PolarProtPred

PolarProtPred predicts membrane-domain localization of proteins in polarized epithelial cells by using deep learning to detect short linear motifs that direct sorting to apical or basolateral membranes.

Key Features:

• Deep Learning Approach: PolarProtPred employs neural networks to identify recurrent sequence patterns, including short linear motifs, associated with protein localization.
• Focus on Cell Polarity: The method targets protein sorting in polarized epithelial cells and predicts apical versus basolateral localization relevant to cell polarity.
• Application to Drug Transporters and Viral Entry Receptors: It assists analysis of vectorial drug transport by drug transporters that govern pharmacokinetics and can characterize surface receptors, including viral entry receptors implicated in COVID-19.
• Support for Experimental Research: PolarProtPred provides localization predictions to support experimental studies of molecular networks controlling transporter and receptor distribution across cellular membranes.

Scientific Applications:

• Epithelial Cell Biology: Studying protein sorting mechanisms within polarized epithelial cells, including apical versus basolateral targeting.
• Pharmacology and Drug Delivery: Investigating distribution and function of drug transporters to inform vectorial drug transport and pharmacokinetics.
• Virology and Receptor Characterization: Characterizing surface and viral entry receptors, including receptors relevant to COVID-19, to inform antiviral research.

Methodology:

PolarProtPred uses neural networks trained on datasets of transmembrane proteins to detect short linear motifs associated with sorting into apical or basolateral membranes, with predictions based on pattern recognition of these motifs.