Computable Data Challenge community

EOL is holding a competition for exploring the use of EOL to enable computable data for scientific analyses. This community will be a place for a Frequently Asked Questions and discussion about the challenge. For detailed information, please go to the EOL Computable Data Challenge page.

The Newsfeed for this EOL Community gathers updates associated with the items belonging to its Managed Collections, including activities of its members and comments from other EOL users.

Add a new comment

Newsfeed

  • Profile picture of Kuldeep Singh who took this action.
    Kuldeep Singh joined the community "Computable Data Challenge community".

    about 2 years ago

  • Profile picture of Fawsy Bendeck who took this action.
    Fawsy Bendeck joined the community "Computable Data Challenge community".

    about 2 years ago

  • Profile picture of Diogo Silva who took this action.
    Diogo Silva joined the community "Computable Data Challenge community".

    about 2 years ago

  • Profile picture of orovellotti Olivier who took this action.
    orovellotti Olivier joined the community "Computable Data Challenge community".

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    @Quentin Groom: Sorry for the confusion -- we are providing funds, but if there are subcontracts to others, you would need to handle that administration. Also, you cannot count on EOL developers to make modifications in order to accommodate your work - we can't guarantee the changes would be top priority.

    about 2 years ago

  • Profile picture of Quentin Groom who took this action.

    Quentin Groom commented on "Computable Data Challenge community":

    @Roderic Page: Ah, that makes more sense. Thanks for responding.

    about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    @Quentin Groom: EOL is offering money, in the budget for your proposal you would state how much you are asking for (up to a limit of $US 50,000) and what you will spend it on.

    about 2 years ago

  • Profile picture of Quentin Groom who took this action.

    Quentin Groom commented on "Computable Data Challenge community":

    The data challenge budget is a little confusing. If this is being funded using our own personnel funds what difference does it make to EOL what the budget is? BTW: the is a interesting challenge and I applaud EOL for setting it. Unfortunately, I've only just discovered it and don't think I will get an entry together for the end of the month. I'm obviously not linked in to the right networks to find out about this earlier.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    ANNOUNCEMENT!! We have extended the deadline to next Thursday, May 31, 2012.


    Also to answer publicly a question I got by email: as long as an entry is submitted on that day (in any time zone) it will be eligible.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    Folks, if you do not get an acknowledgement that I have received your entry within a few hours, please send me a message at parrc@si.edu. Thanks.

    about 2 years ago • edited: about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    For the curious my entry is available on figshare http://hdl.handle.net/10779/4630d01074e61735a5443786377dcb83.

    about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    @Cyndy Parr: I've just submitted mine. Hope the challenge attracts the kind of interest you're looking for.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    Any last questions on submitting entries for the Computable Data Challenge?

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    @Roderic Page: We're actively working on how to clean up and recommend standardized formatting of associations data. That's a good example.


    Looks to me like both "subject" and "subjects" parameter work but we'll be sure to fix the docs so it isn't confusing.

    about 2 years ago

  • Profile picture of Bob Corrigan who took this action.

    Bob Corrigan commented on "Computable Data Challenge community":

    Rod, if there are areas where the EOL API docs need help, let me know and we'll get a resource on that.

    about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    @Cyndy Parr: Not easy indeed. I had a quick look at Taenia taeniaeformis. After discovering that the EOL API documentation is inconsistent (is the query parameter subject or subjects?) I grabbed the associations for this parasite (see https://gist.github.com/2720339.). Lots of HTML formatting, and unlike the EOL web pages the taxon names aren't linked to EOL pages.

    Looks like a fun exercise in text mining for someone.

    about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    @Roderic Page: I picked wasp associations frivolously but I do hope to get that data out of a Word doc soon and onto EOL. EOL has large amounts of association info from multiple sources (tho mostly trophic not host-parasite) that could be text mined once you grab from API (see http://eol.org/info/toc_subjects). We also have additional unnamed taxa other than what we get from NCBI (tho I agree they are the vast majority) and as well as contextual info. Of course, I won't say it will be easy...

    about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    @Cyndy Parr: Great questions, but how, exactly, would we go about answering them? Unless I'm mistaken, EOL itself doesn't provide the data needed.

    OK, one could tackled unnamed taxa, but I wouldn't do that using EOL, I'd go straight to NCBI and grab their taxonomy (I can download the whole thing, not try and scrape via an API).

    Similarly, host-specificity assumes that we have data on host-parasite associations. Host-parasite databases are pretty thin on the ground (certainly digitised ones are) - ironically GenBank is probably the best source of large-scale data for that purpose: http://iphylo.blogspot.co.uk/2011/03/visualising-symbiome-hosts-parasites.html. Does EOL expose host-parasite associations via its API (I haven't seen any examples)?

    Perhaps I'm missing something, but if those were my questions I wouldn't use EOL.

    about 2 years ago • edited: about 2 years ago

  • Profile picture of Cyndy Parr who took this action.

    Cyndy Parr commented on "Computable Data Challenge community":

    @Roderic Page: Part of my agenda here is to help developers build tools or functionality with clear biological questions in mind. I have no doubt that it will be useful, but a question such as "What is the phylogenetic distribution of unnamed taxa and why?" focuses effort (and it will be interesting to know the answer), just as does "What are the phylogenetic patterns in host-specificity among wasps and what processes might explain them?"

    about 2 years ago

  • Profile picture of Roderic Page who took this action.

    Roderic Page commented on "Computable Data Challenge community":

    @Cyndy Parr: Thanks for the feedback. I'm reworking the proposal a little to make the vision clearer.

    In terms of "specific aspect of biology", perhaps I'm being naïve, but I'd argue that having (a) the primary taxonomic literature for a taxon, (b) recent literature about that taxon, and (c) one or more phylogenies for that taxon available in one place is a useful thing. For example, anyone tackling "dark taxa" would want some way to see named and unnamed taxa in context (i.e., a phylogeny), and would want to see the available taxonomic literature on that group.

    Perhaps this falls into the "projects advancing informatics alone may be less competitive" category, in which case so be it. From my own perspective I want a "biodiversity dashboard" which enables me at a glance to see a summary of the taxonomic and phylogenetic information we have for a given taxon (as well as information linked to that data).

    I'll try and address some of this in the proposal. Thanks again for the comments, hope the challenge attracts some cool projects. I'm also hoping that some will show their hand (perhaps after the deadline) so we can get a sense of the range of things people envisage doing with EOL.

    about 2 years ago