MEM-FET: Essential protein prediction using membership feature and machine learning approach

Area of Science:

• Bioinformatics
• Computational Biology
• Systems Biology

Background:

• Proteins are vital for cellular functions, and identifying essential proteins is key to understanding organisms.
• Experimental identification of essential proteins is time-consuming and expensive.
• Computational methods offer an efficient alternative for essential protein prediction.

Purpose of the Study:

• To develop and validate a novel computational method for predicting essential proteins.
• To reduce the cost and time associated with experimental essential protein identification.
• To improve the accuracy of essential protein prediction in understudied organisms.

Main Methods:

• A new methodology, MEM-FET (membership feature), was developed.
• MEM-FET utilizes features such as edge clustering coefficient, average clustering coefficient, subcellular localization, and Gene Ontology within common neighbors.
• Machine learning approaches, including ensemble methods, were employed for prediction and validation.

Main Results:

• The MEM-FET algorithm achieved accuracy values of 0.79, 0.74, 0.78, and 0.71 for YHQ, YMIPS, YDIP, and YMBD datasets, respectively.
• An enriched set of essential proteins was predicted using MEM-FET.
• The proposed MEM-FET model demonstrated superior performance, achieving 80% accuracy on the yeast dataset, outperforming existing algorithms.

Conclusions:

• MEM-FET is an effective computational approach for predicting essential proteins.
• The method significantly reduces experimental costs and time.
• MEM-FET offers a promising tool for essential protein identification in various organisms, especially understudied ones.