FASTGenomics/hello_genomics_viz_py36

Name: hello_genomics_viz_py36

Owner: FASTGenomics

Description: Sample Python visualization

Created: 2017-10-11 13:42:05.0

Updated: 2017-10-13 11:14:38.0

Pushed: 2018-01-04 14:23:54.0

Homepage:

Size: 686

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

o      O        o  o              .oOOOo.
O      o       O  O              .O     o                              o
o      O       o  o              o
OoOooOOo       O  O              O
o      O .oOo. o  o  .oOo.       O   .oOOo .oOo. 'OoOo. .oOo. `oOOoOO. O  .oOo  .oOo
O      o OooO' O  O  O   o       o.      O OooO'  o   O O   o  O  o  o o  O     `Ooo.
o      o O     o  o  o   O        O.    oO O      O   o o   O  o  O  O O  o         O
o      O `OoO' Oo Oo `OoO'         `OooO'  `OoO'  o   O `OoO'  O  o  o o' `OoO' `OoO'

---

How to write a FASTGenomics App

Writing an visualization is easy - there are some conventions you need to know, but otherwise you are free to use the tools you like, e.g. Python, R, Plotly.

This document explains the basic concepts using our “Hello Genomics” visualization app as an example.

TL;DR

• Visualizations have a lot in common with our Hello Genomics Calculation
• A visualization serves a website on port 8000
• Visualizations use our Input/Output conventions, just use our fastgenomics-py module
• Don't write files during runtime (not even a summary)

Core concepts

In FASTGenomics visualizations are realized by visualization apps. These apps generate a website (HTML, Javascript), which is shown as part of the analysis. Hence you can use the full power of the web and equip your visualization with every element you can imagine.

In our example visualization, we use plotly.js as open source JavaScript graphing library together with Flask for serving the data.

How to start

Use this example as a skeleton and follow the instructions of our example calculation Hello Genomics Calculation but ignore any output and summary.

Best practice

• Create small docker images (< 500 MB)
• Pay attention to startup times: A visualization should not perform any computation-intensive tasks and start within milliseconds to a few seconds. If you need to perform computation-intensive tasks anyway, please inform the user to be patient or consider creating a separate calculation app.
• Ship all static files your backend needs (like plotly.js, bootstrap, etc.) as you won't have access to the internet.
• Ship and respect licences
• Don't use absolute urls like “http://localhost:8000/jquery/version-1.js”. Since your app will rune inside our own Website, you would get our Jquery Libary or nothing, not the version you shipped. Pleaseuse relative URLs like “/jquery/version-1.js”
• Expect that your App will run using a non-Root user. So you cannot use Port 80 or similar. Develop your app using a non-root user, e.g. guest.

What are the differences to Calculations?

In contrast to our Hello Genomics Calculation you don't have to write a summary nor any output-file. In fact you should not even write any file except for temporary files. Furthermore serve the website on port 8000 and test your application by starting it via docker-compose -f <my_docker_compose> up and navigate to http://localhost:8000 to see the your visualization.

Miscellaneous

Heath check

Docker provides a mechanism called healthcheck which allows you to check if a container is still working. In our example we serve two websites:

• http://localhost:8000/: serves the visualization itself
• http://localhost:8000/health: serves the heath-status “OK” or raises an HTTP 503 error.

The latter can be used together with the command line tool curl to implement a Docker healthcheck as follows:

Dockerfile:

THCHECK CMD curl --fail http://localhost:8000/health || exit 1

Restrictions

Currently we do not support WebSockets as bokeh does. Please let us know if you want this restriction removed.

Licence

This visualization is licenced under the MIT licence and uses plotly.js v1.30.0, licensed under the MIT license.