This section explains the methods for the paper.
We used CPTAC's data on ccrcc, luad, hnscc, lscc, and ucec tissues.
Proteomic data was obtained from "umich" database.
Transcriptomic data was obained from "washu" database.
- Δ_corr
- Transmutation Effects
- Proteomic and Transcriptomic Tables
- Cancer Stages and Correlations
- Set Enrichment
We calculated Spearman correlations between rna and protein abundances for all genes found in the tissues for both healthy and diseased samples. Δ_corr is the the difference between the healthy and diseased correlations. A permutation label-swap test was done for determining the significance of each gene's Δ_corr. A table showing all the Δ_corr and p-values can be found here
The correlation scripts and tables used to create that table can be found in this folder here.
That table is used to produce the figures 2, 4, 5, S2, and S4.
The transmutation effects scripts and tables can be found in the data folder here.
You can find a table with the transmutation data in this csv
This was used to create figure 3.
Proteomic and Transcriptomic data was obtained from CPTAC and the script can be found here.
Proteomic differential expression can be found in this csv
Transcriptomic differential expression can be found in this csv
This was used to create figure 2, S2.
Cancer stages were obtained from CPTAC and the script can be found here.
We colapsed the different stage name into 1-4 stages. (Ex: ["IA", "Stage 1", "Stage 1B", "Stage IB", "Stage IA" or "Stage IA3"] into "Stage I")
We then calculated the mRNA/Correlation in each of the stages and saved it as a csv
This was used to create figure 1.
A complete tumor-normal Correlation script can be found here
The table is saved in this csv
This was used to create figure 4.
The Script calculate the Set Enrichment can be found here.
We used the KEGG source for the enrichment. You can use the script to find tissues with altered pathways found using significant Δ_corr genes.
This script's code is used to generate figure 5.