Frangakis Group

Research


The main focus of the group is to reveal the macromolecular organisation of living cells by means of cryo-electron tomography. Cryo-electron tomography is the only technique that can obtain molecular resolution images of intact cells in a quasi-native environment. The tomograms contain an imposing amount of information; they are essentially a three-dimensional map of the cellular proteome and depict the whole network of macromolecular interactions. Information mining algorithms exploit structural data from various techniques, identify distinct macromolecules and computationally fit atomic resolution structures in the cellular tomograms, thereby bridging the resolution gap.

A multitude of biological questions can be answered by electron tomography; visualization of the cellular structure at molecular resolution is largely an uncharted territory. The group works with a wide spectrum of specimens, including prokaryotic and eukaryotic cells, but also model systems. Prokaryotic cells are smaller and can therefore be easily penetrated by the electrons. Eukaryotic cells have compartments in which the protein density is lower, facilitating pattern recognition techniques. Model systems, on the other hand, are particularly helpful in improving computational algorithms and in providing solutions for cell systems that are too complex to be investigated by electron tomography.

Future projects and goals

Achieving these goals will enable us to visualise macromolecules in an unperturbed cellular environment and to chart the network of interactions underlying cellular functions. The aim of the group is to prove that a cell is not an envelope of freely diffusing enzymes and substrates, but rather a highly organized and coordinated machine.

 

The scheme visualises the ultimate goal of the group and the true power of cryo-electron tomography. We are using cryo-electron tomography in conjunction with pattern recognition techniques in order to match atomically resolved structures in the context of living cells. Practically, we integrate the information from X-ray crystallography, structural genomics and single-particle electron microscopy in order to computationally search for macromolecular complexes in the three-dimensional cryo-electron tomograms.