codespaces_RStudio

Reproducible teaching labs for RNA-seq and R programming built around GitHub Codespaces. This repo ships a full RStudio Server IDE, an optional desktop-lite workflow, and a mamba-managed toolkit so every student lands in the same environment regardless of their laptop.

Launch a brand-new Codespace

1. Click Open in GitHub Codespaces. GitHub provisions a Codespace from .devcontainer/devcontainer.json.
2. Initial creation installs Ubuntu packages, R/Bioconductor libraries, and CLI tools�plan for 30 minutes. Keep the tab open until the build completes.
3. Once VS Code opens, run:

   rstudio-server start

4. Port 8787 auto-forwards. Use the notification or Ports tab to open RStudio (user rstudio, password rstudio).

Resume an existing Codespace

1. Click Reopen My Codespace button on top and pick the workspace you created earlier.
2. Because packages are already cached, it resumes in seconds.
3. Restart RStudio with rstudio-server start and reopen port 8787.

RStudio IDE overview

A screenshot annotated with these callouts will be inserted here.

• Source edit R scripts/Rmd notebooks, run chunks, and view rendered docs.
• Console executes commands interactively; shows warnings/errors and rstudio-server messages.
• Terminal full shell access (bash) for mamba commands, git, or custom scripts.
• Environment lists objects currently in memory (data frames, vectors, models); lets you inspect or remove them.
• History chronological log of executed commands with buttons to re-run or send to Source.
• Files browse project folders, upload/download files, and open notebooks.
• Plots displays the current graphics device; use navigation arrows to flip through prior plots.
• Packages see installed libraries, enable/disable them, and launch help pages.

Quick start in RStudio

1. Open the .Rmd you want (see Lessons below) via the Files pane.
2. Set the working directory with setwd() or the gear icon if needed.
3. Use Run / Run All buttons to execute chunks, or press Ctrl+Shift+Enter (Cmd+Shift+Enter on macOS) to knit.
4. Keep an eye on the Console for errors; restart R (Session -> Restart R) if memory gets cluttered.

Lessons in practicals/

1. R_Programming_basics_practical.Rmd introduces R syntax, data frames, and visualization fundamentals tailored for bioinformatics students.
2. Gene_expression_DEG_practical.Rmd walks through differential gene expression analysis with DESeq2, including QC, normalization, and volcano plots.
3. DEXSeq_salmon_DTU_practical.Rmd covers isoform-level quantification with Salmon and DEXSeq for differential transcript usage.

What you get

• RStudio Server 2023.x exposed on port 8787 via the Rocker devcontainer feature.
• XFCE desktop + noVNC bridge (desktop-lite feature) so you can run GUI apps inside the Codespace when enabled.
• Conda/mamba toolchain (Miniforge feature) for FastQC, MultiQC, Salmon, STAR, samtools, seqtk, bedtools, etc. (env/mamba-environment.yml).
• Pre-installed R/Bioconductor stack covering all supplied practicals (DESeq2, airway, IsoformSwitchAnalyzeR, clusterProfiler, ...).
• Ready-to-teach content: Rmd notebooks, docs, and data mirrored from the original Gitpod-based course.

Repository layout

• docs/ grading material and supporting markdown.
• env/ shared config such as mamba-environment.yml and R data sets.
• figures/ reference screenshots.
• rocker/ legacy Dockerfiles.
• *.Rmd the teaching notebooks students run in RStudio.

Pre-installed R packages

The post-create hook runs .devcontainer/scripts/install-r-packages.R, ensuring:

• airway, SummarizedExperiment, DESeq2, vsn, apeglm, AnnotationDbi, org.Hs.eg.db
• clusterProfiler, enrichplot, DOSE, IsoformSwitchAnalyzeR
• tidyverse, ggplot2, ggnewscale, ggbeeswarm, ggrepel, pheatmap, RColorBrewer, dbplyr, tidyr, readr, GGally Add more via that script as needed.

Local VS Code / Dev Container usage

# Requires Docker + VS Code Dev Containers extension
git clone https://github.com/FairTeach/codespaces_RStudio.git
cd codespaces_RStudio
code .
# When prompted, "Reopen in Container"

Same devcontainer features run locally, so behavior matches Codespaces.

Managing CLI dependencies with mamba

Miniforge installs mamba under /opt/conda. The bootstrap script creates/updates rnaseq-tools defined in env/mamba-environment.yml:

mamba activate rnaseq-tools
multiqc --help

Add extra tools (e.g., gffcompare, bwa, hisat2) to the YAML and rebuild or run mamba env update inside the Codespace.

Troubleshooting tips

• Rebuild the container (Codespaces / Rebuild container) after changing .devcontainer/**, env/mamba-environment.yml, or bootstrap scripts.
• If RStudio Server doesn�t start, check the Ports tab for 8787; restart with rstudio-server stop / rstudio-server start.
• Desktop-lite can be toggled via the command palette (Codespaces: Open in Browser) if you need the VNC desktop.

¿Happy learning? Questions or suggestions? Open an issue or reach out on @FairTeach!

---

🧬 Contact and Support

Professor: Igor Ruiz de los Mozos Aliaga PhD Institution: NASERTIC / UPNA / UNAV
Office hours: By appointment
Facilities: IRIS Polo de Innovación Digital – Molecular Biology, Synthetic Biology, and HPC laboratories

---