Blog

abstract for genegames.org

I wrote this poster abstract up for an upcoming conference and thought it might be useful to share it here.  If it gets accepted, you can come see me (and my iPad stand) in person at USCD in September.  If not, well, you can read it here, play the games over there, and see me virtually anywhere.

genegames.org: High-throughput access to biological knowledge and reasoning through online games 

Games are emerging as a powerful organizational and motivational tactic throughout many areas of society.  Wherever people have a goal that they are having trouble reaching, be it getting their chores done [1], learning all the functions of Microsoft Visual studio [2], or finishing a 10K [3], many are finding success by posing the required tasks as elements of games.  Games can turn small units of work, that alone might seem boring, into fun steps taken towards a meaningful success.  In doing so, they can sometimes dramatically increase individuals’ chances of reaching their objectives.  The process of translating elements of non-game contexts (e.g. most traditional work, learning, exercise, etc.) into aspects of games is now known as ‘gamification’.

Gamification is now being used to meet a variety of scientific goals by serving as an effective way to organize and incentivise large-scale volunteer labor.  The protein-folding game Foldit was the first of a growing wave of applications of games in the context of biological research.  From this well-publicized [4] beginning in protein structure, we now have a variety of biological games about, for example, RNA structure design [5], multiple sequence alignment [6], and neural connectivity mapping [7].  In these games, players help advance scientific objectives by performing tasks that can not be completed successfully by computers alone.

At genegames.org we are exploring the use of games to access the knowledge and reasoning abilities of biologist players.  Through the gene annotation game ‘GenESP’, players can contribute their knowledge of gene function and disease relevance to a new kind of public gene annotation database.  In the ‘COMBO’ game, players help to identify biomarker gene sets that can be used to improve predictions of various complex human phenotypes.  The poster will provide details about the design of the games as well as preliminary results from ongoing experiments.  In addition, the game prototypes will run live during the conference allowing attendees to play and provide the developers with important feedback.

[1] http://www.chorewars.com/
[2] http://channel9.msdn.com/Blogs/C9team/Announcing-Visual-Studio-Achievements
[3] http://blog.triplepointpr.com/the-gamification-of-running
[4] http://genomebiology.com/2011/12/12/135
[5] http://eterna.cmu.edu/
[6] http://phylo.cs.mcgill.ca/
[7] https://eyewire.org/

Social machines at ISMB 2012

This year’s 20th annual conference on Intelligent Systems for Molecular Biology (ISMB) was a busy one for me and the rest of the Su Lab.  As a group of 5, we were responsible for 4 oral presentations, three posters, and the administration of one s…

Data chart plugin version 2.0

The data chart plugin on BioGPS is by far the most popular, and as a result we've been working on some big changes to make it even more useful. Yesterday we deployed these changes in version 2.0 of the data chart plugin, which now includes the following: 1. Much more...

Grant Proposal for your Review

Dear Internet,
A penny for your thoughts on the grant proposal accessible here.  If you were on the committee that decided the fate of this proposal (and in a very real way, its authors), what would you say?  Go jump in a lake?  This is fantastic?  Why?

In a nutshell, we* propose to build some serious games that will translate key biological data curation and interpretation tasks into aspects of entertaining and educational games.  Here are the first few paragraphs to give you a feeling for its flavor.

   “Seven million hours of human labor were spent building the Empire State Building. The Panama Canal took twenty million human-hours to complete. The construction of these monumental structures required the coordinated efforts of tens of thousands of people over many years. By comparison, it has been estimated that nine billion human-hours are spent playing Solitaire annually (Von Ahn 2006. Human Computation). More generally, up to 150 billion hours (the equivalent of 17 million years of human effort) are spent playing games every year (McGonigal 2011. Reality is broken). Obviously people play Solitaire and other games because they are enjoyable and fun. But aside from that enjoyment, the time spent playing these games largely results in no tangible benefit, neither to the individual nor to society at large.

   Here, we propose to build several “games with a purpose”, a class of games that focuses on collaboratively harnessing gamers for productive ends. This proposal will use scientific games to engage a broad community of individuals in advancing biological research. Specifically, our games will focus on creating biological value on three distinct levels. First, we will annotate the function and disease relevance of individual human genes. Second, we will construct a functional gene network that represents consensus biological knowledge. Third, we will discover multi-gene biomarker panels for human disease.
   This proposal will provide a mechanism whereby the entire biological research community and motivated individuals from society at large can participate in improving our understanding of gene function in health and disease. We will use online games as the mechanism to enable and encourage broad community participation. Our games will be based on existing game designs that are proven to be fun and engaging. If we can harness even a fraction of the “Solitaire time” from the gamer community, then already we will have access to a substantial, novel and free resource to advance scientific research. This proposal will test the hypothesis that biological games can substantially improve genome-scale science.
   This objective will be achieved through the following three Specific Aims:

  • Specific Aim #1: Use biological games to annotate human genes. Identification of novel structured gene annotations for diseases and functions will improve many statistical analyses of genetic and genomic data.
  • Specific Aim #2: Use biological games to create a “community consensus network” of gene relationships. The resulting network of biological knowledge will define both the existence of gene-gene edges, as well as the specific nature of each gene-gene relationship.
  • Specific Aim #3: Use biological games to discover biomarkers of human disease. These games will identify molecular features for studying disease mechanisms, diagnosis, prognosis, and treatment selection.”

The rest of the proposal is accessible here. Any feedback would be most appreciated.

* Me and Andrew Su

Thanks for your support

 A couple weeks ago, we posted an open invitation on BioGPS to write a letter in support of our proposal to continue BioGPS development. We also emailed some of our most trusted and loyal uses directly. "I find BioGPS to be the best portal on the...

Subscribe

Enter your email address:

Archives

Categories