PRRpred

PRRpred predicts Pattern Recognition Receptors (PRRs) from protein sequences by combining similarity-based and machine-learning approaches that leverage sequence composition and evolutionary information.

Key Features:

• Data Source: Trained on the largest non-redundant dataset of PRRs and Non-PRRs from the Pattern Recognition Receptor Database (PRRDB 2.0).
• Hybrid Model: Integrates similarity-based BLAST with machine learning models to enhance prediction accuracy.
• Sequence Features: Utilizes sequence composition for model input.
• Evolutionary Features (PSSM): Incorporates Position-Specific Scoring Matrix (PSSM) profiles as evolutionary information.

Scientific Applications:

• PRR Prediction: Facilitates identification of PRRs to support immune response studies and drug discovery targeting immune pathways.

Methodology:

Combines BLAST similarity searches with machine-learning models trained on sequence composition and PSSM-derived evolutionary features using a non-redundant PRRDB 2.0 dataset.