Online

Snakemake Hackathon

Snakemake is a common entry tool in the world of computational workflows, especially in Bioinformatics. 

The tool is designed to support you in setting up a workflow system, which opposed to a ”your favorite language here” script, helps you keep track of your processes.

For Snakemake, the user has to define inputs and outputs and the tool will figure out the order of steps and which steps can be run in parallel. 

This Hackathon is open for all, free of charge and independent of field of Science and it is all about getting your own workflows to the supercomputer (Puhti, Mahti, Lumi or other).

Content

We will offer short introduction to snakemake and a supercomputing concepts refresher session in the beginning, which you can attend if needed.

Then we will talk about opportunities and challenges with moving workflows from laptop to the supercomputer and give tips on how to go about it. 

The afternoon is reserved for your own work. You can either work on your own ”Snakefile”, moving your workflow from laptop to supercomputer or work on our toy example, to help understand the concepts better before moving to your own work. 

Prerequisites

Tentative schedule

All times in EEST (Helsinki time zone) and including breaks

10-10.30     Supercomputing concept refresher
10.30-11.30   Intro to Snakemake
11.30-12   Snakemake on HPC
12-13   Lunch break
13-15   Hackathon

Lecturers and support

Antoni Gołoś (CSC), Laxmana Yetukuri (CSC), Samantha Wittke (CSC), Teemu Ruokolainen (Aalto University – Aalto Scientific Computing), Yonglei Wang (ENCCS), Diana Iusan (UPPMAX, NAISS), Johan Hellsvik (PDC, NAISS) 

Expectations for the Hackathon

The hackathon part of this event is about moving your own workflow to the supercomputer. You can work on your own projects and specialists will try to help you if you run into issues or have any questions. We will provide a dummy example for those who don’t have any own project to work on during the hackathon.

Learning outcomes:

  • What is snakemake
  • When does it make sense to use Snakemake, when not
  • Implementation of workflow with Snakemake
  • Moving own workflow to the supercomputer