National Centers for Biomedical Computing Software and Data Integration Working Group (SDIWG)

1. National Centers for Biomedical ComputingSoftware and Data Integration Working Group (SDIWG) Peter Lyster and Zohara Cohen NA-MIC All Hands Meeting Tuesday January 10, 2006

2. Brief Journey Through the Seven Centers

3. NIH Roadmap National Centers for Biomedical Computing (NCBC) Physics-Based Simulation of Biological Structures (SIMBIOS) Russ Altman, PI Informatics for IntegratingBiology and the Bedside (i2b2) Isaac Kohane, 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

4. Physics-based Simulation of Biological Structures (SIMBIOS)PI: Russ Altman, M.D., Ph.D. PI Institution: Stanford University This Center will develop, disseminate, and support a simulation tool kit (SimTK) that will enable biomedical scientists to develop and share accurate models and simulations of biological structures from atoms to organisms. SimTK will be an open-source, extensible, object-oriented framework for manipulating data, models, and simulations. The software will include advanced capabilities for modeling the geometry and physics of biological systems, generating the governing differential equations of these systems, integrating the equations to simulate the system dynamics, and interpreting the simulation results through comparison with experimental data. http://cbmc-web.stanford.edu/simbios/

5. National Alliance for Medical Imaging Computing (NAMIC)PI: Ron Kikinis, M.D.PI Institution: Brigham and Women's Hospital The National Alliance for Medical Image Computing (NAMIC), proposed here, will integrate the efforts of leading researchers with a shared vision for development and distribution of the tools required to advance the power of imaging as a methodology for quantifying and analyzing biomedical data. This shared vision is based on a thorough composition of computational methods, from image acquisition to analysis, that builds on the best available practices in algorithm development, software engineering, and application of medical image computing for understanding and mitigating the effects of disease and disability. NAMIC’s goal is to develop, integrate, and deploy computational image analysis systems that are applicable to multiple diseases, in different organs. To provide focus for these efforts, a set of key problems in schizophrenia research has been selected as the initial Driving Biological Projects (DBPs) for NAMIC. http://www.na-mic.org/index.htm

6. Informatics for Integrating Biology and the Bedside (I2B2)PI: Isaac Kohane, M.D., Ph.D.PI Institution: Brigham and Women's Hospital I2B2 (Informatics for Integrating Biology and the Bedside) is an NIH-funded National Center for Biomedical Computing based at Partners HealthCare System. The I2B2 Center is developing a scalable informatics framework that will bridge clinical research data and the vast data banks arising from basic science research in order to better understand the genetic bases of complex diseases. This knowledge will facilitate the design of targeted therapies for individual patients with diseases having genetic origins. http://www.partners.org/i2b2

7. Center for Computational Biology (CCB) PI: Arthur Toga, Ph.D. PI Institution: University of California at Los Angeles The Center for Computational Biology (CCB) was established to develop, implement and test computational biology methods that are applicable across spatial scales and biological systems. Our objective is to help elucidate characteristics and relationships that would otherwise be impossible to detect and measure. The CCB employs an integrative approach, both in terms of the biology and the participating disciplines. The Center focuses on the brain, specifically on neuroimaging, and involves research in mathematics, computational methods and informatics. It also is involved in the development of a new form of software infrastructure – the computational atlas – to manage multidimensional data spanning many scales and modalities. This will be specifically applied to the study of brain structure and function in health and disease, but will have much broader applicability to both biomedical computing and computational biology. http://www.loni.ucla.edu/CCB/

8. National Center for Biomedical Ontology (NCBO) PI: Mark A. Musen M.D., Ph.D. PI Institution: Stanford University Capture and index experimental results Open Biomedical Ontologies (OBO) Open Biomedical Data (OBD) Revise biomedicalunderstanding BioPortal Relate experimental data to results from other sources

9. Ontologies are essential to make sense of biomedical data

10. Core 3 Prostate Cancer Core 1: Computational Technology Core 2: Bioinformatics Technology Diabetes Complications Diabetes Genetics Bipolar Genetics National Center for Integrative Biomedical Informatics (NCIBI) PI: Brian D. Athey Ph.DPI Institution: University Of Michigan Core 5:Training and Education Core 6: Outreach and Dissemination Core 7: Administrative Infrastructure Core 4: Computing and Data Infrastructures

