caBIG Pilot Project Selection Process

1894 days ago, 707 views
PowerPoint PPT Presentation

Presentation Transcript

Slide 1

Vision and Infrastructure Behind the Cancer Biomedical Informatics Grid 0 Peter A. Covitz, Ph.D. Chief, Core Infrastructure National Cancer Institute Center for Bioinformatics

Slide 2

The Center for Bioinformatics is the NCI's vital and strategic arm for research data administration We work together with both intramural and extramural gatherings Mission to coordinate and fit divergent research information Production, benefit arranged association. Assessed based upon client and accomplice fulfillment.

Slide 3

NCICB Operations groups Systems and Hardware Support Database Administration Software Development Quality Assurance Technical Writing Application Support and Training caBIG Management

Slide 4

Relieve enduring and passing because of malignancy by the year 2015 National Cancer Institute 2015 Goal

Slide 5

Origins of caBIG Need: Enable agents and research groups across the nation to join and influence their discoveries and ability so as to meet NCI 2015 Goal. Procedure: Create adaptable, effectively oversaw association that will interface individuals from the NCI-upheld growth venture by building a biomedical informatics organize

Slide 6

Scenario from caBIG Strategic Plan A specialist required in a stage II clinical trial of another focused on remedial for cerebrum tumors watches that malignancies got from one particular tissue begetter have all the earmarks of being emphatically influenced. The trial has been creating proteomic and microarray information. The specialist might want to distinguish potential biochemical and flagging pathways that may be distinctive between this cell sort and other potential ancestors in tumor, derive whether anything comparative has been seen in other clinical trials including operators known to influence these particular pathways, and recognize any studies in model living beings including tissues with comparative pathway action.

Slide 7

caBIG Governance and Organization

Slide 8

Feudalism caBIG Governance Models X Warlord culture offers minimal impetus to collaborate

Slide 9

Governance Models Forced Collectivization X Centralized solid approach not adaptable or versatile

Slide 10

Federal Democracy Governance Models Balance between focal administration and neighborhood control. Best fit for caBIG Principles. Federalist Papers Alexander Hamilton, James Madison, John Jay

Slide 11

caBIG Organization Structure caBIG Oversight General Contractor Clinical Trial Mgmt Integrative Cancer Research Tissue Banks & Pathology Tools = Project Working Group Working Group Working Group Architecture Vocabularies & Common Data Elements Working Group Working Group Strategic Working Groups

Slide 12

Interoperability capacity of a framework to get to and utilize the parts or hardware of another framework Syntactic interoperability Semantic interoperability

Slide 13

SEMANTIC SYNTACTIC caBIG Compatibility Guidelines

Slide 14

Model-Driven Architecture

Slide 16

MDA Approach Analyze the issue space and build up the ancient rarities for every situation Use Cases Use Unified Modeling Language (UML) to institutionalize display representations and antiques. Plan the framework by creating antiquities in view of the utilization cases Class Diagram – Information Model Sequence Diagram – Temporal Behavior Use meta-demonstrate devices to produce the code

Slide 17

Limitations of MDA Limited expressivity for semantics No office for runtime semantic metadata administration

Slide 18

MDA in addition to a mess more! caCORE

Slide 19

S E C U R I T Y Bioinformatics Objects Common Data Elements Enterprise Vocabulary caCORE

Slide 20

Use Cases Description Actors Basic Course Alternative Course

Slide 21

Bioinformatics Objects

Slide 22

Common Data Elements What do each one of those information classes and characteristics really mean, in any case? Information descriptors or "semantic metadata" required Computable, ordinarily organized, reusable units of metadata are "Regular Data Elements" or CDEs. NCI utilizes the ISO/IEC 11179 standard for metadata structure and enrollment Semantics all drawn from Enterprise Vocabulary Service assets

Slide 23

Description Logic Enterprise Vocabulary Concept Code Relationships Preferred Name Definition Synonyms

Slide 24

Semantic metadata illustration: Agent <Agent> <name>Taxol</name> <nSCNumber>007</nSCNumber> </Agent>

Slide 25

Why do you require metadata?

