Entering edit mode
11.0 years ago
bioinfo
▴
830
Identify Organisms from a Stream of DNA Sequences
This Challenge seeks innovative algorithms to analyze samples that contain mixed segments of genetic sequence from next-generation sequencing instruments and report the identities of each organism represented in the sample and characterization of all non-host organisms. Technical details and requirements are available in the full description.
The basic premise of metagenomics: fix all the issues with databases and come up with better identification algorithms to identify the unknown. Contest ends in 6 days: ready, set, GO!
The contest ends on 31st May so still almost a month to go.
I'm not sure if I mistook the date or it was extended, but I still feel -- as someone who works in this area -- that this is an impossible task. So much of our identification abilities rely on databases and for some environmental samples we're able to taxonomically identify only 25% of the reads. You can design computational algorithms until you are blue in the face, but if you don't know what to compare an unknown read to you won't be able to identify it. I would rather see this money be put towards better database design, curation, and availability.