DoubleTwist.com

home	CV	personal	research	other

DoubleTwist.com: Applying software agents
to biocomputational analysis

What The DoubleTwist.com system explored the application of software agents to the following issues in biocomputational analysis:

reducing the user-complexity of biocomputational tools - Because they operate with a lot of background, agents require very little user input to operate.
increasing the understandability of biocomputational results to non-experts - Agents can post-process output extensively before presenting it, and can do so according to the level of user sophistication.
resolving the data freshness problem: Given that the content of databases is constantly evolving, how can one ensure that results remain correct? - Agents automatically repeat analyses following database update and flag changed results.

New York Times

story

Example screenshots

Agent assignment (specifying a given agent to work on a given sequence)
Agent results (global view)
Agent results (detail view)

Where/When DoubleTwist Inc., 1999-2002.

References

Ling, B.X., Pouliot, Y., Kiruluta, E., and Tsinoremas, N. (2000). "Interpreting working draft human genome sequences using DoubleTwist.com. 13th Annual Genome Sequencing & Biology Conference, Cold Spring Harbor, NY.

Pouliot, Y., Felkins, K., Bernstein, J., Rule, J., Kiruluta, E. and Mader, C. "Method and Apparatus for Simplified Research of Multiple Dynamic Databases", patent submitted in the U.S.A. in February 2001, serial number 09/778,181, describing the application of intelligent agents to biocomputational analysis. This is the basis of the www.doubletwist.com biocomputational system.

Collaborators

James Bernstein
Kelly Felkins
Edward Kiruluta
Bruce Xuefeng Ling, Ph.D.
Christopher Mader
Jeffrey Rule

Acknowledgements

Douglas Brutlag, Ph.D.