70ProPred: a predictor for discovering sigma70 promoters based on combining multiple features.

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Promoters are critical DNA sequences regulating gene transcription initiation.
• Sigma70 promoters are essential for transcribing most genes in prokaryotes.
• Accurate identification of sigma70 promoter regions is challenging due to large-scale genome sequencing.

Purpose of the Study:

• To enhance the prediction accuracy of sigma70 promoters in prokaryotic DNA sequences.
• To develop a robust computational model for promoter recognition.

Main Methods:

• Developed the 70ProPred model for promoter recognition.
• Utilized sequence-based features: position-specific trinucleotide propensity (PSTNPss) and electron-ion potential values (PseEIIP).
• Optimized the model by combining 79 PSTNPss features and 64 PseEIIP features.

Main Results:

• The 70ProPred model achieved a prediction accuracy of 95.56%.
• The model demonstrated a Matthews correlation coefficient (MCC) of 0.90.
• Feature combination yielded superior performance in sigma70 promoter identification.

Conclusions:

• 70ProPred surpasses existing methods in accuracy and stability for sigma70 promoter prediction.
• The model's approach is adaptable for predicting promoters in other species.
• An accessible online web server (http://server.malab.cn/70ProPred/) is available for researchers.