Code Repository for the Data Analyses of Co-Transcriptional Pseudouridylation
In order to succesfully execute the code in this repository, the following software and packages are needed:
- R >= 4.1.0
- R packages (available via Bioconductor):
- Rsamtools
- GenomicAlignments
- GenomicFeatures
- GenomeInfoDb
- rtracklayer
- data.table
- ggpubr
- cowplot
- ggsci
- scales
- ggthemes
- ggrepel
- ggVennDiagram
- viridis
- tximport
- DESeq2
- IHW
- org.Hs.eg.db
- UpSetR
- GGally
- outliers
- [OPTIONAL] deepTools
- [OPTIONAL] SparK
Dowload or clone the repository and move to the working directory:
cd Dyskerin_manuscript
To install Miniconda, a free minimal installer for conda, follow these instructions.
Assuming that conda is available on the system, use the environment.yml
file to generate a minimal working environment, comprising R and some basic packages
conda env create -f environment.yml
The installation of the additional pacakges has to be perfomed from within the newly generated environment:
conda activate dyskerin_manuscript
Open an R session:
R
Install Bioconductor
if (!require("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install(version = "3.16")
Install missing packages
BiocManager::install(c("GenomicFeatures", "GenomeInfoDb", "org.Hs.eg.db", "tximport", "DESeq2", "IHW", "apeglm"))
Download the Gencode v27 annotation file (GFF3
format), rename it gencode.v27.annotation.gff3.gz
, and place it into data/Annotation
.
Generate the RepeatMasker annotation (TSV
format) by using the TableBrowser, rename it hg38_repeatMasker.tsv.gz
, and place it into data/Annotation
.
Some of the files that are employed by the script were too large to be uploaded on this repository and, therefore, they must be either retrieved from the GEO repository (GSE211202) or generated (e.g. mapped BAM
files, normalised BigWig
coverage files, etc.) before they can be read into R or deepTools.
NOTE: The code can be executed even without BAM files, but some steps might not return the expected output.
In the main R markdown file, which contains the procedure on the majority of the analysis and visualisations, there are some headers marked as USER_ACTION; these define sections/statements that have to be changed/modified by the user in order to locate and import the files required for these steps. Briefly, these include the RepeatMasker annotation, the Salmon quantification paths, and the BAM files for the different datasets. For convenience, the rosetta table file, which contains labels and paths to the required BAM files, is read into R; so, make sure to modify it according to your setup.
All genomic coverage profiles were generated using SparK (see createGenomicPlots.sh script for procedure).
Genome coverage normalisations and genomic metadata matrices were generated using deepTools (see createNormalisedTracks.sh and createMatrixPlots.sh scripts for procedure). The metadata matrices were imported into R and plotted.