In this project, I have built an interactive dashboard to explore the Belly Button Biodiversity dataset, which catalogs the microbes that colonize human navels.
The dataset reveals that a small handful of microbial species (also called operational taxonomic units, or OTUs, in the study) were present in more than 70% of people, while the rest were relatively rare.
I took following steps to create the dashboard.
-
Use the D3 library to read in
samples.json
. -
Create a horizontal bar chart with a dropdown menu to display the top 10 OTUs found in that individual.
-
Use
sample_values
as the values for the bar chart. -
Use
otu_ids
as the labels for the bar chart. -
Use
otu_labels
as the hovertext for the chart.
- Create a bubble chart that displays each sample.
-
Use
otu_ids
for the x values. -
Use
sample_values
for the y values. -
Use
sample_values
for the marker size. -
Use
otu_ids
for the marker colors. -
Use
otu_labels
for the text values.
-
Display the sample metadata, i.e., an individual's demographic information.
-
Display each key-value pair from the metadata JSON object on the page.
-
Update all of the plots any time that a new sample is selected.
Here are a couple of screenshots of the dashboard for 2 different subject ids, 940 and 1294 respectively.
To load the dashboard, use the following steps.
- Run a simple HTTP server with python. Open gitbash and type python -m http.server
- Open your web broswer and go to http://localhost:8000/.
The apps is deployed to a free static page hosting service at Github pages. Please click here.
Additional reference materials:
Best-README-Template Retrieved from: https://github.com/othneildrew/Best-README-Template
Refer to the Plotly.js documentation when building the plots.
Hulcr, J. et al.(2012) A Jungle in There: Bacteria in Belly Buttons are Highly Diverse, but Predictable. Retrieved from: http://robdunnlab.com/projects/belly-button-biodiversity/results-and-data/