A novel method for similarity analysis and protein sub-cellular localization prediction

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Biological sequences, including DNA and proteins, hold crucial information for understanding biological systems.
• Current methods for protein sequence analysis, often involving machine learning, are used for function classification, localization, and structure prediction.

Purpose of the Study:

• To develop a simpler and more effective method for biological sequence analysis.
• To introduce a novel approach for assessing protein sequence similarity.

Main Methods:

• A new 2D graphical representation of protein sequences based on the properties of 64 genetic codes.
• A Euclidean-distance method applied to the 2D graphical expression for sequence similarity analysis.

Main Results:

• The proposed method effectively captures more sequence information compared to traditional approaches.
• A phylogenetic tree constructed using this method demonstrated its validity.
• The method achieved reasonable accuracy in predicting protein sub-cellular localization on standard datasets.

Conclusions:

• The novel 2D graphical representation and Euclidean-distance method offer a simple yet effective approach to protein sequence analysis.
• This method shows promise for various bioinformatics applications, including phylogenetic analysis and sub-cellular localization prediction.