Protein-protein interaction network rewiring across environments

Published: 21 August 2020| Version 2 | DOI: 10.17632/9ygwhk5cs3.2


We have developed to quantitatively assay protein-protein interactions (PPIs) by barcode sequencing (Schlecht et al., Nature Comm. 2017, 8: 1-9). Here, we scale up this technology to quantify changes in relative in vivo PPI abundance of 1.6 million protein pairs across 9 growth conditions, with replication, for a total of 44 million measurements. To the best of our knowledge, this is the first large-scale study to move beyond a “static” view of the protein interactome by examining how PPI networks change across environments. This new dynamic view of a cell’s protein interactome yields several headline findings: (1) Screening multiple conditions discovers ~3-4 times as many PPIs as a single-condition screen, indicating that previous work has underestimated the size of the protein interactome emergent from a cell genome. (2) Most PPIs are found in only a handful of conditions -- we call these PPIs “mutable”-- and these have been underrepresented in single-condition screens. (3) Mutable PPIs populate a newly-discovered and distinct “accessory” module of the protein interactome that is loosely connected, highly dynamic, and enriched for proteins involved in transcription, RNA processing, and translation (4) Mutable PPIs have several features that distinguish them from immutable PPIs: they are less likely to co-express, co-localize, and be explained by simple mass action kinetics, and more likely to contain intrinsically disordered regions, evolve quickly, and be of low abundance in standard conditions. These results settle a long-standing debate concerning the statistical robustness of a “date hub/party hub” dichotomy and its implications for modularity of PPI networks (Han et al., Nature, 2004, 430: 88-93). Taken together, our results suggest that protein interactomes contain previously uncharacterized and highly dynamic regions that reorganize in response to cellular changes, and that this reorganization is due, to a larger extent, by post translational modifications.


Steps to reproduce

Here you can find all the datasets needed to make figures in the manuscript. Refer to Read_me.txt in the folder for the explanation of some important datasets. The scripts for the analyses can be found in this repository: The documentation of the analyses can be found in the manuscript. The output figures can be found in the folder "Figures".


SLAC National Accelerator Laboratory, Stony Brook University


Systems Biology of Microorganism, Protein-Protein Interaction Networks