1. produce a comprehensive taxonomy
2. enable curation of phylogenies
3. create a supertree - the "synthetic tree"
4. enable users to browse the synthetic tree and "graph of life"

• provided study curation user interface
• primary study database

Phylografter Sortable, Searchable Study Dashboard:

Phylografter OTU Mapping:

Phylografter Tree Editing:

Synthetic tree browsing:

Synthetic tree commenting:

• Command-line tool written/run by Jonathan Rees
• Rule-based combination of:
  • taxonomic info. from NCBI, GBIF, IRMNG, Index Fungorum, SILVA
  • taxonomic edits by curators
• Produces a versioned Open Tree Taxonomy (OTT)

• Takes OTT as input
• Provides name-to-OTT-ID service and other taxonomic APIs
• neo4j backend
• written by Cody Hinchliff and others in Smith lab.

• Front-end for curation of phylogenetic estimates
• Started by Ree lab before the Open Tree project, but significantly extended in years 1 and 2.
• imports OTT and trees into a relational database
• provides views on the trees, OTU mapping, and some conflict summaries
• can export studies to NexSON

• git (version control system) repository of phylografter exports
• a script can pull down the newly-updated studies from phylografter.
• used as the input for treemachine
• decouples phylografter and treemachine

• reads OTT as a tree and trees from NexSON
• reads a list of study ranking to be used when resolving conflicts.
• creates a "graph of life" of all of the input trees in a neo4j database
• implements a novel supertree algorithm for selecting which branches in the "synthetic tree of life"
• treemachine plugin provides web services to explore the tree.

• JavaScript front end (runs in the user's web-browser) for the synthetic tree
• uses taxomachine and treemachine services
• enables
  • navigation of the tree
  • commenting
  • searching for taxa in the tree

• Entire corpus of studies in a git repository
• a python app. provides web-services for data access
• "satellite" services (e.g. our study indexing tool oti) via:
  • sync'-ing using git
  • web-hooks on GitHub
• Planned support for letting the community submit edits via git "pull requests"

• JavaScript using taxomachine, phylesystem API, oti
• more flexible OTU mapping
• easier tree operations (rerooting, ingroup specification)
• displays multiple perspectives on the study
• still undergoing initial tests

• 3 primary data-backed services:
  • taxonomy (taxomachine serving OTT)
  • input trees (phylografter, phylesystem API)
  • graph of Life (treemachine)
• 2 primary user-facing web applications:
  • synthetic tree browsing
  • new curation

• current architecture -> all substantive calls are already done through APIs
• we need to document APIs much better
• authentication via GitHub (we don't control a user database)
• hackathon planned for Sept. 2014

• Free and Open code licenses (of course)
• very little "behind the scenes":
  • git repo as study database
  • taxonomy always available as archive
  • every part of deployment (except private keys) publicly visible
  • open mailing lists, trello board, google docs, issue trackers and IRC