Cryo-electron tomography (cryo-ET) is a unique method to reveal macromolecular structures in their native state and context within cells. Biological samples of interest are plunge frozen in a near-native state and imaged at many different tilt angles. The resulting tilt-series of images are used to generate 3-dimensional (3D) tomograms with molecular resolution. When many copies of a structure of interest are imaged, sub-tomogram averaging is used to create even higher, sub-nanometer resolution images to help unravel the arrangement among structural subunits. Cryo-ET has emerged as a unique method to bridge the resolution gap between light microscopy and atomic resolution methods (e.g. crystallography and nuclear magnetic resonance).
Since an electron microscope can take 20-40 tilt-series per day and manual 3D reconstruction of a single tilt-series takes about 30 minutes, it would be extremely beneficial for researchers to be able to continuously collect large amounts of tilt-series data without needing to manually process the tilt-series over several days. In addition, researchers need easy access to previous tomograms since new discoveries in different fields can cause unnoticed structures in existing tomograms to become new targets of interest. As a result, any institution conducting cryo-ET research would benefit from automated tomography data processing in addition to maintaining database access to existing tomograms.
My project over the summer was to create such a high-throughput cryo-ET pipeline and database at the University of Pennsylvania. My constructed system has already enabled cryo-ET researchers to image novel structures and focus on data analysis by automating the entire process from data collection, to data processing, and finally data storage. In addition, the automated real time tilt-series processing has allowed researchers to view their initial 3D tomograms during data collection and accordingly adjust their collection parameters in a timely fashion, optimizing valuable electron microscope time. I also created a web server accessible from anywhere in the world that allows researchers to easily view every tomogram and the associated files they have collected.
Unlike my classes where I am given the needed prerequisite knowledge and already established solutions, my project has required me to solve a large number of novel problems. No one could tell me exactly what I needed to know to successfully construct my cryo-ET system. As a result, my project made me much more comfortable with and skilled at dealing with ambiguity and solving original problems. In addition, my project has improved a wide range of my technical skills. All my work has been on Linux machines, so I have become an advanced Linux administrator. I have also greatly strengthened my Python, HTML, PHP, and MySQL database administration skills. Overall, I have greatly enjoyed solving the dynamic and novel problems posed by my summer research project and plan on continuing during the school year.
Cryo-electron tomography (cryo-ET) is a unique method to reveal macromolecular structures in their native state and context within cells. Biological samples of interest are plunge frozen in a near-native state and imaged at many different tilt angles. The resulting tilt-series of images are used to generate 3-dimensional (3D) tomograms with molecular resolution. When many copies of a structure of interest are imaged, sub-tomogram averaging is used to create even higher, sub-nanometer resolution images to help unravel the arrangement among structural subunits. Cryo-ET has emerged as a unique method to bridge the resolution gap between light microscopy and atomic resolution methods (e.g. crystallography and nuclear magnetic resonance).
Since an electron microscope can take 20-40 tilt-series per day and manual 3D reconstruction of a single tilt-series takes about 30 minutes, it would be extremely beneficial for researchers to be able to continuously collect large amounts of tilt-series data without needing to manually process the tilt-series over several days. In addition, researchers need easy access to previous tomograms since new discoveries in different fields can cause unnoticed structures in existing tomograms to become new targets of interest. As a result, any institution conducting cryo-ET research would benefit from automated tomography data processing in addition to maintaining database access to existing tomograms.
My project over the summer was to create such a high-throughput cryo-ET pipeline and database at the University of Pennsylvania. My constructed system has already enabled cryo-ET researchers to image novel structures and focus on data analysis by automating the entire process from data collection, to data processing, and finally data storage. In addition, the automated real time tilt-series processing has allowed researchers to view their initial 3D tomograms during data collection and accordingly adjust their collection parameters in a timely fashion, optimizing valuable electron microscope time. I also created a web server accessible from anywhere in the world that allows researchers to easily view every tomogram and the associated files they have collected.
Unlike my classes where I am given the needed prerequisite knowledge and already established solutions, my project has required me to solve a large number of novel problems. No one could tell me exactly what I needed to know to successfully construct my cryo-ET system. As a result, my project made me much more comfortable with and skilled at dealing with ambiguity and solving original problems. In addition, my project has improved a wide range of my technical skills. All my work has been on Linux machines, so I have become an advanced Linux administrator. I have also greatly strengthened my Python, HTML, PHP, and MySQL database administration skills. Overall, I have greatly enjoyed solving the dynamic and novel problems posed by my summer research project and plan on continuing during the school year.