Protein Remote Homology Detection and Fold Recognition Based on Sequence-Order Frequency Matrix

Area of Science:

• Bioinformatics
• Computational Biology
• Structural Biology

Background:

• Protein remote homology detection and fold recognition are vital for understanding protein structure and function.
• Current profile-based methods, like PSFM and PSSM, overlook sequence-order effects.

Purpose of the Study:

• To introduce a novel sequence-order frequency matrix (SOFM) that captures sequence-order information.
• To develop new predictors for protein remote homology detection and fold recognition using SOFMs.

Main Methods:

• Developed the sequence-order frequency matrix (SOFM) to extract neighboring residue information from MSAs.
• Integrated SOFMs with top-n-grams and Smith-Waterman algorithm for feature extraction.
• Created SOFM-Top and SOFM-SW predictors.

Main Results:

• SOFM demonstrates higher information content compared to existing profiles.
• The SOFM-Top and SOFM-SW predictors achieved superior performance in homology detection and fold recognition.
• Experimental results validate the effectiveness of SOFM.

Conclusions:

• SOFM effectively captures crucial sequence-order information, outperforming traditional profiles.
• The proposed SOFM-based predictors represent a significant advancement in protein analysis.
• SOFM is expected to be a valuable tool for studying protein structures and functions.