Visual Proteomics

To determine the density maps of protein complexes from tomograms it is necessary to align and average subtomograms of these complexes. Given the typically large number of subtomograms in a whole cell tomogram, exhaustive search of all possible combinations of subtomograms is computationally challenging. We have developed an efficient alignment method for large sets of subtomograms, which provides the key step in subtomogram classification and averaging. Our method also considers image distortions due to missing wedge effects and high noise levels. The method starts from a fast rotational search to enumerate a small number of candidate rotations, and then performs a gradient-based stochastic parallel refinement to achieve high precision.

High precision alignment of cryo-electron subtomograms through gradient-based parallel optimization. BMC Systems Biology (2012)

High-throughput subtomogram alignment and classification by Fourier space constrained volumetric matching. Journal of Structural Biology (2012)

Gradient-based high precision alignment of cryo-electron subtomograms. IEEE International Conference on Systems Biology (2011)

Electron cryotomography enables 3D visualization of cells in a near-native state at molecular resolution. The produced cellular tomograms contain detailed information about a plethora of macromolecular complexes, their structures, abundances, and specific spatial locations in the cell. However, extracting this information in a systematic way is very challenging, and current methods usually rely on individual templates of known structures. We have developed methods and software tools for de novo discovery of different complexes from highly heterogeneous sets of particles extracted from entire cellular tomograms without using information of known structures. These initially detected structures can then serve as input for more targeted refinement efforts.

3D Rotation Invariant Features for the Characterization of Molecular Density Maps. IEEE International Conference on Bioinformatics and Biomedicine (2009)

Template-free detection of macromolecular complexes in cryo-electron tomograms. Bioinformatics (2011)

Automated target segmentation and fast alignment methods for high-throughput classification and averaging of crowded cryo-electron subtomograms. Bioinformatics (2013)

De Novo Structural Pattern Mining in Cellular Electron Cryotomograms. Structure (2019) (Highlighted by Nature Methods)

Information is increasingly available about the membrane bound compartmentalization in the cell (from cryoET), the identity and number of cell components (from genomics, mass spectrometry), the spatial distributions of protein complexes (from cryoET imaging) and the protein interaction and reaction rate constants (from biochemical experiments). There is a pressing need to integrate such information into spatially explicit predictive models of whole cells. A goal in my research is to develop such a computational method to study the higher-order systematic behavior of the proteome dynamics in biological processes.