11. Granular Overview of NCIBI Activities Underway: Cores 1-3 workflow BDW Client MiMI BDW A B C D Data MiMI Onco- mine MBI Track user actions MiMI MBI Tools for large- Scale analysis Driving Biological Problems SAGA: Patel Lab Feldman Lab: Diabetes Type 1, NRF2 signaling T1DM pathways First User Environment Database Technologies for Deep Integration of Biological Information User Environment Analyses: Mirel, Ackerman labs Bayesian Nets: Woolf Lab McInnis lab: Bipolar Disorder Interactions Between Genetic Linkage Peaks, WNT signaling BDW: Biomedical Data Workbench prototype: Weymouth Workflows: Weymouth, McEachin, & Programming team MiMI: Michigan Molecular Interactions DB: Jagadish lab, Tarcea Boehnke Lab: Diabetes Type II SNP WGA workflow Genetic Interaction linkage workflow Natural Language Processing: States, Meng, Radev Labs Chinnaiyan and Omenn Labs: Prostate Cancer Bayesian, NLP, Oncomine Concept Mapper: Rhodes, Patel, Woolf MBI: Molecular Biology Integration DB: States lab, Phillips

12. Multiscale Analysis of Genomic and Cellular Networks (MAGNet) PI: Andrea Califano Ph.D. PI Institution: Columbia University • To study the organization of the complex networks of biochemical interactions whose concerted activity determines cellular processes at increasing levels of granularity. • To provide an integrative computational framework to organize molecular interactions in the cell into manageable context dependent components. • To develop interoperable computational models and tools that can leverage such cellular interaction maps to elucidate key biological processes and to dissect complex diseases.

13. Core I, II, III BISON MPI Algorithms Data BISON MAGNet/C2B2 BISON geWorkbench Core 1: Computational SciencesCore 2: BioinformaticsCore 3: Cadherin binding; Cancer; Genetic determinants of common heritable disorders (Alzheimer’s, Autism)

14. Goals of Software and Data Integration Working Group (SDIWG) The RFA states the goal of creating “the networked national effort to build the computational infrastructure for biomedical computing for the nation”. In furthering this, the goals of the SDIWG in concert with the Project Team and Centers staff are: 1. To advance the domain sciences, and promote software interoperability and data exchange. 2. To capture the collective knowledge of software engineering and practices among the Centers and publish this knowledge widely

15. Mechanics… • Google “SDIWG” (on NA-MIC!) • Monthly last Friday Breeze/Tcon 2:30 – 3:30 PM ET • Open (anyone and open minutes) • Chair Peter Lyster lysterp@mail.nih.gov

16. NCBC Staff serving on SDIWG • Bill Lorensen (NA-MIC) • Mike Sherman (Simbios) • Henry Chueh (I2B2) • Ivo Dinov (CCB) • Aris Floratos (MAGNet) • Daniel Rubin (NCBO) • Brian Athey (NCIBI) • Many others—Suzi Lewis, David States, Steve Pieper, Tina Kapur, Shawn Murphy, Center PIs…

17. NIH Staff serving on SDIWG Peter Lyster (NIGMS, Chair) Michael Ackerman (NLM) Carol Bean (NCRR) Art Castle (NIDDK) German Cavelier (NIMH) Larry Clarke (NCI) Zohara Cohen (NIBIB) Elaine Collier (NCRR) Jennifer Couch (NCI) Peter Covitz (NCI) Valentina Di Francesco (NIAID) Dan Gallahan (NCI) Peter Good (NHGRI) John Haller (NIBIB) Donald Harrington (NIBIB) Peter Highnam (NCRR) Michael Huerta (NIMH) Donald Jenkins (NLM) Jennie Larkin (NHLBI) Yuan Liu (NINDS) Michael Marron (NCRR) Richard Morris (NIAID) Bret Peterson (NCRR)Salvatore Sechi (NIDDK) Karen Skinner (NIDA) Michael Twery (NHLBI) Terry Yoo (NLM)

