LSID Mashup (Morphster: Morphological Phyloinformatics)

1
LSID Mashup(Morphster Morphological
Phyloinformatics)

• Professor Daniel P. Miranker
• Department of Computer Sciences
• University of Texas at Austin
• Austin, Texas, USA
• http//www.morphster.org

Collaborators Tim Rowe, Julian Humphries
2
Morphster Workflow
uBio
Catalog of Taxonomic Names
UTCT
Treebase




Morphster
Image Database (one of many)
Matrix (Nexus file)
ZFIN
Anatomical Ontology (one of many)
3
Problem Integer Coding of Matrices

• Input for tree reconstruction algorithms
• Otherwise, all the biology is gone.

4
Later - Go Back and Make Sense of it All
uBio
Rosa minutifolia
Catalog of Taxonomic Names
Treebase
UTCT
Image Database (one of many)
ZFIN
?
Anatomical Ontology
5
The Time Space Continuum and Pointers

• In RAM, pointer // In C
  variable
• Functions at
• Fixed time
• Fixed Space
• In a database, surrogate (RID)
• Functions at
• Any time
• Fixed space
• In a networked world, LSID
• Functions
• Any time
• Any where

6
Every Record Gets an LSID

• Where every record means,
• each encoded part of a NEXUS file
• Morphster LSID resolution yield LSIDs back to the
  source data

LSID1 Character State Assignment LSID2
Taxon (uBio) LSID4 Anatomical Feature,
ZFIN ontology LSID3 Labeled Image
in Morphbank
7
Morphster LSIDs Embody All the Provenance of a
Study
uBio
L2
Rosa minutifolia
Catalog of Taxonomic Names
Build a data structure, Label L1
L3
UTCT
Treebase L1
Image Database (one of many)
L1
ZFIN
L4
L2 L3 L4
Anatomical Ontology
uBio UTCT Ontology ID
Data Provenance
8
Morphster - Demo (on request)

• Productivity tool
• Targeted activity
• Image Driven Ontology Editor
• Compare to something familiar, Mesquite
• Browse, collect information from diverse sources
• Image annotation themselves are recorded in the
  ontology

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
Other Major Aspects

• Ontobrowser
• publication of illustrated ontologies as web
  sites.
• V1 done.
• V2 nearly done.
• OBO ----gt OWL translation
• Ours does a round trip
• done
• Schema Driven LSID Assignment
• Given your data is in a relational DBMS
• Specify an export schema (in SQL)
• Compiler and runtime does all the work.
• At the feasibility stage.

17
The Full Picture Could be Implemented by a
Compiler