tranSMART Community Meeting 5-7 Nov 13 - Session 3: The TraIT user stories for tranSMART
1 like1,371 views
This document provides an overview of the TraIT project and existing demonstrators using tranSMART. It discusses the TraIT roadmap and user stories being implemented at the Netherlands Cancer Institute. Key points include: - TraIT aims to support translational research through integrated data and tools across clinical, imaging, biobanking and experimental domains. - Existing demonstrators using tranSMART include DeCoDe (colorectal cancer) and PCMM (prostate cancer). - The roadmap involves enhancing tranSMART functionality based on user needs and integrating additional data sources. - At NKI, tranSMART will provide an integrated research data warehouse with clinical and research data from various sources and departments.
tranSMART Community Meeting 5-7 Nov 13 - Session 3: The TraIT user stories for tranSMART
- 1. TraIT user stories for tranSMART tranSMART User Meeting; Paris Jan-Willem Boiten; Jelle ten Hoeve 7 Nov 2013
- 2. Contents • Introduction TraIT project • A taster of the existing tranSMART demonstrators – DeCoDe: colorectal cancer – PCMM; prostate cancer • Current user stories on TraIT roadmap • Implementation within Netherlands Cancer Institute (Jelle ten Hoeve)
- 3. Global positioning of TraIT Facts & figures: • Netherlands (AKA Holland) 300 Km • 40.000 km2 • 17 million people • 8 UMCs ( 150 Km )
- 4. CTMM, TIPharma and BMM offer an integrated approach for innovations in the Dutch health care sector TIPharma: drugs • Translational research on novel pharmaceutical therapies CTMM: diagnosis • Early detection of disease by invitro and in-vivo diagnostics Biomarkers • Target finding, animal models and lead selection • Stratification of patients for personalized treatment • Drug formulation, delivery and targeting • Assessing efficiency and efficacy of medicines by imaging Image guided drug delivery • Image guided delivery of medication • Focus on cancer, cardiovascular, neurodegenerative and infectious /autoimmune disease. • Special Theme focusing on the efficiency of the process of drug development Imaging for regenerative medicine Drug delivery BMM: devices • Smart drug delivery systems • Innovations in contemporary organ replacement therapies • Passive and active scaffolds, including cell signalling functions
- 5. CTMM projects Stroke Heart Failure Breast Arrhythmia Diabetes Kidney Failure Lung Thrombosis Peripheral Vascular Disease Prostate Colon Leukemia Alzheimer Rheumatoid Arthritis Sepsis
- 6. Growth of active participation in TraIT: 2011 2013: increase from 11 26 partners EUR 16 million / 4 years Growing TraIT project team
- 7. TraIT aims to support the translational research process by means of IT Epi/Genetics DNA Variants, Copy Number modifications Transcriptome mRNA, ncRNA miRNA Peripheral Markers Proteins, Metabolites Cells, Microbes Organ Systems
- 8. Patient enters medical center Clinical Procedures Electronic Health Record Imaging Samples Experiments Clinical database Image database Biobank database Experimental data Data Integration External data Scientific Output Downstream analysis Intellectual Property Improved Healthcare
- 9. Connecting initiatives z 13 November 2013 9
- 10. the middle ages the 21st century
- 11. TraIT incentives • Increase efficiency of translational research – End to end workflow – Multicenter studies – Connect initiatives (ESFRI, IMI, national programs, etc) • Cope with data challenges – Volume – Silo’s – Interoperability – Stewardship – (open) access • QA/QC – Improve validity of proof of concepts – Diminish scientific misconduct
- 12. TraIT tools & applications: the landscape Hospital (IT) HIS PACS LIS Samples (IT) BIMS Public Data P s e u d o n y m i z a t i o n Translational Research (IT) data domains clinical data integrated data OpenClinica translational analytics workbench imaging data tranSMART/ cohort explorer NBIA + XNAT biobanking CBM-NL tranSMART/i2b2 dataware house R experimental data e.g. Galaxy, Chipster e.g. PhenotypeDB, coLIMS Galaxy
- 13. Uptake of OpenClinica 55 studies 84 sites 300 users OpenClinica Use number of studies 35 31 studies 30 sites 185 users Pre TraIT effect: all multicenter VUmc studies 30 25 20 15 Also multicenter studies UMCU, UMCN, EMC, Meander MC 10 5 0 1 2 2008 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 calendar years July 2008 Oct 2011 Start DeCoDe OpenClinica Start TraIT OpenClinica Oct 2012 Sept 2013 Today
