Feb 05

EOL/BHL Research Sprint at NESCent - Update 1

A group of researchers has gathered this week in Durham, North Carolina at the National Evolutionary Synthesis Center (NESCent) to tackle some fascinating projects about the ecology, evolution, and conservation of biodiversity. What these projects have in common is that that they are all using information from the Encyclopedia of Life or the Biodiversity Heritage Library to answer their questions.

Here are some of the projects:

  1. Esther Quintero and Anne Thessen are text mining data from EOL and BHL in order to statistically assess the conservation risk of species of Mexican amphibians.
  2. Elise Larsen and Yan Wong will figure out how to automatically extract color information from EOL images of butterflies. Variation in color may determine species’ responses to habitat edges and to climate change.
  3. Gordon Burleigh, Scott Chamberlain, Craig McClain, and Meghan Balk are asking if co-occuring sister species in vertebrates show more divergence in body size (for example) than sister species that do not co-occur.
  4. Nicole Angeli and colleagues, along with Javier Otegui, are considering potential co-extinction of mutualistic species due to range shifts in order to inform conservation planning efforts for at-risk species.

Aside from consumption of copious amounts of coffee, we’re consuming and generating vast amounts of data.

Please stay tuned to the EOL blog for some hints of what we are learning, and follow our progress on Twitter this week via the #eolbhl hashtag.

EOL and BHL acknowledge the Richard Lounsbery Foundation for supporting this exciting research sprint.

Jan 27

TraitBank release notes

OK, we’ve now officially launched TraitBank after a period of quiet testing and over a year of planning, development and late-night Taco Bell visits.
New to TraitBank?   Learn the TraitBank basics then try out our nifty new TraitBank data search facility for starters.
We are eager to hear your responses to the new functionality and data.  Please note that there are already quite a few things on our list of “things we’d like to do”. We would value your feedback to help us focus on what’s most important to you.
  1. Data records and attributes in TraitBank are just the tip of the iceberg so far.  We’ll planning to add many more sources, including detailed morphological data, so stay tuned. If you want your own dataset to appear on EOL pages, please contact us.
  2. We’re very interested in basic visualizations of TraitBank data that could help people explore and draw meaning from large amounts of data. If you develop such visualizations using EOL, please do let us know.
  3. We are aware that we still have more than one attribute with the same label and the same or similar definitions. Where we can we plan to merge these. In some cases we can re-import the data but in others we’ll need to tell our system that the two attributes are similar enough that we can merge them,
  4. It isn’t possible yet to search or download ecological association data records. 
  5. As with most text mining projects, the data from one of our largest sources, ENVIRONMENT-EOL are not perfectly clean.  They will be improving over time.  Please leave comments on all wrong or awkward data records, and if you are a curator please do use the hide link to keep them from being shown to all users.
  6. We know that some attributes are subsets of others or of some higher category (such as “size”) but we haven’t enabled any interesting semantics that help out machine reasoning.
  7. While we have some information about type specimens in TraitBank, we haven’t come to a final solution on the best way to deal with types.  Your input would be welcome.
  8. Some basic statistics about TraitBank holdings are already in our site statistics, but there’s much more we could do with this.
  9. We know that it will be useful to have an advanced search that allows you to combine attributes with each other or with a geographic search.  
  10. For those of you who want machine access to the data we are considering a SPARQL endpoint and/or API access.  For the moment, we are facilitating downloads that are as complete as possible.
  11. We will be developing richer documentation over the next few months. Please let us know what you are most interested in knowing.
Please add your comments to this post, or send an email to secretariat@eol.org with your comments.  Thanks!

