PRRpred: A Hybrid Model for Predicting Pattern Recognition Receptors Using Evolutionary Information

PRRpred: A Hybrid Model for Predicting Pattern Recognition Receptors Using Evolutionary Information PRRpred is a computational web server developed for predicting Pattern Recognition Receptors, also known as PRRs, from protein sequences. Pattern Recognition Receptors are important components of the innate immune system. They recognize pathogen-associated molecular patterns and damage-associated molecular patterns, helping the immune system detect infection, tissue damage, and inflammation. PRRpred uses similarity-based search, machine learning, and evolutionary information to classify proteins as PRRs or non-PRRs. Web Server: http://webs.iiitd.edu.in/raghava/prrpred/ Citation Kaur, D., Arora, C., and Raghava, G. P. S. A Hybrid Model for Predicting Pattern Recognition Receptors Using Evolutionary Information. Frontiers in Immunology, 11, 71, 2020. https://doi.org/10.3389/fimmu.2020.00071 About the Research Pattern Recognition Receptors are germline-encoded immune receptors that recognize conserved molecular patterns present in pathogens or damaged host cells. These receptors play a major role in innate immunity and help initiate inflammatory and antimicrobial responses. Major classes of PRRs include: Toll-like receptors NOD-like receptors RIG-I-like receptors C-type lectin receptors Scavenger receptors Mannose receptors Peptidoglycan recognition proteins Secreted PRRs such as collectins, ficolins, and pentraxins PRRs are involved in pathogen recognition, inflammation, autoimmune disorders, cancer, immunodeficiency, and vaccine adjuvant research. PRRpred was developed to computationally identify PRRs from protein sequences and support innate immunity research. Data Compilation: PRR sequences were collected from PRRDB 2.0, while non-PRR sequences were obtained from Swiss-Prot. After removing identical sequences, the final dataset contained 179 unique PRRs and 274 non-PRRs. Methodology: PRRpred uses a hybrid prediction strategy combining BLAST-based similarity search with machine learning models trained on protein sequence composition and evolutionary information.