pca-pwa

pca-pwa, a simplified manner for insights and decision-making by visualizing complex relationships with PCA web application.

The Purpose of the Package

• The purpose of the package is to offer a simple way of visualizing relatationships between items of any given dataset. The user could easily obtain a pca plot without needing to configure or compile the application.

Installation

To install pca_pwa, you can use pip. Open your terminal and run:

pip install pca_pwa

Open IPython or Jupyter Notebook

>>> from pca_pwa import app
>>> app.app.run(debug=True, use_reloader=True, host='0.0.0.0', port=8082)
>>> # * Serving Flask app 'app'
>>> # * Debug mode: on
>>> # * Running on http://127.0.0.1:8082

Open the url: http://127.0.0.1:8082

Upload xslx/slx file (Excel)

• e.g.:
  • Click here to download the excel file

    • Items/Observations should be in rows

    • Variables/Features should in columns

      • Standard Data (table) Format

      The example of standard data format to be used while uploading to pca-pwa web app is a dataframe from sample names in the first column, and the rest (e.g.: metabolites, genes, RNA, etc.) for each sample in the following columns (see Table 1).

      Table 1: Standard data table format.

      Sample	Met 1	Met 2	Met 3	...	Met N
      S1	99,380	10.177	51.484	...	71.882
      S2	101.195	10.786	50.446	...	73.318
      S3	102.165	9,375	49.668	...	72,056
      S4	99.481	8.291	48.111	...	73.282
      S5	101.282	10.867	50.209	...	73,572
      S6	99.43	9.95	47.602	...	71,983

Choose a method of imputation for missing values.

Then run the pca by clicking Perform PCA button.

---

Otherwise you can use git clone:

Here is the Usage:

Clone the github repository

git clone https://github.com/danymukesha/pca-pwa.git

Run the app

cd pca-pwa
python3.1 pca-pwa/app.y

# * Serving Flask app 'app'
# * Debug mode: on
# * Running on http://127.0.0.1:8082

Open the url: http://127.0.0.1:8082

License

This project is licensed under the MIT License.

Credits

Author: MIT © Dany Mukesha

Email: [email protected]

Thank you for using pca_pwa!