DeepScalePPIs · Massively scaling-up protein-protein interaction mutagenesis.

Understanding, predicting and engineering protein-protein interaction (PPI) affinity and specificity is a key problem in molecular biology. In particular, we do not know how sequence-to-affinity encoding works at a quantitative and mechanistic level and we cannot predict and engineer the effects of mutations on PPIs. To tackle this problem, we need to generate large amounts of quantitative biophysical measurements. Until now this was not possible due to the low throughput of classic biophysics techniques. Here we propose to expand high-throughput deep mutational scanning (DMS) methodology to quantify the effects of tens of millions of mutations on hundreds of PPIs. To generate a dataset of sufficient size and diversity to train models to predict changes in binding affinity from sequence, we will perform combinatorial trans mutagenesis by combining single mutants in the interaction interface of each partner of a given PPI and quantify the change in binding using a highly-validated protein complementation assay (PCA). To establish the platform, we will first focus on mutating PPIs of the KRAS oncoprotein. We will then scale up this approach to mutate at least one hundred structurally diverse PPIs. The resulting dataset will be of unprecedented size and standardisation, serving as a landmark reference dataset for us and others to train and evaluate models to predict the effects of mutations on interaction affinity. These data and models should transform our ability to understand clinical variants affecting PPIs and allow us to better understand, drug and engineer PPIs critical for medicine and biotechnology.“DeepScalePPIs” will be supervised by Prof. Ben Lehner (CRG, Barcelona), whose group has pioneered the use of deep mutagenesis to understand protein structures, dynamics and interactions. In this unique environment I will be able to learn and combine knowledge from different fields s as genomics, systems biology, biophysics and computational biology.