Skip to content

higlass/scipy19

Repository files navigation

Demos from our Scipy 2019 talk

Binder

Presentation Video

Slides: slides.pdf or Speakerdeck

Demos:

  • basic-setup.ipynb: Simplest example possible to check that HiGlass loads properly in Jupyter Lab.
  • temperature.ipynb: Introduction to the python API for HiGlass using temperature measurements over the last ~50 years.
  • point-data.ipynb: An example of how the tileset API works and how one can build their own API from scratch.
  • nyc-taxi.ipynb: Demonstrate advanced features using the NYC Taxi dataset.
  • genomics.ipynb: Show how to work with and extend published HiGlass dashboards for collaboration and reproducibility.

Getting started

Requirements

  • Conda v4.6.14 (or higher)
  • FUSE: libfuse v2.9.7 (or higher) or OSXFUSE v3.9.2 (or higher)

Other versions might work too but we only tested the above mentioned versions.

Installation

First, install the environment:

git clone https://github.com/higlass/scipy19
cd scipy19
conda env create -f environment.yml

Next, install HiGlass' jupyter extension:

conda activate higlass-scipy19
jupyter labextension install @jupyter-widgets/[email protected]
jupyter labextension install [email protected]

Finally, start Jupyterlab:

jupyter-lab

Trouble shooting

  • If Conda fails to set up the environment please make sure you're using the latest version of Conda and update if necessary using:

    conda update -n base conda
    
    • If you end up with a half-created environment:
    CondaValueError: prefix already exists: /Users/me/miniconda3/envs/higlass-scipy19
    
    • You may have to delete it before continuing:
    $ rm -rf /Users/me/miniconda3/envs/higlass-scipy19
    
  • If Conda seems to have gotten stuck, first check your processes using top or alike. Conda is dead slow these days and might just need some extra time. In the meantime you can make yourself a nice cup of coffee, go out for lunch, or head to Hawaii for your summer vacation.

  • If starting HiGlass fails and you see an issue related to FUSE telling you a RuntimeError popped up with the following super helpful and elaborate error message: 1, then you most likely need to update FUSE. If HiGlass does start then you can ingore the error. The error goes away if you keep unmounting previously mounted values using umount HttpFs.

References

Weather Data: The data was obtain from the National Centers for Environmental Information. https://www.ncei.noaa.gov/access/search/data-search/global-hourly

Point Data: The data is from Wang et al. Multiplexed imaging of high-density libraries of RNAs with MERFISH and expansion microscopy, Nature Scientific Reports, 2018.

NYC Taxi Data: The data is NYC Taxi & Limousine Commission, https://www1.nyc.gov/site/tlc/about/tlc-trip-record-data.page.