Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features
Since proteins work in the context of many other proteins and rarely work in isolation, it is higly important to study protein-protein interactions to understand proteins functions. The interactions data that have been identified by high-throughput technologies like the yeast two-hybrid system are k...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | en |
| Published: |
2006
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/8754/1/ICOCI-2006.pdf http://eprints.utm.my/8754/ https:dx.doi.org/10.1109/ICOCI.2006.5276519 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1845472071443808256 |
|---|---|
| author | Alashwal, Hany Taher Ahmed Deris, Safaai Othman, Muhamad Razib |
| author_facet | Alashwal, Hany Taher Ahmed Deris, Safaai Othman, Muhamad Razib |
| author_sort | Alashwal, Hany Taher Ahmed |
| building | UTM Library |
| collection | Institutional Repository |
| content_provider | Universiti Teknologi Malaysia |
| content_source | UTM Institutional Repository |
| continent | Asia |
| country | Malaysia |
| description | Since proteins work in the context of many other proteins and rarely work in isolation, it is higly important to study protein-protein interactions to understand proteins functions. The interactions data that have been identified by high-throughput technologies like the yeast two-hybrid system are known to yield many false positives. As a result, methods for computational prediction of protein-protein interactions based on sequence information are becoming increasingly important. In this study, computational prediction of protein-protein interactions (PPI) from domain structure and hydrophobicity properties, is presented. Protein domain structure and hydrophobicity properties are used separately as the sequence feature for the support vector machines (SVM) as a learning system. Both features achieved accuracy of about 80%. But domains structure had receiver operating characteristic (ROC) score of 0.8480 with running time of 34 seconds, while hydrophobicity had ROC score of 0.8159 with running time of 20,571 seconds (5.7 hours). These results indicate that protein-protein interaction can be predicted from domain structure with reliable accuracy and acceptable running time. |
| format | Conference or Workshop Item |
| id | my.utm.eprints-8754 |
| institution | Universiti Teknologi Malaysia |
| language | en |
| publishDate | 2006 |
| record_format | eprints |
| spelling | my.utm.eprints-87542017-08-30T01:25:27Z http://eprints.utm.my/8754/ Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features Alashwal, Hany Taher Ahmed Deris, Safaai Othman, Muhamad Razib T Technology (General) Since proteins work in the context of many other proteins and rarely work in isolation, it is higly important to study protein-protein interactions to understand proteins functions. The interactions data that have been identified by high-throughput technologies like the yeast two-hybrid system are known to yield many false positives. As a result, methods for computational prediction of protein-protein interactions based on sequence information are becoming increasingly important. In this study, computational prediction of protein-protein interactions (PPI) from domain structure and hydrophobicity properties, is presented. Protein domain structure and hydrophobicity properties are used separately as the sequence feature for the support vector machines (SVM) as a learning system. Both features achieved accuracy of about 80%. But domains structure had receiver operating characteristic (ROC) score of 0.8480 with running time of 34 seconds, while hydrophobicity had ROC score of 0.8159 with running time of 20,571 seconds (5.7 hours). These results indicate that protein-protein interaction can be predicted from domain structure with reliable accuracy and acceptable running time. 2006 Conference or Workshop Item PeerReviewed application/pdf en http://eprints.utm.my/8754/1/ICOCI-2006.pdf Alashwal, Hany Taher Ahmed and Deris, Safaai and Othman, Muhamad Razib (2006) Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features. In: IEEE International Conference on Computing & Informatics (ICOCI'2006), 6-8 June 2006, Kuala Lumpur. https:dx.doi.org/10.1109/ICOCI.2006.5276519 |
| spellingShingle | T Technology (General) Alashwal, Hany Taher Ahmed Deris, Safaai Othman, Muhamad Razib Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title | Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title_full | Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title_fullStr | Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title_full_unstemmed | Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title_short | Support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| title_sort | support vector machines for predicting protein-protein interactions using domains and hydrophobicity features |
| topic | T Technology (General) |
| url | http://eprints.utm.my/8754/1/ICOCI-2006.pdf http://eprints.utm.my/8754/ https:dx.doi.org/10.1109/ICOCI.2006.5276519 |
| url_provider | http://eprints.utm.my/ |