MIAMEEnv

1
MIAME/Env

• Towards a minimum environmental meta-data
  specification developed for functional genomics
• Norman Morrison
• University of Manchester and NEBC, Oxford.

2
Outline

• Where did MIAME/Env come from?
• Why was a yet another data standard needed?
• How we actually went about it developing it?
• Future considerations
• Functional Genomics Data Standards
• Env as a stand alone specification

3
Remit

• To contribute to the systematic support of
  environmental transcriptomic data management.

4
Diversity of projects

• Stickleback and flounder treated with a range of
  pollutants.
• Arabidopsis (including specific ecotypes and
  cross strains) at particular stages of
  development subjected to various environmental
  conditions.
• Earthworms (sentinel species) subjected to
  various environmental toxins.
• Plants mounting defence to caterpillar attack as
  a result of chemical signals mediated by soil
  micro-organisms.

5
Can we use something that already exists?
Potentially

• MIAME is the Minimal Information for the
  Annotation of Microarray Experiments.
• The result of a MGED (www.mged.org) driven effort
  to codify the description of a microarray
  experiment.
• MIAME aims to define the core that is common to
  most experiments.
• It tries to specify the collection of information
  that would be needed to allow somebody to
  completely reproduce an experiment that was
  performed elsewhere.

6
Data Safari
7
7 Step Development Process

• Identification of the need for a standard
  (MIAME/Tox and MIAME/Nut)
• The formation of a working group (community)
• Selection of case studies
• Development of the specification checklist,
  specification attributes, definitions, and
  allowed terms
• Knowledge acquisition (discussion of requirements
  within the community, case studies)
• Selection and definition of a set of attributes
• Development of allowed terms / controlled
  vocabularies
• Development of a suitable implementation
• Development of a repository to store, view, and
  distribute annotations
• Final annotation of meta-data to compliant format
  and submission to the repository

8
Circadian rhythmicity in tidal worms
9
Algal blooms affected by viruses
10
7 Step Development Process

• Identification of the need for a standard
• The formation of a working group (community)
• Selection of case studies
• Development of the specification checklist,
  specification attributes, definitions, and
  allowed terms
• Knowledge acquisition (discussion of requirements
  within the community, case studies)
• Selection and definition of a set of attributes
• Development of allowed terms / controlled
  vocabularies
• Development of a suitable implementation
• Development of a repository to store, view, and
  distribute annotations
• Final annotation of meta-data to compliant format
  and submission to the repository

11
What is my scope?

• Minimal criteria
• Context dependencies
• Some attributes that are minimally sufficient for
  describing a particular strain of Mouse will not
  apply to the description of a particular strain
  of Bacteria, vice-versa.
• Not known / not-applicable
• Derived meta-data versus Primary meta-data
• Some types of Primary meta-data can be derived
  but will they ever be as accurate?
• What are the overheads in producing derived
  meta-data?
• It would be nice to only have to do it once.
• Will the derived data change with better methods?
• Record the method.

12
Investigation categories

• Field Trials
• wild organism/biosource
• natural environment
• Conditioned field trials
• wild organism/biosource
• natural environment then conditioned in the lab
• animal husbandry conditions (preconditioning)
• treatments (conditioning)
• Lab experiments
• lab reared or obtained from a standard provider
• animal husbandry conditions (preconditioning)
• treatments (conditioning)

13
Environmentally important concepts

• Individuals, Populations and Communities.
• Geographic Parameters
• Topography
• Phenotypic Characteristics
• Behavioural
• Physiological
• Anatomical
• Environmental Parameters
• Climate?
• Photoperiodicity
• Lunar Phase?
• Experimental Phase
• Discrete, Relative and Absolute time
  considerations.

14
(No Transcript)
15
7 Step Development Process

• Identification of the need for a standard
• The formation of a working group (community)
• Selection of case studies
• Development of the specification checklist,
  specification attributes, definitions, and
  allowed terms
• Knowledge acquisition (discussion of requirements
  within the community, case studies)
• Selection and definition of a set of attributes
• Development of allowed terms / controlled
  vocabularies
• Development of a suitable implementation
• Development of a repository to store, view, and
  distribute annotations
• Final annotation of meta-data to compliant format
  and submission to the repository

16
Considerations

• Prescriptive and Specific vs. Flexible and
  Generic.
• Too prescriptive and specific -gt page after page
  of information.
• Too flexible and generic -gt Thing.

17
Meta-data quality

• Accuracy
• Completeness
• Currency
• Portability
• Credibility

• Important to be able to reference external
  sources rather than duplicate them
• Functional annotation that is not updated
• Gene names can change or obtain synonyms, without
  this being reflected in the data
• Chip files can be out of date even on the
  manufacturers web-site

18
Generic Attribute Construct

• Entity or Thing
• A concept that represents an entity that exists,
  potentially described in another ontology.
• Property or Modifier (Measured)
• A characteristic of the entity that is measured,
  for example, size, weight, loudness, gestation
  period.
• Value
• The value - not necessarily quantitative.
• Unit
• Unit where appropriate.
• Assay
• The assay used to measure the property of the
  entity.

