douglasdrake / myheroku

Heroku app for a visualization dashboard

myheroku

Description

Repo for a dashboard to explore the data described in Belly Button Biodiversity DataSet with graphs built using Plot.ly.

Methods

1. Flask is used for the GET/POST methods.
2. SQLAlchemy is used to connet to and query the database of samples.
3. JavaScript (D3) is used to respond to menu changes and update the contents of the dashboard.
4. Plot.ly is used to generate the plots.

Notes:

• A unique color is assigned to each otu_id using the function newMakeColorScale() in the file color-scheme.js. This function interpolates 3666 values from the hsl(hue, saturation, lightness) color model. This allows the colors of the different otu_ids to remain constant between the plots and between the samples.
• If there are no observations recorded for a sample (as in sample # 1495), the pie chart and bubble plot will be empty. This is a result of using let declarations instead of var statements in the functions in app.js.
• The dashboard is deployed as an app on Heroku