This repository summarises work for the publication:
Multi-omics and biophysical phosphoproteomics upon BRAF inhibition uncover functional networks of BRAFV600E-driven signaling
Read the paper: \
Dysregulated kinase activity drives oncogenic signaling, disrupts cellular homeostasis, and promotes tumour progression. Although kinase inhibitors have achieved clinical success, the downstream effects of inhibition and mechanisms of resistance remain incompletely understood. The BRAFV600E mutation constitutively activates the MAPK pathway and is a key therapeutic target in melanoma and other cancers, but the functional relevance of most downstream phosphorylation events remains unclear. To address this, a global multi-omic model of BRAF inhibition response was established in BRAFV600E-mutant melanoma cells by integrating time-resolved and biophysical phosphoproteomics, transcriptomics, and thermal proteome profiling. Ultradeep phosphoproteomics revealed extensive phosphorylation changes upon BRAF inhibitor treatment, while biophysical phosphoproteomics identified phosphorylation events linked to altered solubility and subcellular localization, suggesting changes in nucleic acid interactions and nuclear reorganisation. Network-based integration of these datasets prioritised functionally relevant phosphorylation sites and kinases. Experimental validation identified CDK9, CLK3, and TNIK as critical regulators of BRAFV600E signaling and candidate targets for combinatorial inhibition capable of re-sensitising resistant cells. The transcription factor ETV3 emerged as a previously unrecognised effector of BRAF signaling. Biophysical proteomics data confirmed that ETV3 phosphorylation modulates DNA-binding, while functional assays combining knockdown, metabolomics, and drug screening demonstrated its role in coordinating transcriptional and metabolic adaptations to BRAF inhibition. This study provides a systems-level framework linking phosphorylation dynamics to protein function and phenotype, identifies ETV3 as a new node in oncogenic BRAF signaling, and illustrates how integrated, site-resolved models can reveal mechanisms of kinase-driven oncogenesis.
All scripts can be found in code, which contains scripts for the analysis of the different datasets as well as scripts to generate figures and supplementary tables. Fo the python scripts a yaml file to build the required environment is provided.
The data analysis for (phospho)proteomics data starts with psm/protein files which are the output of an MSFragger search (https://msfragger.nesvilab.org0) and can be retrieved from the PRIDE repository of the study (PXD071324).
The psm/protein files of the the different datasets are processed in the respective scripts and standardized result objects for further analysis and plotting are generated and saved for every dataset. These files are provided in results.7z which can be downloaded here (https://oc.embl.de/index.php/s/FiJI0NKFgVB9uDM).
In code code used to generate all data analysis figures, tables and supplementary data is deposited. Code either starts from the processed results or results of the analyses described in theh other scripts. Please reach out in case of questions. Please note that figure color, legends, axis title etc were modified in Inkscape to be included in the final manuscripts, so slight differences arise from that.