We would like to construct a quantitative, context-dependent scaffold for all interactions in the human proteome. Using a combination of computational and experimental techniques, we have a model for efficiently targeting important protein nodes for interaction mapping, beginning with an initial computational assembly from extant data and moving towards a high resolution map with embedded cell context-dependent information.

We propose to leverage recent advances in the Proteomics Platform for reproducibly executing protein-protein interaction experiments using ready-made genetic reagents and laboratory automation (“draft-level interactomics”) to:

  • Obtain a complete wiring diagram of protein-protein interactions in human cells: Using molecular biology, quantitative mass spectrometry, and computational techniques, we will strive to develop a “complete” scaffold of protein-protein interactions.
  • Learn the rules of how cellular context and type affect protein interactions so that context-dependent maps can be imputed.
  • Study protein interaction dynamics in perturbed systems: Using knowledge of the wiring diagram, we will make predictions about interaction dependencies and stabilities in the setting of chemical and genetic perturbations (including the presence of disease alleles), and go on to test them using next-generation quantitative mass spectrometry techniques.

Our overarching goal is to produce a series of statements of the form: Protein A interacts with Protein B with probability (or strength) P in context X