Multi-omics integration of malignant peripheral nerve sheath tumors (MPNST) identifies potential targets based on chromosome 8q (chr8q) status
These scripts are designed to be run after those in the proteomics directory, where the primary proteomics ws done.
The analysis for this manuscript integrates copy number and RNASeq measurements of diverse MPNST together with proteomics measurements. The proteomics analysis is stores in the proteomics directory and stores the files on Synapse at http://synapse.org/chr8_mpnst. Once you request access to the data files you can reproduce the analysis in the paper below.
We first calculate the chr8q copy numbers and correlations across molecules. NOTE: always return to this directory as a working directory after every script, as there are many subdirectories created that can make things very confusing.
#end to end analysis runs and stores lots of files
source('00_computeChr8Expression.R')
source("01_enrichments_copynumber.R")
source('02_enrichments_protRNA.R')
source('03_DMEA_correlations.R')
source('04_networkAnalysis.R') #this requires Cytoscape to be open to run
source("05_tableCompilation.R")
This generates the data and tables needed for the figures. Below is a breakdown of the scripts, scroll down for figure instructions
The first step to analyzing the data is to calculate the copy number of chr8 across the samples. This is currently run by the 00_computeChrCorrelations.R script.
This script downloads the relevant data to the environment, creates a folder with today's date and saves the copy number data. It also creates a folder with Supplemental tables 1 and 2.
We then calculate positional enrichment to ensure that chromosome 8 activity is indeed altered in the samples in 01_enrichments_copynumber.R. This will save files to the working directory with the current date.
The next step is to compute the correlations between copy number computed in the previous step and mRNA/protein levels to identify functional pathway enrichment. This is calculated by 02_enrichments_protRNA.R.
The next step is to carry out the drug correlations used for Figure 3, which are slightly different than the mechanism enrichment analyses. There's a chance that the DMEA is run more than once when we run 03_DMEA_correlations.R
We then used the stored files to build the networks used in Figures 2 and 3. This is 04_networkAnalysis.R. If you want to use updated data, then you must update the three synapse IDs that pull the most recent data.
This requires that you run figure 3 code first, but will collect the files in the local directory so that they can be supplemental tables 05_tableCompilation.R
Now that the calculations have been made we can generate the figure panels in order through the following scripts. The scripts assume that the above scripts have been run and uploaded to synapse. However, if you run the following scripts they will refer to previously run synapse files (currently from the 2026-01-22 run of the code).
To build the figures in a figures directory, first start with
Figure1_correlations.R. This requires 5 files from synapse
generated from above, so you will need to update it if you re-ran the analysis
To visualize the enrichment analysis run Figure2_enrichment_centrality.R
Next run Figure3_drugCors.R Creates the figures for Figure 3.
FigureS1_histograms.R
Kinase substrates are plotted here: FigureS3_kinase.R
Transcription factor targets are plotted here but we don't reference in paper: FigureS4_TF.R
There are many files that might be used later that are uploaded to synapse using the 06_save_to_synapse.R script. It is designed to take the current working directory and run immediately after the first three scripts. NOTE: you will require upload access to this repository.
The following files were not validated but also run parts of the analysis.
- drugScreening.R: only used for IC50 t-tests
- fit_curve.py: calculates area under the curve values; not currently used; adapted from: https://github.com/PNNL-CompBio/coderdata/blob/main/coderbuild/utils/fit_curve.py