OBO-Edit: The Browser - PowerPoint PPT Presentation

Loading...

PPT – OBO-Edit: The Browser PowerPoint presentation | free to download - id: 5d225-ZDc1Z



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

OBO-Edit: The Browser

Description:

Berkeley Bioinformatics and Ontology Project / Gene Ontology. Before We Do Anything! ... GREEDY: Anything that has no visible parents should be displayed as a root. ... – PowerPoint PPT presentation

Number of Views:53
Avg rating:3.0/5.0
Slides: 29
Provided by: geneon
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: OBO-Edit: The Browser


1
OBO-Edit The Browser

The Browser John Day-Richter Berkeley
Bioinformatics and Ontology Project / Gene
Ontology
2
Before We Do Anything!
  • When in doubt, check the User's Guide!

3
Intro to the OBO-Edit Interface
  • How to read the ontology editor panel
  • How to split the ontology editor panel
  • Viewing term details

4
Searching Keywords
  • The simplest OBO-Edit search is a keyword
    search. Keyword searches work kind of like a
    google search.

5
Search Example Keywords
  • Keyword search kinase

6
Working with Search Results
  • Selecting search results
  • Sorting search results
  • Naming sets of search results
  • Tabs or windows?
  • Closing results

7
The DAG Viewer
  • The DAG Viewer displays every single path from a
    term to the root.
  • This view can get very confusing in complex
    ontologies. There are a number of ways to
    simplify the view. (See the User's Guide)

8
Searching Full Term Search
  • It is possible to do a much more precise search
    using the full-featured search capabilities.
  • The full-featured search works by specifying
  • NOT (whether or not to negate the search)
  • A search aspect (we'll talk about this later)
  • Which field to search (term name, comment, id,
    namespace, whatever)
  • What comparison to use (equals, contains, starts
    with, gt, etc)
  • The value to look for

9
Search Example Full Term Search
  • Term search Find terms with comments that
    contain the word kinase

10
Searching The Aspect Field
  • The Aspect field allows you to specify which
    OTHER terms should be looked at to determine
    whether a term should be matched. (OR Sometimes
    you need to look at terms A, B, C, and D to see
    if term X matches your search)
  • The available aspects are
  • Self X is included in the search results if X
    satisfies the search criteria
  • Ancestor X is included in the search results if
    any ancestor of X satisfies the search criteria
  • Descendant X is included in the search results
    if any descendant of X satisfies the search
    criteria

11
Search Example The Aspect Field
  • Term search Find all children of development

12
Searching Compound Searches
  • It is possible to create complex search criteria
    by combining simpler searches with boolean
    operators
  • OBO-Edit's interface defines a complex search
    using prefix notation (sometimes called reverse
    polish notation) rather than the more common
    infix notation. The infix statement a AND (b OR
    C) would be written AND a (OR b c) in prefix
    notation.

13
Searching Compound Searches
  • Keyword search kinase AND cell
  • Term search Term name contains the words
    kinase OR activity
  • Term search Term name contains the word kinase
    AND contains the words activity OR complex

14
Searching Link search
  • Term searches find terms (duh). Link searches
    find relationships between terms.
  • Links searches can be defined in terms of a
    link's child, a link's parent, a link's type, or
    properties of the link itself.

15
Search Example Link search
  • Link search Find all part_of links
  • Link search Find all links with type part_of
    that occur in the molecular function branch of
    the GO

16
Saving/Loading Searches
  • Searches can be saved to disk as an XML file, and
    loaded later.
  • This can be really useful for frequently used or
    complex searches. Share your favorite search with
    your spouse or rabbi!
  • Saved searches can be used by the obo2obo command
    line tool to do quick, file-based ontology
    filtering.

17
Rendering
  • Rendering allows you to specify a special visual
    style for terms that match your search.

18
Rendering Example
  • If a term name contains the word kinase, draw
    that term in red

19
Filtering
  • Filtering changes the main OBO-Edit display so
    that only matching terms (or links) are shown.

20
Filtering Example
  • Only show terms that belong to the
    molecular_function ontology
  • Only show terms that belong to goslim_generic
  • Only show is_a links

21
Root Detection Algorithm
  • The Root Detection Algorithm decides what counts
    as a root. There are (currently) 2 different Root
    Detection Algorithms
  • GREEDY Anything that has no visible parents
    should be displayed as a root.
  • STRICT Only display a term as a root if it
    actually has no parents.
  • STRICT will sometimes hide terms that actually
    match a filter. GREEDY will never do that.

22
The Graph Viewer Plugin
  • The Graph Viewer Plugin uses ATT's GraphViz
    library to draw a graphical version of the
    information shown in the DAG Viewer
  • This plugin requires GraphViz, which can be
    downloaded for your platform from
    http//www.graphviz.org
  • See the User's Guide for information on
    configuring OBO-Edit to use your installation of
    GraphViz

23
Graph Viewer Demo
  • How the Graph Viewer displays selections
  • Selecting terms using the Graph Viewer
  • Saving pictures
  • If time permits Future directions for the Graph
    Viewer

24
Oh the Things You Can Do!
  • Filtered save using the Advanced Save Dialog
  • Filtered save using obo2obo
  • All kinds of cool filtering searching options
    when you use the OBO-Edit reasoner

25
How to Get OBO-Edit
  • http//sourceforge.net/projects/geneontology

26
Acknowledgments
27
Acknowledgments
  • The Good People at Berkeley BOP
  • Seth Carbon
  • Karen Eilbeck
  • Mark Gibson
  • Chris Mungall
  • Suzi Lewis
  • She Shengqiang
  • Nicole Washington
  • The Gene Ontology Consortium
  • AND...

28
More Acknowledgments
Logan Isaac August 22nd, 2006
About PowerShow.com