Related Experiment Video
Updated: Jun 6, 2026
16:41
A Protocol for Computer-Based Protein Structure and Function Prediction
Published on: November 3, 2011
POLYAR, a new computer program for prediction of poly(A) sites in human sequences.
Malik Nadeem Akhtar1, Syed Abbas Bukhari, Zeeshan Fazal
1Department of Biosciences, COMSATS Institute of Information Technology, Islamabad, Pakistan.
BMC Genomics
|November 25, 2010
Summary
A new classification and computer program, POLYAR, improve the prediction of strong polyadenylation sites in human pre-mRNA. While POLYAR shows high accuracy for strong sites, further research is needed for weak and poly(A) signal-less sites.
Area of Science:
- Molecular Biology
- Bioinformatics
Background:
- Messenger RNA (mRNA) polyadenylation is a critical step in eukaryotic gene expression.
- Accurate prediction of polyadenylation sites is essential for defining gene boundaries and understanding gene regulation.
Purpose of the Study:
- To classify human polyadenylation sites into distinct categories.
- To develop a novel computational tool, POLYAR, for predicting polyadenylation sites.
Main Methods:
- Classification of 28,761 human mapped poly(A) sites into PAS-strong, PAS-weak, and PAS-less groups.
- Development and comparative analysis of the POLYAR program against existing tools like polya_svm.
Main Results:
- POLYAR demonstrated significantly higher prediction sensitivity (80.8%) and specificity (66.4%) for PAS-strong poly(A) sites compared to polya_svm.
- Both POLYAR and polya_svm exhibited low accuracy for predicting PAS-weak and PAS-less poly(A) sites, indicating incomplete understanding of their regulatory factors.
Conclusions:
- A new classification and the POLYAR program enhance the prediction of strong polyadenylation sites.
- Further investigation is required to elucidate the factors governing PAS-weak and PAS-less polyadenylation sites.
- The POLYAR program is available for download.
