MAGNet Center: Andrea Califano

1
Creating a DBP Community to Enhance the NCBC
Biomedical Impact NCBC Work Group Report, 18
July 2006

• MAGNet Center Andrea Califano
• NCIBI Brian Athey
• Simbios Russ Altman

2
Workgroup Goals

• Problem The NCBCs have no workgroup to help
  build the DBP community
• Goal 1 To determine the Mission and Goals for
  the Applications of Systems Biology, Modeling and
  Analysis Working Group
• Goal 2 To determine how this group would
  interact with the 2 other NCBC Working Groups to
  define key sets of
• Data
• Tools
• Methodologies
• Ontologies
• DBPs
• Goal 3 Identify NCBC DBPs that are highly
  motivated to participate in the Working Group
• Goal 4 How to link to external communities
  (e.g., DREAM-like activities)?
• Additionally Discuss as a focused example
  Molecular Interaction Maps in the context of
  the DBPs

3
DBP Success Depends on the Availability of an
Integrated Resourceome (Not a priority for Core
I/Core II Projects)

• Integrated Computational Biology Platform
• Support for gene expression data, physics-based
  simulations, image analysis, sequences, pathways,
  structure, etc. (40 visualization and analysis
  modules).
• Access to local and remote data sources and
  analytical services.
• Support for workflow scripting.
• Integration with grid infrastructures.

The integration process must be driven by the DPB
requirements rather than by Core I/II activities.

• Development framework
• Open source development.
• Modular/extensible architecture, supporting
  pluggable components with configurable user
  interface.
• Easy integration of 3rd party components.

4
We Must Work With the Yellow Pages WG to
Assemble an Indexable List of Most Useful Tools
and Platforms

• Many Toolkit and platforms
• Internal
• SimTk
• Genopia
• VTK/ITK
• Brainsuite
• geWorkbench
• GenePattern
• MiMI
• SAGA
• miBLAST
• MarkerInfoFinder
• External
• GenePattern
• Systems Biology Workbench
• myGRID
• Cytoscape and ISB tools
• How do we make these tools interoperable? This
  must be DBP-driven because other cores (I/II) do
  not necessarily depend on tool interoperability.

5
GenePattern/geWorkbench InteroperabilityAn
Opportunity and a Starting Point
KNN
PCA
GSEA
WV
SVM
ARACNE
SOM
SPLASH
GenePattern Module Repository
Execute GenePattern modules from within
geWorkbench
Wrap geWorkbench modules as GenePattern tasks
6
Interactions with the Scientific Ontology Working
Group
Component A
_at_Publish public DSDataSet publish(. . .)
DSDataSet dataSet // do some work that
assigns a value to dataSet. return
dataSet _at_Subscribe public void
receive(DSDataSet dataSet, Object source)
// Consume the argument dataSet, as
appropriate
Component B

• Provide re-usable models of common bioinformatics
  concepts
• Data sequence, expression, genotype, structure,
  proteomics
• Complex data structures patterns, clusters,
  HMMs, PSSMs, alignments
• Algorithms Clustering, matching, discovery,
  normalization, filtering
• Provide a foundation for the development of
  interoperable geWorkbench components
• Endorsed by multiple communities (caBIG, AMDeC,
  NCBCs)

7
Identifying Specific Tools

• There are tools, databases, and methods that have
  universal value across different DBPs.
• What are they?
• Which NCBC or external community is producing
  them
• What can we do to standardize their use across
  the community.
• An Example
• Molecular Interaction Maps

8
A Relevant Example That Was Discussed

• Molecular Interaction Maps are becoming the
  equivalent of an anatomy atlas to map specific
  measurements in a functional context e.g. QTLs,
  expression profiles, etc.

