We are developing sensitive methods to detect subtle differences between protein structures, by leveraging large datasets and expanding iCn3D distance functions analysis feature. As a result, we may identify different interactions which can cause difference in structures. Recognizing different structures will allow us to dig deeper and determine mechanisms that cause the difference.
In the iCn3D command line, input their structure name and defined sets command for distance set analysis. Users may use proteinset_command.py to generate this command. Next, in iCn3D, select Analysis > Distance > Among Many Sets and save the generated table as an HTML file. The HTML file can then be run on [INSERT SCRIPT NAME] that will parse the HTML file information, gather distances for all possible combinations of sets, and compile output to a table mapping structure identification to distance vector. The output table should be saved as a CSV file and run cluster.py to cluster data and generate a dendrogram.
- Evaluate experimental structures in reference to AlphaFold structures
- Understand variations in protein conformation within families
- Analysis of molecular dynamics trajectories
- Discern subtle conformational differences eliduded by large dataset analysis
- Create an interactive Dendrogram and intergrate the figure to iCn3D
- Expand analysis parameters to handle arbitrary alignments such as different chain identifers and different residue numbering between structures
- Tom Madej (Team Lead),[email protected]
- David Bell, [email protected]
- Chase Freschlin, [email protected]
- Gabby Vento, [email protected]
- Lianna Khachikyan,[email protected]