Data Resources

From Evolutionary Informatics Working Group
Revision as of 09:58, 21 November 2008 by Arlin.stoltzfus@nist.gov (talk) (Data Resources)
Jump to: navigation, search

Quick analysis of data resources to target for interop hackathon

The idea is that each of these resources has comparative data that we might want to make interoperable.

questions to ask for each resource

  • What is the scope of the resource?
  • Who controls it?
  • How can users access data?
  • How are data organized? Is there an explicit data model, schema, or format description?
  • Is there evidence that this is an important resources (registered users; citations)?

Phylogeny Services

Phylogeny.Fr

http://www.phylogeny.fr

output formats are Newick, NHX and Phylip, but apparently no means to export alignment and tree together.

Data Resources

The resources below provide comparative data and are listed in order of apparent level of re-use, using literature citations as the measure of apparent data re-use. This is a crude way of prioritizing targets because i) literature citations sometimes do not indicate data re-use; and ii) past re-use is not as important as the potential for future re-use of a resource when semantic tools are used to increase its value.

Panther

The Panther system project is an online data resource for protein families with trees, HMMs, metabolic pathways, and other "functional" information. According to the latest publication [1], "PANTHER is a freely available, comprehensive software system for relating protein sequence evolution to the evolution of specific protein functions and biological roles."

  • 279 citations for 2003, 2005 and 2007 papers, but its not clear how data re-use is based on the comparative aspect of this resource, as opposed to using this resource to mine relations between individual proteins and metabolic pathway annotations.

Hovergen, hogenom

Hovergen, hogenom

  • hovergen 1994 paper cited 157 times

TreeBase

TreeBaseII has the kind of granular schema that would be a good challenge to try to accommodate using cdao. What is missing from cdao (and nexml) is the notion of a "study" with various types of metadata, including publication/reference metadata (which is somewhat dc-like).

Its not clear what reference to site. The TreeBase "Intro" page cites about 7 different references up to 2000, including posters and papers. Some of the papers apparently are scientific studies and not implementation papers. Some of the most likely candidates for a TreeBase citation are as follows (perhaps we should include all of these):

  • Sanderson, et al., 1994 (Am. Jour. Bot.), cited 59 times, can't track citing papers
  • Sanderson, et al., 1993 (Syst. Biol.) cited 29 times, most citing papers are meta-analyses ! (that's good)
  • Morel, 1996, cited 21 times

TreeFam

TreeFam

  • citations for 2006 paper: 36

Pandit

Pandit

Tree o' Life

Microbes Online

Arkin lab's MicrobesOnline server has cool tree-based view of sequence families

PhylomeDB

PhylomeDB

MorphBank

MorphBank

PhyLoTA

PhyLoTA

PhyloFacts

PhyloFacts

TimeTree

TimeTree

MorphoBank

MorphoBank

pPOD

pPOD (not really a data resource, its a db tech project led by computer scientists)