The critical importance of algorithms for orienting proteins in the lipid bilayer stems from the extreme difficulty in obtaining experimental data about the membrane boundaries. Here, we present a computational method for positioning protein structures in the membrane, based on the sole alpha carbon coordinates and, therefore, compatible with both high and low structural resolutions. Our algorithm follows a new and simple approach, by treating the membrane assignment problem as a binary classification. Compared with the state-of-the-art algorithms, our method achieves similar accuracy, while being faster. Finally, our open-source software is also capable of processing coarse-grained models of protein structures.
Keywords: binary classification; coarse-grained models; lipid bilayer assignment; low-resolution structures; membrane proteins.
© The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.