Related Experiment Videos
Sequence and structure-based prediction of eukaryotic protein phosphorylation sites
N Blom1, S Gammeltoft, S Brunak
1Department of Biotechnology, The Technical University of Denmark, Lyngby, DK-2800, Denmark.
Journal of Molecular Biology
|December 22, 1999
Summary
This study introduces an artificial neural network to predict protein phosphorylation sites, achieving high sensitivity. It identifies novel sites in p300/CBP and potential yin-yang regulation by glycosylation.
Area of Science:
- Molecular Biology
- Biochemistry
- Bioinformatics
Background:
- Protein phosphorylation is crucial for cellular signaling.
- Specificity in kinase-substrate recognition remains a key question.
- Understanding phosphorylation patterns is vital for deciphering cellular processes.
Purpose of the Study:
- To develop a computational method for predicting protein phosphorylation sites.
- To identify novel phosphorylation sites in p300/CBP.
- To investigate the interplay between glycosylation and phosphorylation.
Main Methods:
- Development of an artificial neural network (ANN) model.
- Prediction of phosphorylation sites in independent protein sequences.
- Analysis of p300/CBP protein for phosphorylation and glycosylation sites.
Main Results:
- The ANN achieved high prediction sensitivity (69-96%).
- Novel phosphorylation sites in p300/CBP were predicted, potentially affecting its functions.
- Serine and threonine residues in p300/CBP susceptible to O-linked N-acetylglucosamine glycosylation were identified.
Conclusions:
- The developed ANN is an effective tool for predicting phosphorylation sites.
- Predicted sites in p300/CBP may regulate transcription factor interactions and HAT activity.
- A potential 'yin-yang' regulatory mechanism between glycosylation and phosphorylation was proposed.