hide
Free keywords:
Cryo-electron tomography, template matching, support vector machines, spherical harmonics
Abstract:
Detection and identification of macromolecular complexes
in cryo-electron tomograms is challenging due to the extremely
low signal-to-noise ratio (SNR). While the state-ofthe-
art method is template matching with a single template,
we propose a 3-step supervised learning approach: (i) predetection
of candidates, (ii) feature calculation, and (iii) final
decision using a support vector machine (SVM). We use
two types of features for SVM: (i) correlation coefficients
from multiple templates, and (ii) rotation invariant features
derived from spherical harmonics. Experiments conducted
on both simulated and experimental tomograms show that
our approach outperforms the state-of-the-art method.