18. There are a number of similar efforts to SDIWG at NIH • Integrated Cancer Biology Program (ICBP) • NIAID (Allergy/Infectious Diseases) Bioinformatics Resource Centers (BRC) • Clinical Research Networks (NECTAR) • Chronic Disease Outcomes (PROMIS) • BIRN/caBIG/Sysbiol/Glue/PSI/TCNP/P41/PGA/…

19. Discussions in the First Year • Demonstration projects • Collaborative software development environment (yellow pages, knowledge environment, full development environment or Framework) • NIH-forge • Emerging efforts in the community (Neuroscience Information Framework, Clearinghouse) • Intangibles—pairwise interactions, consciousness raising • Example—SimTK<->ITK

20. What would you do?

21. A Biomedical Software Ontology? • ACM categorization • Open Directory Project • Home grown, tipping point? • Why? • To help query tools (computer-computer, • human-computer) • To help describe tools (human-human)

22. NA-MIC Software Classification Applications Slicer2 LONI Pipeline Toolkits Insight Toolkit itk Ontology Visualization Toolkit KWWidgets Software Engineering Tools Testing Dart2 CTest Cross-Platform Build CMake Cross-Platform Distribution CPack Cross-Language Wrapping Cable SWIG National Alliance for Medical Imaging ComputingRon Kikinis, M.D.Brigham and Women's Hospital

23. Simbios Software Classification Deliverable software modules Applications Models Computational components Model-building toolsets Application-building toolsets User and developer documentation Application areas Driving Biological Problems RNA folding Myosin dynamics Cardiovascular fluid dynamics Neuromuscular biomechanics Molecule modeling Electrostatics Physics-based Simulation of Biological StructuresRuss Altman, M.D., Ph.D. Stanford University

24. Simbios Software Classification User categories Clinician Experimentalist Application developer Modeler Algorithm developer SimTK software developer Platform support Shared memory multiprocessors Clusters Language support Scripting languages (tcl, perl, python) Java C++, C, Fortran

25. Simbios Software Classification Computation components Linear Algebra (dense, sparse) Numerical Integrators (ODE, DAE, PDE, Stochastic) Monte Carlo simulation Multibody dynamics Contact modeling Meshing PDE solvers Controllers Optimizers/root finders Molecular force field calculations Computational geometry Software development tools Source control Multiplatform build system Regression test support Document generation

26. Simbios Software Classification Software dissemination tools User interaction Bug reporting Feature requests Mailing lists News Software and documentation delivery support Installation Download Upgrade Education Tutorials Course material and software Online courses

27. CCB Software Classification Analysis EDA Feature Analysis Shape Analysis Pattern Recognition Genomic & Phenotypic Data Analysis Statistical Analysis E.g., R Integration DB Efficient DB Traversal & Querying Graphical e.g., HIVE Pipeline Grid Computing Resources Mappers e.g., BrainGraph BrainMapper Portals Resource Integration Components Web Services Center for Computational BiologyArthur Toga, Ph.D. University of California at Los Angeles

28. CCB Software Classification Modeling Algorithms Image Processing Atlas Generation Cortical Modeling Registration Segmentation Engineering Open Source Sequence Annotation Simulation SW Development

29. CCB Software Classification Pre-Processing Data Transforms Spectral Transforms Fourier Transform Wavelet Transform Filtering Skull Stripping Inhomogeneity Correction Visualization Clinical Charts e.g., Demographics Graph Viewers Hyperbolic Graphs Hierarchical Trees Imaging Cross-Sectional Viewers Manifold Viewers 2D, 3D, 4D, ND Sequences

30. I2B2 Software Classification Computer Code Source Application Domain(s) License Tested on platforms Binaries Application Domain(s) License Tested on platforms Informatics for Integrating Biology and the BedsideIsaac Kohane, M.D., Ph.D.Brigham and Women's Hospital

31. I2B2 Software Classification Documentation Biological Experimentation Protocol Algorithm Description Publication PUBMED ID or DOI Human Cohort Description Non-Human Experiment Description Organism Software/Technology protocols Technology white papers Application Domains (Software)