Announcing TraitBank

FOR IMMEDIATE RELEASE

Encyclopedia of Life releases TraitBank

Landmark open repository of species facts and measurements now available to support research, education and conservation goals

Washington, D.C., January 27, 2014 – The Encyclopedia of Life (EOL) today announces the first release of TraitBank™, a comprehensive, searchable and open digital repository for organism traits, measurements, interactions and other facts for all species and groups of species.  It fills an urgent need among global research, education and conservation communities for a reliable, curated, comprehensive source of trusted data across the tree of life, as well as the need for a public platform for sharing scientific data in a consistent, accessible manner.

With TraitBank, EOL can help scientists develop and test hypotheses explaining patterns in nature, or estimate food webs by identifying organisms known to interact with each other.  It can also be used to address “easy to ask but difficult to answer” questions such as “Which whales weigh more than 10,000 kg?” and “Which plants have blue flowers?”

Developed with support from the Alfred P. Sloan Foundation, TraitBank (http://eol.org/traitbank) is integrated into the fabric of the EOL website (http://eol.org), where it leverages EOL’s existing infrastructure for taxonomic names, content organization, curation roles and search.  Data records are aggregated from a wide variety of sources, including databases, literature tables, spreadsheets and individual contributors.  At launch, TraitBank features over 3 million records related to more than 250 attributes for over 250,000 taxa (species and groups of species) obtained from 27 data sources.  It is expected to grow substantially as existing and new trait data records are loaded into TraitBank over time.

“TraitBank is unique in serving up reliable real-world data for the entire tree of life,” said Dr. Drew Purves, head of the Computational Ecology and Environmental Science Group at Microsoft Research Cambridge. “To those of us trying to build Global Ecosystem Models, it’s like someone just handed us a list of parts.”

“TraitBank will help accelerate the research process by providing a free, open and accessible way for investigators to share and discover species traits, measurements and interactions,” said Bob Corrigan, Director of Operations for the Encyclopedia of Life.  “The motivation behind TraitBank was to make it easy for users to get a comprehensive picture of these data for any given species or group of species, and then give them access to the underlying publications and data sets quickly.  We’re delighted to share TraitBank with the world, and we’re looking forward to working with scientists and educators to make it as useful to them as possible.”

The EOL team has the capacity to help researchers make the best use of existing TraitBank capabilities and provide expertise for data mobilization, extraction and analysis efforts.  EOL is also interested in reviewing opportunities to extend TraitBank’s capabilities to support project-specific requirements, and it invites biodiversity researchers to take advantage of the significant investments and expertise behind TraitBank as part of their grants and other sponsored projects.  To learn more, please contact the EOL Secretariat at secretariat@eol.org.

About the Encyclopedia of Life: The Encyclopedia of Life (EOL) operates as an ongoing collaboration among its international partners with the mission to raise awareness and understanding of living nature.  Participants include the Atlas of Living Australia (Australia), the Biodiversity Heritage Library, the Chinese Academy of Sciences (China), CONABIO (Mexico), the Global Biodiversity Information Facility (GBIF), Harvard University, the New Library of Alexandria (Egypt), the Marine Biological Laboratory, the Norwegian Biodiversity Information Centre (Norway) and the Smithsonian Institution. With additional support from a global network of content partners, curators and users, EOL works to provide free, open, multilingual, digital access to trusted information on all known species at http://eol.org.

###

Contact:

Mr. Robert Corrigan, +1-202-633-8730; bcorrigan@eol.org

Nov 20

EOL Podcast Highlights - Women in Science

                  Image Credit: Randi D. Rotjan, Ph.D., Associate Research Scientist, New England Aquarium

The following podcasts from the Encyclopedia of Life’s One Species at a Time Podcast series feature female scientists in the field!

Listen to each podcast and make sure to check out the “Meet the Scientist” interview links.

