Related Experiment Video
Updated: Mar 22, 2026
11:13
Identification of Kinase-substrate Pairs Using High Throughput Screening
Published on: August 29, 2015
8.7K
Constructing and Validating High-Performance MIEC-SVM Models in Virtual Screening for Kinases: A Better Way for
Huiyong Sun1,2, Peichen Pan1, Sheng Tian3
1College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, Zhejiang 310058, P. R. China.
Scientific Reports
|April 23, 2016
Summary
The MIEC-SVM approach effectively identifies potential drug inhibitors by analyzing molecular interactions. This method significantly outperforms traditional tools like Autodock in structure-based virtual screening for kinase targets.
Area of Science:
- Computational chemistry
- Drug discovery
- Bioinformatics
Background:
- The Molecular Interaction Energy Components-Support Vector Machine (MIEC-SVM) approach is effective for protein-peptide recognition.
- Its efficacy in identifying small molecule inhibitors for drug targets requires experimental validation.
Purpose of the Study:
- To assess and validate the MIEC-SVM approach for identifying small molecule inhibitors against kinase drug targets.
- To optimize MIEC-SVM model construction protocols.
Main Methods:
- Developed and optimized MIEC-SVM models for ABL, ALK, and BRAF kinase targets.
- Compared optimized MIEC-SVM models against default SVM parameters and Autodock.
- Screened the Specs database using the optimized MIEC-SVM strategy to identify ALK kinase inhibitors.
Main Results:
- Optimized MIEC-SVM models demonstrated superior performance over default SVM and Autodock for the tested kinase targets.
- The MIEC-SVM strategy identified 7 active ALK inhibitors (14% hit rate) from 50 compounds, including 4 in the nM range.
- Autodock identified only 3 active inhibitors (6% hit rate) from the same set, with 2 in the nM range.
Conclusions:
- The optimized MIEC-SVM approach is a powerful and validated tool for structure-based virtual screening in drug discovery.
- This strategy significantly improves the hit rate and identification of potent inhibitors compared to conventional methods.
