Programmed_readthrough

Programmed translational readthrough produces C-terminally extended protein isoforms via decoding of stop codons by near-cognate tRNAs. Here we use deep mutational scanning to quantify ~5,000 sequence variants for four examples of human readthrough in the genes AQP4, MAPK10, OPRK1 and OPRL1.

Project Workflow and Execution

This project consists of a sequential pipeline. To ensure reproducibility and simplify execution, you can run the entire workflow with a single command.

Prerequisites

1. Clone the repository:

git clone https://github.com/lehner-lab/Programmed_readthrough.git
cd Programmed_readthrough

2. Create and activate the Conda environment:

conda env create -f environment.yml
conda activate Programmed_readthrough

How to Run the Pipeline

After setting up the environment and placing the data in the correct directories, you can run the entire pipeline using the Run_all.R script:

Rscript Run_all.R

Running files individually

Files can also be run one at a time, but the order must be preserved and all commands must run in the same R session: each file creates objects reused by the files after it. The 01–07 prefixes indicate the intended order.

From the repository root:

conda activate Programmed_readthrough
R

Then, in that same R session:

dir.create("html_reports", showWarnings = FALSE)
rmarkdown::render("01_Setup.Rmd",           envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("02_Data_processing.Rmd", envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("03_Fig1.Rmd",            envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("04_Fig2.Rmd",            envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("05_Fig3.Rmd",            envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("06_Fig4.Rmd",            envir = globalenv(), output_dir = "html_reports")
rmarkdown::render("07_Fig5.Rmd",            envir = globalenv(), output_dir = "html_reports")

envir = globalenv() makes each file evaluate in the shared workspace, so its objects remain available to the next one. Do not quit R in between.

Exploring the code in RStudio

For interactive exploration we recommend opening the .Rmd files in RStudio and run chunks in order. Note that RStudio's Knit option will not work for files 02–07: it starts a separate R process, so objects created by earlier files are not visible to it.