EOL Podcast Highlights - Women in Science

The One Species at a Time podcast series is supported by the Harvard Museum of Comparative Zoology

Image Credit: Randi D. Rotjan, Ph.D., Associate Research Scientist, New England Aquarium

Nov 18

New One Species at a Time Podcast: Dolphins

Image Credit:Howard Goldstein, NOAA Fisheries. World Register of Marine Species. CC BY-NC-SA

You have probably seen cans of tuna in your local supermarket marked “dolphin safe.” That label means the tuna was fished in a way that spares most dolphins from being killed in the tuna fleet’s giant nets. In this podcast, biologist and guest reporter Matt Leslie brings us a story about tuna, the intertwined fate of fisheries and dolphins, and the work of scientists.

Listen to the podcast

The One Species at a Time podcast series is supported by the Harvard Museum of Comparative Zoology.

Oct 24

(New and Improved!) Memory Game from the Encyclopedia of Life

image

This is the classic game of Memory - remember where you saw a species and try to find its match! There is also a quiz that challenges you to match species names with the pictures. 

Try one of our memory games by visiting this site and looking under “Featured Games”. You don’t need to login to play, simply click on a game icon. To create your own Memory Game, you will need to create a login and password.

You’ll see some changes with the latest update to Memory. In addition to speed improvements, you can now also play against Elephas, the computer with a good memory. Good luck!

Note: EOL Memory works best in Google Chrome, Firefox and Safari internet browsers

Oct 22

EOL by the Numbers

In the recent NESCent Research Sprint Call for Proposals we invited  biologists to think of questions that EOL and BHL could help answer.  We’ll provide an informatician to extract and integrate the data that are needed. Some fantastic ideas have already been suggested by Rod Page in his iPhylo post.

To help you come up with do-able, interesting, big data questions, here is a snapshot as of early October 2013 of what we have on EOL. Some numbers are rounded to the nearest 1000. Also see our daily statistics for more information and feel free to ask questions in the EOL Forum.

3.6 million taxon pages (aka taxon concepts):

Of these 3.6 million pages, 1,368,301 pages have some content associated with them.  The balance are pages with a name, but no content…yet.

These pages represent the aggregation and reconciliation of:

31.8 million name usages provided to EOL by its partners (name-taxon-source combinations) of which…

From 19 taxonomic hierarchy providers available via the API, as well as 
250+ other taxonomic sources we’ve mapped across 

EOL has:

6 million data objects

also 30,000 sounds and 10,000 videos

For your text-mining pleasure, EOL’s text objects are distributed across the following top 6 subjects:

image

Early next year we’ll introduce support for structured data.  So far we have:

21 trait datasets

268,000 taxa with some trait data

170 types of traits (including types of relations between taxa)

2.8 million trait data records

Common trait types include:

Beta testing for these new features starts Monday, October 28, 2013. If you would like to be a beta tester, please email us at tech@eol.org to sign up. If you know of a great dataset of traits that is freely shareable, please let us know!

Questions about the NESCent-EOL-BHL Research Sprint? Contact Cyndy Parr or Craig McClain.

Oct 21

New One Species at a Time Podcast: Honey Bees

Image Credit:Treesha Duncan, Flickr: EOL Images. CC BY

In this podcast, we venture into a cloud of honey bees to learn about the unique way one bee scientist is managing to help bees and fund his research at the same time.

Listen to the podcast

Learn more about Honey Bees on EOL

Oct 17

clusterpod:

Thick-tailed Gecko Underwoodisaurus milii.
Also known as a Barking Gecko.
Jaurdi, Western Australia

clusterpod:

Thick-tailed Gecko Underwoodisaurus milii.

Also known as a Barking Gecko.

Jaurdi, Western Australia

New One Species at a Time Podcast: Bats

The batman of Mexico has his own bat-cave. He just shares it with 4,000 Mexican long-nosed bats. In this episode, join researcher Rodrigo Medellin as he descends into the Devil’s Cave just north of Mexico City. It’s a journey that started decades ago when Medellin was on a game show as a boy. He lost the game show, but won a prize far more valuable—for himself, his students, and Mexico’s bats. Ari Daniel Shapiro reports from Tepoztlán.

Listen to the podcast

Learn more about bat diversity on EOL