Technical Report No. 252 - Abstract

L.Y. Ma, M. Reisert, H. Burkhardt
RENNSH: A Novel alpha-helices Identification Approach for Intermediate Resolution Cryo-EM Maps

Accurate identification of protein secondary structures is beneficial to understand three-dimensional structures of biological macromolecules. In this paper, a novel refined classification frame is proposed, which can convert the alpha-helices identification into a matter of pure machine learning problem by representing each voxel in the dentsity map with its Spherical Harmonic Descriptors (SHD). We call it the RENNSH approach. Comparing with other existing alpha-helices identification methods for intermediate resolution cryo-EM maps, the experimental results demonstrate that the RENNSH approach gives the best identification accuracy. Also a voxelwise evaluation function is defined for evaluating the identification quality, which can be used by all the alpha-helices identification methods.