Next Article in Journal
The Emerging Role of GLP-1 Receptors in DNA Repair: Implications in Neurological Disorders
Next Article in Special Issue
CytoCluster: A Cytoscape Plugin for Cluster Analysis and Visualization of Biological Networks
Previous Article in Journal
Potential Diagnostic and Therapeutic Applications of Oligonucleotide Aptamers in Breast Cancer
Previous Article in Special Issue
IonchanPred 2.0: A Tool to Predict Ion Channels and Their Types
Open AccessArticle

PSFM-DBT: Identifying DNA-Binding Proteins by Combing Position Specific Frequency Matrix and Distance-Bigram Transformation

by and *
School of Computer Science and Technology, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen 518055, China
Author to whom correspondence should be addressed.
Int. J. Mol. Sci. 2017, 18(9), 1856;
Received: 28 July 2017 / Revised: 19 August 2017 / Accepted: 22 August 2017 / Published: 25 August 2017
(This article belongs to the Special Issue Special Protein Molecules Computational Identification)
DNA-binding proteins play crucial roles in various biological processes, such as DNA replication and repair, transcriptional regulation and many other biological activities associated with DNA. Experimental recognition techniques for DNA-binding proteins identification are both time consuming and expensive. Effective methods for identifying these proteins only based on protein sequences are highly required. The key for sequence-based methods is to effectively represent protein sequences. It has been reported by various previous studies that evolutionary information is crucial for DNA-binding protein identification. In this study, we employed four methods to extract the evolutionary information from Position Specific Frequency Matrix (PSFM), including Residue Probing Transformation (RPT), Evolutionary Difference Transformation (EDT), Distance-Bigram Transformation (DBT), and Trigram Transformation (TT). The PSFMs were converted into fixed length feature vectors by these four methods, and then respectively combined with Support Vector Machines (SVMs); four predictors for identifying these proteins were constructed, including PSFM-RPT, PSFM-EDT, PSFM-DBT, and PSFM-TT. Experimental results on a widely used benchmark dataset PDB1075 and an independent dataset PDB186 showed that these four methods achieved state-of-the-art-performance, and PSFM-DBT outperformed other existing methods in this field. For practical applications, a user-friendly webserver of PSFM-DBT was established, which is available at View Full-Text
Keywords: PSFM-DBT; DNA binding protein; distance bigram transformation; PSFM PSFM-DBT; DNA binding protein; distance bigram transformation; PSFM
Show Figures

Graphical abstract

MDPI and ACS Style

Zhang, J.; Liu, B. PSFM-DBT: Identifying DNA-Binding Proteins by Combing Position Specific Frequency Matrix and Distance-Bigram Transformation. Int. J. Mol. Sci. 2017, 18, 1856.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map

Back to TopTop