PoreWalker: A Novel Tool for the Identification and Characterization of Channels in Transmembrane Proteins from Their Three-Dimensional Structure

Abstract

Transmembrane channel proteins play pivotal roles in maintaining the homeostasis and responsiveness of cells and the cross-membrane electrochemical gradient by mediating the transport of ions and molecules through biological membranes. Therefore, computational methods which, given a set of 3D coordinates, can automatically identify and describe channels in transmembrane proteins are key tools to provide insights into how they function. Herein we present PoreWalker, a fully automated method, which detects and fully characterises channels in transmembrane proteins from their 3D structures. A stepwise procedure is followed in which the pore centre and pore axis are first identified and optimised using geometric criteria, and then the biggest and longest cavity through the channel is detected. Finally, pore features, including diameter profiles, pore-lining residues, size, shape and regularity of the pore are calculated, providing a quantitative and visual characterization of the channel. To illustrate the use of this tool, the method was applied to several structures of transmembrane channel proteins and was able to identify shape/size/residue features representative of specific channel families. The software is available as a web-based resource at http://www.ebi.ac.uk/thornton-srv/software/PoreWalker/. Transmembrane channel proteins are responsible for the transport of ions and molecules through biological membranes and are pivotal for the physiology of the cell. In fact, their incorrect functioning is involved or related to several diseases (diabetes, myotonia, Parkinson's disease, etc.). Moreover, their specificity and selectivity to different ions or molecules have been hypothesized and sometimes shown to strongly depend on the shape and size or amino acid composition of the channel. Therefore, computational methods to identify and quantitatively characterise channel geometry in transmembrane protein structures are key tools to better understand how they function. We have developed PoreWalker, a new method to detect and describe the geometry of these channels in transmembrane proteins from their 3D structures. The method is fully automated, very user-friendly, identifies the location of the channel and derives a number of channel features: diameter profiles at given heights along the channel, all the residues lining the channel walls, size, shape and regularity of the channel. These features can be very helpful in the study of how these channels might function. We have applied PoreWalker to several channel protein structures and were able to identify shape/size/residue features that were representative of specific channel families.