Map of Collaborations

An interactive world map of iGEM teams from 2016 is shown below. In this visualization, one may hover over each team’s marker to see the team name in addition to any keywords that appeared frequently on the team’s wiki. This project was made in collaboration with UofCalgary iGEM (see Collaborations page for details).

The process followed was:

  • Scrape all team wikis to extract text (parsing team wiki links from https://igem.org/Team_Wikis?year=2016) using the Beautiful Soup Python library.
    • Crawling was limited to the team’s domain to prevent crawling external websites
  • Tokenize text and identify nouns, using the Natural Language Toolkit (NLTK) library in Python.
  • Use fuzzy string matching (computing the Levenshtein distance between two strings to allow for slight variations/typos) for:
    • Identifying biological terms that appear most frequently on each team’s wiki (a list of biological terms was curated beforehand)
    • On the “Collaborations” pages of wikis, any team names were identified and tracked as collaborations for that specific team
  • The geographical location of teams was obtained by extracting the geojson file from the Google Maps document posted in http://2016.igem.org/Community. These addresses were then fed into a Google geocoder API using GeoPy, in order to obtain the exact latitude and longitude of each team.
  • The data obtained was displayed using the Dash library in Python, powered by Plotly for a visualization of the teams on a world map.

iGEM 2016 Team Collaborations