NRProF

NRProF predicts protein function by applying a neural response algorithm to quantify protein subsequence similarity and assign Gene Ontology (GO) terms for functional annotation of sequences from high-throughput sequencing.

Key Features:

• Neural Response Algorithm: Employs a neural response algorithm inspired by neuronal behavior in the human visual cortex to detect similarities among protein subsequences.
• Distance Metric Definition: Defines a distance metric that quantifies similarity between subsequences for functional prediction.
• Functional Annotation Assignment: Predicts the most similar target protein for a query protein and assigns Gene Ontology (GO) terms based on that prediction, providing improved specificity over BLAST for detecting remote homologues.
• High Accuracy: Ranked the actual leaf GO term among the top five probable terms with an 86.93% success rate in testing.
• Integration of HMM Profiles and Secondary Structure Information: Incorporates HMM profiles and secondary structure information to enhance the neural response mechanism and annotation accuracy.

Scientific Applications:

• Protein Function Prediction: Predicts functions for newly sequenced proteins and for proteins lacking experimental characterization by detecting remote homologues.
• Comparative Genomics and Evolutionary Studies: Enables exploration of evolutionary relationships between proteins through sensitive sequence-similarity analysis.

Methodology:

Uses a neural network–inspired neural response algorithm and a defined distance metric to quantify subsequence similarity, integrates HMM profiles and secondary structure data, and assigns GO terms based on the most similar target protein.