19
Phenotypic Characteristic

• Free text
• Calipers were employed to measure the length of
  the dorsal fin. The fin was measured to be 1.2 cm.

Can also be applied to relative characteristics,
ie dissolved oxygen content in mg/l
20
7 Step Development Process

• Identification of the need for a standard
• The formation of a working group (community)
• Selection of case studies
• Development of the specification checklist,
  specification attributes, definitions, and
  allowed terms
• Knowledge acquisition (discussion of requirements
  within the community, case studies)
• Selection and definition of a set of attributes
• Development of allowed terms / controlled
  vocabularies
• Development of a suitable implementation
• Development of a repository to store, view, and
  distribute annotations
• Final annotation of meta-data to compliant format
  and submission to the repository

21
maxdLoad2
22
7 Step Development Process

• Identification of the need for a standard
• The formation of a working group (community)
• Selection of case studies
• Development of the specification checklist,
  specification attributes, definitions, and
  allowed terms
• Knowledge acquisition (discussion of requirements
  within the community, case studies)
• Selection and definition of a set of attributes
• Development of allowed terms / controlled
  vocabularies
• Development of a suitable implementation
• Development of a repository to store, view, and
  distribute annotations
• Final annotation of meta-data to compliant format
  and submission to the repository

23
Circadian rhythmicity in tidal worms
24
Algal blooms affected by viruses
25
The 8th Step

• Submission of compliant meta-data to a public
  repository via a common exchange format.

26
Functional Genomics Standards

• Object Model
• FuGE (Functional Genomics Experiment - Object
  Model)
• http//fuge.sourceforge.net
• Ontology
• FuGO (Functional Genomics Ontology)
• http//mged.sourceforge.net/ontologies/

27
Introduction to FuGE

• A model for developing data standards for
  functional genomics
• General classes for protocols, investigation
  structure, data structure
• Also models equipment, software, contacts etc.
• Can be extended for use in a particular domain
• Uses ontologies extensively, such as MGED
  Ontology (or next version FuGO)

28
Status of FuGE

• Milestone 1 release Sept 2005
• UML (Object Model)
• XML Schema
• Milestone release being tested by MGED
  (Micorarray Gene Expresion Dataand PSI
  (Proteomics Standards Initiative)
• Will form the basis for the next version of
  MAGE-ML and protein separation standards
• Also has been presented to metabolomics community

29
Introduction to FuGO

• An ontology for describing information about a
  functional genomics experiment
• To include a top level structure of general
  concepts for example Investigation, Assay, Study.
• Can be extended for use in a particular domain

30
Status of FuGO

• Historically, FuGO was once MO (MGED Ontology).
• Top level structure to be ratified at MGED8.
  Sept 11-13, 2005.
• Existing classes in MO (transcriptomics) will
  continue to be reorganised as a template for
  other domains to follow suite (proteomics,
  metabalomics, genomics).

31
RSBI - Reporting Structure for Biological
Information
32
RSBI - People

• RSBI Coordinator
• Susanna Sansone
• EBI
• Environmental Genomics WG
• Norman Morrison
• NEBC, NERC Post-Genomics Proteomics programme,
  EBI
• Nutrigenomics WG
• Philippe Rocca-Serra,
• EBI, European Organization NuGO
• Toxicogenomics WG
• Jennifer Fostel,
• NIEHS-NCT, NCTR-FDA, HESI Genomics Committee,
  EBI,
• Contributors/collaborators
• Alex Garcia (EBI, PhD student at Uni of
  Queensland, Australia)
• Chris Taylor (EBI, PSI)

33
(No Transcript)
34
Future considerations

• Plug and play domain specific data standards.
• Q. I want to describe an investigation looking at
  the environmental impact of certain toxins on a
  sentinel species using proteomics. What data
  standard(s) should I be using?
• Similarly
• Q. I want to describe the environmental context
  of a genome sequenced from an environmental
  isolate. Are there existing standards or parts of
  standards I should be using?

35
meta-data overlap
Functional Genomics Investigation
36
Thanks

• Development of the Env specification for
  environmental biology and its application to
  transcriptomics as MIAME/Env
• Norman Morrison, A. Joseph Wood, David Hancock,
  Sonia Shah, Luke Hakes, Bela Tiwari, Peter Kille,
  Andrew Cossins, Matthew Hegarty, Michael J.
  Allen, William H. Wilson, Peter Olive, Kim Last,
  Cas Kramer, Thierry Bailhache, Jonathan Reeves,
  Denise Pallett, Justin Warne, Karim Nashar, Helen
  Parkinson, Susanna-Assunta Sansone, Philippe
  Rocca-Serra, Robert Stevens, Jason Snape, Dawn
  Field, Andy Brass
• NERC Environmental Genomics and Post Genomics and
  Proteomics Programmes.
• http//envgen.nox.ac.uk