IN-MACA-MCC: Integrated Multiple Attractor Cellular Automata with Modified Clonal Classifier for Human Protein Coding...

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Accurate prediction of protein coding and promoter regions is crucial for gene function understanding.
• Existing methods often predict these regions separately, indicating a need for integrated approaches.

Purpose of the Study:

• To develop a single, unified classifier for predicting both protein coding and promoter regions.
• To improve the efficiency and accuracy of gene region identification in bioinformatics.

Main Methods:

• A novel classifier integrating multiple attractor cellular automata (MACA) and modified clonal classifier (MCC) was developed.
• The classifier was trained and tested using established datasets including Fickett and Tung, MMCRI, DBTSS, EID, and UTRdb.
• Performance was evaluated on DNA sequences of varying lengths for protein coding regions and promoter/non-promoter sequences.

Main Results:

• The proposed classifier achieved an average accuracy of 90.5% for promoter region prediction.
• An average accuracy of 89.6% was obtained for protein coding region prediction.
• High specificity (0.89) and sensitivity (0.92) were reported for both prediction types.

Conclusions:

• The integrated MACA and MCC classifier effectively predicts both protein coding and promoter regions with high accuracy.
• This unified approach offers a promising advancement in gene region identification within bioinformatics.
• The model demonstrates strong performance, suggesting its utility in genomic analysis.