Caryocar

Caryocar is a Python package for building Species-Collector Networks (SCNs) and Collector CoWorking Networks(CWNs) models from species occurrence data, as introduced in my MSc thesis. SCNs and CWNs extend the social network analytics and can be used for understanding the social structure behind biological collections. This package is built on top of NetworkX.

Supporting documents

• New perspectives on analyzing data from biological collections based on social network analytics [MSc thesis].
• On the social structure behind biological collections [Preprint].
• Package documentation coming soon...

Example Usage

Create a Species-Collector Network (SCN) from a list of collectors and species:

>>> cols=[ ['col1','col2','col3'],
           ['col1','col2'],
           ['col2','col3'],
           ['col4','col5'],
           ['col4'],
           ['col5','col4'] ]

>>> spp=['sp1','sp2','sp3','sp2','sp3','sp2']

>>> scn = SpeciesCollectorsNetwork( species=spp, collectors=cols )

>>> scn.nodes(data=True)
{ 'sp1': {'bipartite': 1, 'count': 1}, 
  'col1': {'bipartite': 0, 'count': 2}, 
  'col2': {'bipartite': 0, 'count': 3}, 
  'col3': {'bipartite': 0, 'count': 2}, 
  'sp2': {'bipartite': 1, 'count': 3}, 
  'sp3': {'bipartite': 1, 'count': 2}, 
  'col4': {'bipartite': 0, 'count': 3}, 
  'col5': {'bipartite': 0, 'count': 2} }

>>> scn.edges(data=True)
[ ('sp1', 'col1', {'count': 1}), 
  ('sp1', 'col2', {'count': 1}), 
  ('sp1', 'col3', {'count': 1}), 
  ('col1', 'sp2', {'count': 1}), 
  ('col2', 'sp2', {'count': 1}), 
  ('col2', 'sp3', {'count': 1}), 
  ('col3', 'sp3', {'count': 1}), 
  ('sp2', 'col4', {'count': 2}), 
  ('sp2', 'col5', {'count': 2}), 
  ('sp3', 'col4', {'count': 1}) ]    

Create a Collector CoWorking Network (CWN) from a list of collector cliques:

	>>> collectors = [ ['a','b','c'], ['d','e'], ['a','c'] ]
>>> cwn = CoworkingNetwork(cliques=collectors)

>>> cwn.nodes(data=True)
{ 'a': {'count': 2}, 
  'b': {'count': 1}, 
  'c': {'count': 2}, 
  'd': {'count': 1}, 
  'e': {'count': 1} }    

>>> cwn.edges(data=True)
[ ('a', 'b', {'count': 1, 'taxons': None, 'weight_hyperbolic': 0.5}), 
  ('a', 'c', {'count': 2, 'taxons': None, 'weight_hyperbolic': 1.5}), 
  ('b', 'c', {'count': 1, 'taxons': None, 'weight_hyperbolic': 0.5}), 
  ('d', 'e', {'count': 1, 'taxons': None, 'weight_hyperbolic': 1.0}) ]

Install

This package is still experimental, and should ideally be run from a conda virtual environment, which is specified in the environment.yml file. In order to create the virtual environment clone this repository, make sure you have conda installed and use one of the following commands, from the root of the repository:

$ conda env create -f environment.yml

Then you should activate it with the following command:

• On Linux:

  $ source activate caryocar

• On Windows:

  $ activate caryocar