Slide 26

C1708 C1708:C41243 Computable Interoperability Agent Drug name id nSCNumber NDCCode CTEPName approvalDate FDAIndID approver IUPACName fdaCode My model Your model

Slide 27

Desc. Rationale CDEs Concept Codes 2223333 C1708 2223866 C1708:C41243 2223869 C1708:C25393 2223870 C1708:C25683 2223871 C1708:C42614 Bioinformatics Objects Common Data Elements Enterprise Vocabulary Tying everything together: The caCORE semantic administration system

Slide 28

Cancer Data Standards Repository ISO/IEC 11179 Registry for Common Data Elements – units of semantic metadata Client for Enterprise Vocabulary: metadata built from controlled phrasing and explained with idea codes Precise detail of Classes, Attributes, Data Types, Permissible Values: Strong writing of information items. Devices: UML Loader : consequently enlist UML models as metadata parts CDE Curation : Fine tune metadata and oblige reasonable qualities with information norms Form Builder : Create benchmarks based information accumulation frames CDE Browser : hunt and fare metadata segments

Slide 29

S E C U R I T Y Common Security Module Common Authorization Schema

Slide 30

caCORE Architecture Clients Middleware Data HTTP Clients A P I Web Application Server Biomedical Data Interfaces Java SOAP XML A P I SOAP Clients Common Data Elements Domain Objects [Gene, Disease, etc.] Domain Objects [Gene, Disease, Agent, etc.] Data Access Objects A P I Perl Clients Enterprise Vocabulary Data Access Objects A P I Java Applications Authorization

Slide 31

Development and Deployment DEV… … ..… … ..|QA… ..… ....|STAGE...|PROD PRODUCTION Use Cases Design Test Plans Iterative Development Modeling Unit Testing User Guides System Testing Staging Packaging

Slide 32

caCORE Software Development Kit

Slide 33

caCORE SDK Components UML Modeling Tool (any with XMI trade) Semantic Connector (idea restricting utility) UML Loader (display enrollment in caDSR) Codegen (middleware code generator) Security Adaptor (Common Security Module) caCORE SDK Generates a caBIG Silver-Compliant System

Slide 34

Professional Documentation

Slide 35

mzXML mass spec proteomics information scanFeatures Proteomics AML Proteomics statml Statistical markup show CAP College of American Pathologists conventions for Breast, Lung, Prostate GoMiner Text digging apparatus for GO caTISSUE Tissue keeping money protLIMS Laboratory Information Management System for proteomics BRIDG Clinical Trials caBIO General bioinformatics caDSR ISO11179 metadata EVS Vocabulary caMOD Cancer Models MAGE 1.2 Microarray information CSM Security Common Provenance, DBxrefs caTIES Pathology reports. gridPIR Protein Information caBIG UML Models Completed and in the Works at Cancer Centers for Silver Systems

Slide 36

From Silver to Gold: caGrid

Slide 37

caBIG Use Cases Advertisement Service Provider makes benefit metadata portraying the administration and distributes it to matrix. Revelation Researcher (or application engineer) determines seek criteria depicting an administration of intrigue The exploration presents the disclosure demand to a revelation administration, which distinguishes a rundown of administrations coordinating the criteria, and returns the rundown. Question and Invocation Researcher (or application designer) instantiates the matrix administration and get to its assets Security Service Provider limits access to benefit based upon validation and approval rules

Slide 38

Silver Gold Silver OTHER TOOLKITS NCI OTHER caBIG SERVICE PROVIDERS Cancer Center Cancer Center Cancer Center Cancer Center Cancer Center

Slide 39

Mobius Globus BPEL GRAM Globus myProxy OGSA-DAI Globus Toolkit GSI CAS caCORE Globus caGrid Service-Oriented Architecture Functions Management Schema Management Metadata Management ID Resolution Workflow Security Resource Management Service Registry Service Description Grid Communication Protocol Transport OGSA Compliant - Service Oriented Architecture

Slide 40

Service Data Elements Service Data Elements (SDEs) portray benefits so customers can find what they do Two sorts of top-level framework administrations characterized Data Services Analytical Services Three models for SDEs have been composed Data benefit particular Analytical Service-particular Common (all administrations)

Slide 41

Silver to Gold: Data Services caBIG Gold information benefit EVS caGrid Infrastructure Query Adaptor Silver Data Service

Slide 42

Data Object Semantics, Metadata, and Schemas Client and administration APIs are protest situated, and work over very much characterized and curated information sorts Objects are characterized in UML and changed over into ISO/IEC 11179 Administered Components, which are thus enlisted in the Cancer Data Standards Repository (caDSR) Object definitions draw from vocabulary enrolled in the Enterprise Vocabulary Services (EVS), and their connections are in this manner semantically depicted XML serialization of articles stick to XML diagrams enrolled in the Global Model Exchange (GME)

Slide 43

Analytical Services Accept and discharge specifically information questions that adjust to Gold information benefit prerequisites Analytical technique execution is characterized by administration supplier Toolkit to help with making a caGrid Analytical Service will accompany caGrid 0.5 download

Slide 44

Analytical Service Creation Wizard

Slide 45

Method Implementation Insert strategy code here

Slide 46

Test bed Infrastructure caGrid 0.