Two-stage support vector machines for protein structure and solvent prediction
We propose Two-Stage Support Vector Machines (TSSVM) for the prediction of structural properties of amino acid residues, namely, relative solvent accessibilities and protein secondary structure elements. The second stage of TSSVM extends the classical SVM approach to capture the contextual informati...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
2008
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/2407 |
| Summary: | We propose Two-Stage Support Vector Machines (TSSVM) for the prediction of structural properties of amino acid residues, namely, relative solvent accessibilities and protein secondary structure elements. The second stage of TSSVM extends the classical SVM approach to capture the contextual information among the secondary structural elements or the relative solvent accessibilities and thereby improves the accuracies of the predictions. |
|---|