#Phylotastic: Day 5, to Infinity and Beyond!

The final day of the Phylotastic hackathon (and the subsequent weeks afterward) passed in a hectic flurry. We pulled off our runway show, and now we're getting ready to ship the prêt-à-porter collection.

From the initial conception of ideas on Day 1, to the stark realizations (and acceptance of more modest goals) on Day 4, all the hard work finally came together on the last day. Our Shiny group churned out a gleaming--and somewhat functional--web interface that wrapped together key tools and server functions from other groups. We even made a Phylotastic module to fit into the Mesquite software package, complete with explanatory video demo.

You can see the final Shiny group web interface here: http://phylotastic.org This links to our demo cases and screencasts, as well as tool interfaces for Reconcilotastic and the Tree topology server (amongst others).

Each group got up and presented their work, and I was astonished at how much we had accomplished in a few short days.

• The Architecture group was charged with building the computational structure and API for Phylotastic tools. After sketching out a nifty figure that determined how different services would fit together, they decided to pursue a Galaxy architecture to make these tools functional. This is a big job, and still a work in progress, but the vision is there. The group even had time to make an explanatory video screencast! 
• The Taxonomic Name Resolution Service (TNRS) group wrapped together existing tools to produce a super-TNRS service for Phylotastic. This tool is currently available in GitHub, as a command-line service, as well as the TNRastic API.
• The Branch Length group was tasked with a "molecular clock" focus, allowing users to access information from fossil calibrations and decorate trees with dates (showing when taxa in the tree diverged from a common ancestor). The end product was a sleek web interface, although there are many details left to pursue.
• The Tree Store group wanted to build a giant warehouse of tree topologies, accessible by user queries and supporting functions such as metadata retrieval and tree pruning. Sounds simple, but this is a massive task. The group continued to wrestle with data formats, API interfaces, and query syntaxes, and put up a basic web interface to be expanded as the project continues.

We all made a long list of deliverables that would ultimately come out of the workshop: manuscripts, a suite of functional servers, oral and poster presentations at upcoming conferences, demonstrations and use case scenarios for users to explore, and even some supercool workshop t-shirts. After the mad morning rush to finish up and present our results, we took a group photo. And then that was it. Workshop over. We all piled into the airport shuttle and divided up to go our separate ways.

But the end of the hackathon only marked the beginning of Phylotastic. The following weeks saw continued progress--different groups took their ideas, tools, and interfaces and ran with them. Teams held follow-up work sessions on June 20th, using the wonderful magic of Google Hangouts. After recovering and stepping back from the intensity of the hackathon, we continued to tweak and clean up the products produced under pressure. The wiki keeps growing, and the #phylotastic Twitter feed remains active.

The hackathon fostered a sense of community, bringing together researchers with similar needs and struggles. Phylotastic wasn't just a one-off event - the organizers are planning subsequent hackathons to expand the conceptual framework developed at NESCent and incorporate other relevant foci. We'll bring other people into this burgeoning community and make the tools even better.

Many of the Phylotastic groups will be presenting their products at iEvoBio, a satellite meeting to the humongous 2012 Evolution conference in Ottowa, Canada. Challenges and workshops are also being organized to refine and improve on existing tools. If you're interested, I strongly urge you to keep tabs on Phylotastic. With help from researchers from multiple disciplines, we can strive towards our end goal: a bustling suite of web tools, accessible and amenable to the needs of the entire scientific community.