Discussion Goal To determine how relevant these
maps are to the DBPs of the various
NCBCs Limitations Many Interactomes are limited
because they are (1) too generic (e.g. missing
cellular and molecular context), (2) poorly
annotated (e.g. linked only to the specific data
used to produce them), (3) limited to pairwise
interactions, (4) lacking quality
control/validation, and (5) not associated to the
investigation of specific biological/biomedical
problems.
9
Example From Molecular Interaction Maps to
Molecular Interaction Knowledge Bases

• What does it take to turn a ridiculome into a
  relevantome?
• Quality control metrics (recall/precision)
• Context specificity
• Cellular Is the interaction specific to a
  cellular phenotype
• Molecular Is the interaction dependent on the
  availability of other molecular species
• Links to data (and literature)
• Links to analysis of biomedical problems
• Focus on specific features (e.g. mechanisms)

10
A Potential Template for NCBC Knowledge
BasesMAGNet Human B Lymphocytes Dataset

• Integrative Framework
• Bayesian Evidence integration of pairwise
  interactions
• Protein-Protein, Protein-DNA
• Prior Knowledge Incorporation
• Context Specific
• ARACNE, GeneWays, REDUCE
• B-Cell data or B-cell specific criteria
• Linked to one of the largest B-Cell expression
  profiles microarray dataset, ChIP-Chip assays
  (MYC/BCL6), miRNA profiles, and Literature
• Captures Multi-variate dependencies
• Three-way interactions via MINDY and MATRIXReduce
• Post-translational modulation of transcriptional
  regulation
• Combinatorial transcriptional regulation
• Signal transduction control of Transcriptional
  Regulation I.e. the Transferome meets the
  Transcriptome
• Links to literature (via GeneWays, NCIBI, I2B2,
  GATE, etc.)
• Other examples? Oncomine (NCIBI), GenePattern
  ALL/AML, Others?

Example
11
Some Key Observations from Attendees

• Systems Biology name is too narrow. Think of
  Alternatives
• Working group to Biomedical Impacts of
  Computational Biology at NCBCs or
• NCBC Biomedical Impact Workgroup
• Is the intramural program a better place to
  create atlases and knowledge bases, since its
  not RO1 funding? They could implement contract
  mechanisms with extramural researchers to
  leverage outside expertise
• Keep in mind that we need to understand what will
  you deliver at the end of 4 years, positioning
  each NCBC for renewal. Which communities are
  using the tools? Are they better off?
• Individual centers can work to create a specific
  resourceome that can be linked and accessible to
  others
• Many working group members had a strong interest
  in multi-scale modeling and biological context

12
Outcomes

• Create a DBP community within the NCBCs
• ACTION Make an interactome map of the existing
  DBPs with potential synergies to be published in
  Symbios magazine
• Use this forum to inform target biological
  communities (not just NCBCs). E.g. DREAM meeting.
• Organize a coordinated effort to evaluate the
  tools and technologies and make them
  interoperable
• ACTION Coordinate the DBP requirements to drive
  the integration of specific tools and data
  resources
• Integrate data and annotation in knowledge bases
  and models for related DBPs. Identify other
  common tools, data, and methods
• Drop the Systems Biology name Use something
  like
• NCBC Biomedical Impact Workgroup
• Commit to a regular T-con and virtual (Wiki)
  participation
• Consider a yearly retreat of NCBC DBPs possibliy
  in collaboration with other NIH roadmap
  activities (e.g. ICBPs)

13
NCBC DBP Interactome I Useful Starting Point
Peter Woolf, NCIBI
14
Distribution Model How Can the 7 NCBCs
Effectively Interoperate?
Informatics for IntegratingBiology and the
Bedside (i2b2) Isaac Kohane, PI
Physics-Based Simulation of Biological Structures
(SIMBIOS) Russ Altman, PI
National Center for Integrative Biomedical
Informatics (NCIBI) Brian D. Athey, PI
National Alliance for Medical Imaging Computing
(NA-MIC) Ron Kikinis, PI
The National Center For Biomedical Ontology
(NCBO) Mark Musen, PI
Multiscale Analysis of Genomic and Cellular
Networks (MAGNet) Andrea Califano, PI
Center for Computational Biology (CCB) Arthur
Toga, PI