally, the optimal NBC model is employed for virtual screening against the ChEMBL database, leading to the identification of multiple molecules with high potential as active PI3K inhibitors.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Integration of Multicomplex‐Based Pharmacophore Modeling and Molecular Docking in Machine Learning‐Based Virtual Screening: Toward the Discovery of Novel PI3K Inhibitors

Qiu,

Jia,

Yuan

et al. 2024

Advcd Theory and Sims

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our previous studies have demonstrated that eight complexes would be satisfactory to build the multiple structure-based VS model, [22,24,26,28,29] thus, 24 ligand-JAK3 binding complexes were divided into eight categories according to structural characteristics using the k-means algorithm. [30,31] Finally, the protein whose RMSD (root-mean-square deviation) value was closest to the centroid within the same category was selected as the representative structure.…”

Section: Selection Of Representative Jak3 Conformationsmentioning

confidence: 99%

“…[19][20][21] In our previous works, several SBVS models based on multiple protein structures have been established to identify kinase selective inhibitors and some compounds with potent bioactivities were successfully screened out. All these results fully reveal the reliability and practicability of the ensemble-based VS. [22][23][24][25][26][27] Therefore, in this present study, an in silico ensemble strategy that combines molecular docking, pharmacophore, and naïve Bayesian classification (NBC) using multiple JAK3-inhibitor complexes was developed for virtual screening against JAK3 protein (Figure 1).…”

Section: Introductionmentioning

confidence: 99%

Machine Learning‐Enabled Virtual Screening with Multiple Protein Structures toward the Discovery of Novel JAK3 Inhibitors: Integration of Molecular Docking, Pharmacophore, and Naïve Bayesian Classification

Zhu

Sun

Jia

et al. 2023

Advcd Theory and Sims

Self Cite

View full text Add to dashboard Cite

Extensive research has accumulated suggesting that Janus kinase 3 (JAK3) is closely related to the occurrence and development of various human diseases, making JAK3 a highly potential drug target. However, JAK3 has high homology with other members of the JAK family, making the development of JAK3 inhibitors full of challenges. Thus, here, a naïve Bayesian classification (NBC) model based on multiple JAK3 protein conformations, which integrates molecular docking, pharmacophore, and molecular descriptors, is developed to find novel JAK3 inhibitors. First, the validation set is used to prove whether molecular docking or pharmacophore, integrating multiple JAK3 conformations always has higher prediction accuracy than that of any single conformation. Second, external prediction reveals that the NBC model combining molecular docking, pharmacophore, and important molecular features could significantly improve the enrichment of active JAK3 inhibitors. Finally, the optimal NBC model is utilized for virtual screening against a large chemical database and some compounds with high Bayesian scores are identified. Altogether, the machine learning-based virtual screening protocol not only has strong efficiency but also has high screening accuracy. It is hoped that the developed virtual screening strategy could provide valuable guidance for the discovery of novel JAK3 inhibitors.

show abstract

Integration of Consensus‐Based Pharmacophore Mapping and Molecular Docking in Sequential Virtual Screening: Toward the Discovery of Novel PI3Kγ Inhibitors

Yu,

Yuan,

Jia

et al. 2024

ChemistrySelect

View full text Add to dashboard Cite

Numerous research studies have demonstrated the significant correlation of phosphatidylinositol‐3 kinase gamma (PI3Kγ) with the onset and progression of various human diseases, highlighting PI3Kγ as a promising therapeutic target. However, PI3Kγ demonstrates considerable similarity with other isoforms in the PI3K family, presenting significant challenges in the creation of PI3Kγ inhibitors. This study presents an ensemble‐based virtual screening approach to discover novel inhibitors targeting PI3Kγ. Eganelisib (IPI‐549) is the sole selective PI3Kγ inhibitor that has progressed to clinical trials, making it a significant model for the advancement of novel PI3Kγ inhibitors. Initially, common feature pharmacophore and receptor–ligand pharmacophore models were independently developed using IPI‐549 and its potent derivatives, in conjunction with the crystal complex of PI3Kγ/IPI‐549. Both qualitative pharmacophore models proved highly effective at distinguishing between active and inactive compounds. Then, four widely utilized docking programs were chosen for assessment, where the Glide_SP mode demonstrated superior predictive accuracy in sampling ligand conformations during binding, effectively distinguishing between PI3Kγ inhibitors and noninhibitors. Finally, a virtual screening protocol was conducted to screen the ChEMBL database, utilizing similarity search, consensus‐based pharmacophore mapping, and sequential molecular docking. This process resulted in the identification of multiple molecules exhibiting notable promise as potent PI3Kγ inhibitors.

show abstract

A multi-conformational virtual screening approach based on machine learning targeting PI3Kγ

Cited by 12 publications

References 55 publications

Integration of Multicomplex‐Based Pharmacophore Modeling and Molecular Docking in Machine Learning‐Based Virtual Screening: Toward the Discovery of Novel PI3K Inhibitors

Integration of Multicomplex‐Based Pharmacophore Modeling and Molecular Docking in Machine Learning‐Based Virtual Screening: Toward the Discovery of Novel PI3K Inhibitors

Machine Learning‐Enabled Virtual Screening with Multiple Protein Structures toward the Discovery of Novel JAK3 Inhibitors: Integration of Molecular Docking, Pharmacophore, and Naïve Bayesian Classification

Integration of Consensus‐Based Pharmacophore Mapping and Molecular Docking in Sequential Virtual Screening: Toward the Discovery of Novel PI3Kγ Inhibitors

Contact Info

Product

Resources

About