iProm-Sigma54

iProm-Sigma54 predicts σ54 promoter sequences using a convolutional neural network to identify σ54-dependent promoters for studying prokaryotic gene transcription and regulation.

Key Features:

• Biological target: Specifically trained to identify σ54 promoters associated with the σ (sigma) factor and σ54-dependent RNA holoenzymes in prokaryotes.
• Model architecture: Implements a convolutional neural network (CNN) architecture.
• Layer composition: Uses two one-dimensional convolutional layers followed by max pooling layers and dropout layers for feature extraction and regularization.
• Input encoding: Applies one-hot encoding to transform nucleotide sequences into matrix format for neural network input.
• Evaluation protocol: Assessed using four distinct performance metrics and five-fold cross-validation on benchmark and test datasets.
• Comparative performance: Reported to outperform existing methodologies on benchmark and test datasets for σ54 promoter identification.

Scientific Applications:

• σ54 promoter identification: Detects σ54 promoter sequences in prokaryotic genomic sequences.
• Gene regulation studies: Facilitates investigation of σ54-mediated transcriptional regulation and environmentally responsive processes.
• Promoter analysis: Enables comparative evaluation of promoter prediction approaches using benchmark and test datasets.

Methodology:

Input nucleotide sequences are one-hot encoded and processed by a CNN with two 1D convolutional layers followed by max pooling and dropout; performance was measured with four assessment metrics using five-fold cross-validation on benchmark and test datasets.