Proceedings of the 2016 International Conference on Biological Sciences and Technology

Analysis of Effect of the Position on Weighted Degree Kernel for Splice Site Prediction

Authors
Tian-Qi Wang, Yong Xu
Corresponding Author
Tian-Qi Wang
Available Online January 2016.
DOI
https://doi.org/10.2991/bst-16.2016.27How to use a DOI?
Keywords
Splice site prediction, Weighted degree kernel, Position factor, Support vector machine.
Abstract
Prediction of splice sites plays a key role in the annotation of genes. SVM with the weighted degree kernel has been proved to achieve a satisfactory performance. However, this kernel did not consider the effect of the position. In this article, we explored the relationship between the weighted degree kernel and the position of single base match. We defined a position factor to measure the effect of the position on weighted degree kernel, and selected several positions with high position factors to be key positions. Then we constructed a classification model and applied it to the Homo sapiens splice site dataset. To analyze the effect of the position of single base match, we removed the base in the key position and compared the classification accuracy with the accuracy without removing. The result shows that the position of single base match has significant influence on weighted degree kernel.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
The International Conference on Biological Sciences and Technology
Part of series
Advances in Biological Sciences Research
Publication Date
January 2016
ISBN
978-94-6252-161-2
ISSN
2468-5747
DOI
https://doi.org/10.2991/bst-16.2016.27How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Tian-Qi Wang
AU  - Yong Xu
PY  - 2016/01
DA  - 2016/01
TI  - Analysis of Effect of the Position on Weighted Degree Kernel for Splice Site Prediction
BT  - The International Conference on Biological Sciences and Technology
PB  - Atlantis Press
SN  - 2468-5747
UR  - https://doi.org/10.2991/bst-16.2016.27
DO  - https://doi.org/10.2991/bst-16.2016.27
ID  - Wang2016/01
ER  -