Granada, Spain, March 28 March 31, 2004

1
Granada, Spain, March 28 March 31, 2004
2
Task II functional annotation for proteins

• 1.'Recover' text that proves the GO annotation
• We provided the protein, its GO annotation and
  the associated publication and the participants
  had to provide a part of the document that would
  (to a human expert) prove the original
  annotation.
• 2.Provide GO annotation for human proteins
• We provided the protein and the associated
  publication and the participants had to
  'annotate' automatically the protein according to
  the information in this paper and provide a part
  of the document to prove the annotation.
• 3.Selection of relevant papers
• We provided a protein and a (high) number of
  papers of which most are irrelevant for the
  protein. The participants had to detect which
  papers are relevant for a protein in the sense
  that they contain information that would be
  suitable to derive a GO annotation and also
  provide these parts of the papers that would be
  useful for annotation.

3
Challenges

• We did not provide protein name dictionaries
• GO consists of three (non overlapping) parts
• One protein can have many different functions
• The function of a protein can be described in
  many different articles and in different ways
• The GO codes have to be predicted precisely
• One article can describe different functions of
  the same protein AND/OR mention a number of
  proteins
• Full-text articles are long and in general only a
  (small) section of the whole paper is relevant

4
The GO Gene Ontology
5
(No Transcript)
6
GOA annotation at the EBI
7
Journal of Biological Chemistry from HighWire
Press
8
Examples Iclear cases

• Protein RGS4GO0005516 calmodulin binding
  activityPMID 10747990
• Text 'Indeed, Ca2/calmodulin binds a complex of
  RGS4 and a transition state analog of Galpha
  i1-GDP-AlF4-'
• Protein p21waf/cip1GO 0008285 negative
  regulation of cell proliferationPMID 10692450
• Text 'The p21waf/cip1 protein is a universal
  inhibitor of cyclin kinases and plays an
  important role in inhibiting cell proliferation'
• Protein ThrombinGO0006915 apoptosisPMID 106
  92450
• Text 'Induction of Apoptosis by Thrombin'

9
Examples IIindirect/difficult cases

• Proteins RGS1,RGS2,RGS4,RGS16GO
  0008277 regulation of G-protein coupled receptor
  protein signaling pathwayPMID
  10747990
• Text 'We report that calmodulin binds in a
  Ca2-dependent manner to all RGS proteins we
  tested, including RGS1, RGS2, RGS4, RGS10, RGS16,
  and GAIP'
• and later in the text
• 'To investigate the role of Ca2 in feedback
  regulation of G protein signaling by RGS
  proteins, we characterized ...'.
• One would have to establish first the relation
  between the individual proteins and the fact that
  they are all RGS proteins and then interpret from
  the second sentence later in the text that these
  proteins are related to G protein signaling.

10
Examples IIindirect/difficult cases

• Protein MIP-1alphaGO0007186 G-protein coupled
  receptor protein signaling pathwayPMID 10734056
  
• Text 'Taken together, these results indicate
  that CCR1-mediated responses are regulated at
  several steps in the signaling pathway, by
  receptor phosphorylation at the level of
  receptor/G protein coupling and by an unknown
  mechanism at the level of phospholipase C
  activation'
• and later
• 'In this study, the CCR1 receptor, which binds
  RANTES, MIP-1alpha , MCP-2, and MCP-3 with high
  affinity'.
• The first sentence establishes that CCR1 is
  related to a G-protein coupled receptor pathway
  and the second sentence states that MIP-1alpha
  binds to this receptor and it can be deduced that
  it is therefore also related to this process.

11
Examples IIindirect/difficult cases

• Protein CCR1GO0006955 immune
  responsePMID 10734056
• Text 'Thus, the ability of such classes of CC
  and CXC chemokine receptors to selectively
  cross-regulate each other at multiple levels may
  be physiologically relevant in controlling immune
  response'
• In this case one would have to know (from an
  external source) that CCR1 belongs to these
  classes of receptors to deduce the relation to
  immune response (that is questionable from this
  sentence anyway).

12
Data sets

• Training data
• 636 JBC papers
• 136 Nature papers

• Resources created
• evaluated GO/protein relations 22.000
• correct entries 3381

• Test data
• Task 2.1
• all 1076
• proteins 138
• GO terms 580
• papers 113
• Task 2.2
• all 1227
• proteins 138
• papers 99

13
Evaluation schema
14
Results task 2.1
15
Results task 2.2
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)