Integrating docking scores, interaction profiles and molecular descriptors to improve the accuracy of molecular docking: Toward the discovery of novel Akt1 inhibitors

Wen-hu, Zhan; Li, Da‐Qiang; Che, Jinxin; Zhang, Liangren; Yang, Bo; Hu, Yongzhou; Liu, Tao; Dong, Xiaowu

doi:10.1016/j.ejmech.2014.01.019

Cited by 47 publications

(26 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding prospective applications of family‐specific machine‐learning SFs, Zhan et al developed a SVR‐based SF aimed at guiding the optimization of Akt1 inhibitors. From retrospective analysis of the set of 47 Akt1 inhibitors, the authors concluded that none of the five classical SFs tested was suitable for the task.…”

Section: Family‐specific Machine‐learning Sfsmentioning

confidence: 99%

“…A prospective application of a family‐specific machine‐learning SF has already been carried out, leading to the discovery of several low‐nanomolar Akt1 inhibitors. It is noteworthy that this SVR model was trained on the docking poses of a set of known inhibitors, as crystal structures for these ligands were not available.…”

Section: Conclusion and Future Prospectsmentioning

confidence: 99%

See 1 more Smart Citation

Machine‐learning scoring functions to improve structure‐based binding affinity prediction and virtual screening

Ain

Aleksandrova

Roessler

et al. 2015

WIREs Comput Mol Sci

283

234

View full text Add to dashboard Cite

Docking tools to predict whether and how a small molecule binds to a target can be applied if a structural model of such target is available. The reliability of docking depends, however, on the accuracy of the adopted scoring function (SF). Despite intense research over the years, improving the accuracy of SFs for structure‐based binding affinity prediction or virtual screening has proven to be a challenging task for any class of method. New SFs based on modern machine‐learning regression models, which do not impose a predetermined functional form and thus are able to exploit effectively much larger amounts of experimental data, have recently been introduced. These machine‐learning SFs have been shown to outperform a wide range of classical SFs at both binding affinity prediction and virtual screening. The emerging picture from these studies is that the classical approach of using linear regression with a small number of expert‐selected structural features can be strongly improved by a machine‐learning approach based on nonlinear regression allied with comprehensive data‐driven feature selection. Furthermore, the performance of classical SFs does not grow with larger training datasets and hence this performance gap is expected to widen as more training data becomes available in the future. Other topics covered in this review include predicting the reliability of a SF on a particular target class, generating synthetic data to improve predictive performance and modeling guidelines for SF development. WIREs Comput Mol Sci 2015, 5:405–424. doi: 10.1002/wcms.1225For further resources related to this article, please visit the WIREs website.

show abstract

Section: Family‐specific Machine‐learning Sfsmentioning

confidence: 99%

Section: Conclusion and Future Prospectsmentioning

confidence: 99%

Machine‐learning scoring functions to improve structure‐based binding affinity prediction and virtual screening

Ain

Aleksandrova

Roessler

et al. 2015

WIREs Comput Mol Sci

283

234

View full text Add to dashboard Cite

show abstract

“…As a continuation of our virtual screening work which identified new inhibitors targeting NIK, CHK1, Akt, etc. (18)(19)(20)(21)(22)(23)(24). In this study, we performed a traditional VS procedure to identify potential inhibitors of IRAK1.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Artificial Intelligence in Participating Structure-Based Virtual Screening for Identifying Novel Interleukin-1 Receptor Associated Kinase-1 Inhibitors

et al. 2020

Self Cite

View full text Add to dashboard Cite

“…All these methods have been applied successfully for the identification of novel bioactive compounds [22–24] , but which method is the most suitable for a specific target class or research questions still remains largely elusive. In addition, in a recent study, we could observe substantial differences even between programs that rely upon the same methodology.…”

Section: Introductionmentioning

confidence: 99%

Prospective performance evaluation of selected common virtual screening tools. Case study: Cyclooxygenase (COX) 1 and 2

Kaserer

Temml

Kutil

et al. 2015

European Journal of Medicinal Chemistry

View full text Add to dashboard Cite

Computational methods can be applied in drug development for the identification of novel lead candidates, but also for the prediction of pharmacokinetic properties and potential adverse effects, thereby aiding to prioritize and identify the most promising compounds. In principle, several techniques are available for this purpose, however, which one is the most suitable for a specific research objective still requires further investigation. Within this study, the performance of several programs, representing common virtual screening methods, was compared in a prospective manner. First, we selected top-ranked virtual screening hits from the three methods pharmacophore modeling, shape-based modeling, and docking. For comparison, these hits were then additionally predicted by external pharmacophore- and 2D similarity-based bioactivity profiling tools. Subsequently, the biological activities of the selected hits were assessed in vitro, which allowed for evaluating and comparing the prospective performance of the applied tools. Although all methods performed well, considerable differences were observed concerning hit rates, true positive and true negative hits, and hitlist composition. Our results suggest that a rational selection of the applied method represents a powerful strategy to maximize the success of a research project, tightly linked to its aims. We employed cyclooxygenase as application example, however, the focus of this study lied on highlighting the differences in the virtual screening tool performances and not in the identification of novel COX-inhibitors.

show abstract

Integrating docking scores, interaction profiles and molecular descriptors to improve the accuracy of molecular docking: Toward the discovery of novel Akt1 inhibitors

Cited by 47 publications

References 30 publications

Machine‐learning scoring functions to improve structure‐based binding affinity prediction and virtual screening

Machine‐learning scoring functions to improve structure‐based binding affinity prediction and virtual screening

Evaluation of Artificial Intelligence in Participating Structure-Based Virtual Screening for Identifying Novel Interleukin-1 Receptor Associated Kinase-1 Inhibitors

Prospective performance evaluation of selected common virtual screening tools. Case study: Cyclooxygenase (COX) 1 and 2

Contact Info

Product

Resources

About