Snakemake is a flexible workflow engine to build reproducible and scalable bioinformatics data analyses. Workflows are described in Makefile-like syntax and support custom Python functions. In this talk, I will share my experience as a researcher on how to use Snakemake to put together a data processing pipeline easily and scale up the pipeline to large number of samples on multiple computing platforms (e.g., local server, HPC, and cloud). I will also talk about lessons/best practices I learned while adopting Snakemake to my data processing and analyses.
Snakemake 是一個靈活易用的 workflow engine，用來建立一個可再現、可擴展的生物資訊分析流程。它的 workflow 近似 Makefile 語法，並能在流程中呼叫自訂的 Python 函式。我會分享以我一個研究員的經驗，如何用 Snakemake 改寫我研究用的資料處理流程來處理大量的樣本，並且把同個流程跑在不同環境（本地伺服器、HPC、雲端）。我也會分享我在資料處理、資料分析研究時使用 Snakemake 所學到的心得與技巧。
First, set up the node.js environment:
npm install # Install the dependencies npm start # Start live reloading by browsersync npm test # Run CSS style check using stylelint To publish the bundled presentation to GitHub Pages, run:
npm run bundle # Create standalone bundle npm run publish # Push the bundle to GitHub Pages This repo also sets up the GitHub workflow that will automatically bundle and publish the presentations on every git push. The workflow produces a bundled presentation as a compressed file slides.zip.
The bundled presentation contains a standalone webpage that can be viewed offline. Simply open the file index.html in a web browser to start the presentation and view the notes.
The theme is shared under MIT license. It directly modifies the official Material theme (the original license) and will be in sync with upstream updates.
The theme builds on the following packages and resources:
- Shower's official theme, Material, MIT License
- Shower: HTML5 presentation framework by Vadim Makeev et al., MIT license
- highlight.js: Code highlighting library by Ivan Sagalaev et al., MIT license
- Fonts:
- Source Sans, SIL Open Font License 1.1
- Fira Code, SIL Open Font License 1.1