Map of Collaborations Among Primatologists

When I was in the field with Kenny Chiou, we started talking about doing an updated web-based genealogy of primatologists. That project is still on the horizon, but in the meantime here's a map I made of collaborations between authors in the American Journal of Primatology. Click and drag to pan and scroll wheel to zoom, or use the controls in the corner. There's a button for full-screen mode in the corner, too. Larger, red nodes have more unique collaborators. Try to find your favorite primatologist!

If you'd rather, there's also a PDF available for download. I was able to find myself:

There are unfortunately a lot of nodes hiding behind other nodes, so some people may be obscured. Below is a map with the nodes adjusted to have no overlap. There's also a PDF available, at higher resolution.

Technical Notes

1,362 article entries were downloaded from NCBI PubMed in XML format. A search query was used to limit the results to those that matched "Am J Primatol". A Perl script using XML::TreeBuilder pulled out the authors that wrote papers together. Here's a code snippet:


use XML::TreeBuilder;
use List::Util qw(first);

my $file = 'pubmed_ajp.xml';
my $tree = XML::TreeBuilder->new();

$tree->parse_file($file);

my @allGroups;

foreach my $auth_list ($tree->find_by_tag_name('AuthorList')){

   my @authGroup = ();
	
   foreach my $auth ($auth_list->find_by_tag_name('Author')) {
      my $name = '';
      if ($auth->find_by_tag_name('LastName')) {
         $name = $auth->find_by_tag_name('LastName')->as_text;
         $name .= ', ';
         $name .= $auth->find_by_tag_name('Initials')->as_text;
      }

      push @authGroup, $name;
   }
	
   push @allGroups, [ @authGroup ];
}

The code later writes the node and edge data into GEXF format (Graph Exchange XML Format), an XML extension for describing complex networks structures with associated data. You can download the GEXF file here.

I then used the program Gephi to lay things out nice (using a force-based algorithm) and to color the nodes based on number of edges connecting them. The data file is also available to download, though you'll need the program to view it.

Finally, the graph was exported for use in a web plugin using the SeaDragon Web Export plugin for Gephi. The result is above, and here is a screenshot of the zoomed-out graph, in case it's not working in your browser.

The small clusters and singletons were removed, but on the whole, we're a pretty interconnected bunch.