32. I2B2 Software Classification Clinical Information Systems Natural Language Processing Predictive modeling and classification Extract, Transform, and Load Database Schema Ontology Management Service Oriented Architecture Biological Structure Structure:function and structure:disease prediction Protein interaction modeling

33. I2B2 Software Classification Clinical Information Systems (cont.) Expression studies Disease classification/advanced histopathology Outcome prediction Pathway/gene identification for disease Integration of expression w/DNA studies DNA/Genomic studies Population study type (association/linkage etc) Scoring/predicting DNA variant functional and disease significance Comparative genomics Motif studies (TFBS/miRNA/protein domains, etc)

34. Experimental Data Aggregate Human Data DNA RNA Protein Non-image Phenotype Image Individual Human Data DNA (sequence and mapping) RNA Protein Non-image Phenotype Image Usage Non-human DNA RNA Protein Non-image Phenotype Image I2B2 Software Classification

35. I2B2 Software Classification Equipment Genomic Measurement Imaging Imaging Class DNA (sequence and mapping) Gene ID Non-standard descriptor RNA Gene ID Splice variant Non-standard descriptor Protein ID or AA sequence? Non-image Phenotype UMLS or other vocabulary coded Narrative text

36. I2B2 Software Classification Usage Anonymity level Consent level Image DICOM? MIME type? Reagents Chemical Organismal

37. NCIBI Software Classification Data Management TIMBER Michigan Molecular Interactions (MiMI)  Biomolecular Interaction Network Database (BIND) Human Proteome Organization Plasma Proteome Project (HUPO) BioWarehouse BioPAX Community and User Interaction Systems Worm Community System MEDSPACE VEHICLES Answer Garden NJFun National Center for Integrative Biomedical InformaticsBrian D. Athey, Ph.D. University of Michigan

38. NCIBI Software Classification Database and Information Retrieval Algorithms Periscope miBLAST Workflow Management GenePattern Ontologies and Natural Language Processing MEAD NewsInEssense Protégé Image Processing Edgewarp

39. MAGNet Software Classification Analyses Regulatory/Signaling network reconstruction ARACNE (gene expression) GeneClass (gene expression) Network characterization NetClass/NetBoost (molecular interactions networks) InfoMod (molecular interactions networks) Homology-based protein sequence classification Bio-Kernels (protein sequences) RankProp/MotifProp (protein sequences) Structured-based protein classification Protein function pipeline (protein structure) Prediction of regulatory (cis) elements REDUCE (gene expression) MEDUSA (gene expression) National Center for Multi-Scale Study of Cellular NetworksAndrea Califano, Ph.D. Columbia University

40. MAGNet Software Classification Analyses (cont.) Protein structure prediction Protein structure pipeline (protein sequence) META-server (protein sequence) Super-NEST (protein sequence) Protein cavity and binding site prediction (protein structure) Databases Molecular interactions GeneWays Cellular Network KB Gene-phenotype associations PhenotypeML DB

41. NCBO Software Classification Ontology Management Protégé OBO-EDIT Ontology Diff and Alignment PROMPT Ontology Visualization Jambalaya OntoViz PromptViz TGViz National Center for Biomedical OntologyMark A. Musen, M.D., Ph.D. Stanford University

42. NCBO Software Classification Biomedical Data Annotation OBO-EDIT Progammatic Access to Ontologies Protege Script Tab Web Access to Ontologies WebProtege

43. Suggestion for Next Step • Choose one of the existing ontologies as a starting point, perhaps a combination of the MAGNet and CCB ontologies. • Find one person at each NCBC who is interested in the issue of categorizing their software and who is willing to work with the SDIWG on incorporating their taxonomy into the chosen top level ontology. • The NCBC representatives will incorporate their software into the ontology directly or suggest changes needed in order to incorporate their software. • After the software are all integrated into one ontology, a set of attributes to describe the software will be developed. A format for storing the ontology will be agreed upon. • Each NCBC will create and host an instance of the ontology to describe their resources. OWL/Semantic Web may be used.

46. Contact Peter Lyster lysterp@mail.nih.gov 301.594.3928