Welcome to Choragraph, a deep-learning based approach to the spatial mapping and analysis of subcellular proteomics. Choragraph uses an ensemble of deep neural network models to provide context-dependent reconstruction of missing proteomic values, and prediction of a protein's subcellular localisation in a manner that is innately aware of multi-localisation.

Use the table on the left to find and select proteins of interest and the 2D map on the right to interact with compartmental classifications in the context of a whole dataset. Three datasets are currently available for Arabidposis thaliana.

Choragraph was created by Harriet T. Parsons (Quadram Institute) and Tim J. Stevens (MRC Laboratory of Molecular Biology; tstevens@mrclmb.ac.uk)