- 14. TraIT Data Integration Roadmap 2012: Data integration platform evaluation and selection tranSMART 2013: Study driven enhancement of data integration platform using “ready to use” data: => enhanced functionality and robustness (tranSMART++) 2014/2015: Study-driven system integration with TraIT data capturing systems => enhanced interoperability and usability (TraIT platform) 2014/ 2015 2013 2012 2012 2013 2014/2015
- 15. TraIT foundation team Foundation team: • TraIT core development team NKI 2 FTE 2 FTE 4 FTE 2 FTE • Adapt & adopt existing solutions like tranSMART • Distributed Scrum Team • Four core centers and several associated ( ) ones
- 16. Foundation team user stories & epics • User stories are collected for every potential TraIT customer project (large research consortia) • User stories are collected on the TraIT Wiki and broken down in epics that can be taken up by the foundation team • Transformed into an actively maintained TraIT roadmap
- 17. CAIRO studies The Dutch Colorectal Cancer Group (DCCG) provides an excellent infrastructure for the performance of multicentre clinical studies in patients with colorectal cancer CAIRO studies: principal investigator Prof.dr. C.J.A. Punt Collaborative translational research: Prof.dr. G.A. Meijer combine clinical trial information with molecular profiling data
- 18. CAIRO studies Clinical data, e.g.: - TNM staging - gender - age - treatment arm of study Non-omics data, e.g.: - MSI/MSS - MLH1 - KRAS - BRAF Genomics: - Comparative genomic hybridisation microarray (arrayCGH)
- 19. Examine study data Overall survival summary statistics in ‘Results’
- 20. Comparison of different groups Overall survival in subjects with MSI vs MSS
- 21. Comparison of different groups Overall survival in subjects with MSI vs MSS
- 22. Survival analysis Overall survival of subjects < and >60 years of age
- 23. Survival analysis Overall survival of subjects < and >60 years of age
- 24. Comparison of chromosomal alterations between different groups Are there significant differences between two groups, e.g. MSS vs MSI?
- 25. Chromosomal alterations and overall survival
- 26. The Prostate Cancer Crisis: Statistics • Most common cancer in men (>900 K ww cases p.a.) • Every 2.5 minutes a man is newly diagnosed • Every 19 minutes a man dies from prostate cancer • Ageing population Rudolph Guiliani diagnosed at age 56 Andrew Lloyd Webber diagnosed at age 61 Ryan O’Neal diagnosed at age 70 Warren Buffet diagnosed at age 81 26
- 28. Examine study data: summary statistics
- 29. Comparison of readcounts (RNA-Seq) between different groups
- 30. Conclusions demo sessions June-Sep 2013 Praise: • "Oh, wow, you just dragged that in!", "I've never been able to do this“ • "This is already great for exploring data.“
- 31. Conclusions demo sessions June-Sep 2013 But also many new wishes & issues identified: • Improve user interface – Standard navigation for all studies – Zoom in/select (group of) subjects from any plot • Basic functionality for facilitating data exploration to be extended – Better handling of units – Stratification – Combinations • Improve genome/chromosome viewing – Implement standard genome browser • Important data sets are still missing Projects are still not actively using tranSMART
- 32. Further roadmap Current portfolio of projects for a tranSMART implementation: • DeCoDe: Colorectal cancer (demonstrator available) • PCMM: Prostate cancer consortium (demonstrator available) • Maastricht Study: A longitudinal diabetes study • POSEIDON: A national registry for outcome data in lung cancer • NKI: Internal data warehouse Netherlands Cancer Institute • And many more in the queue……. Each project has specific user stories requiring new features Currently app. 200 resulting epics on the roadmap
- 33. Improvement theme: data security Data security is number one concern for principal investigators Inter-study security Intra-study security Study 1 Study 2 Intrusion protection Study …
- 34. Improvement theme: molecular viewing PI / (end)user wet-lab-person tech-operator (bio)informatician PI / (end)user
- 35. Recent work: Include Dalliance Genome browser
- 36. cBioportal example for molecular viewing
- 37. molecular data integration Processed data Import to TranSMART Suitable for molecular data integration Suitable for viewer Suitable for data querying
- 38. Improvement themes: Longitudinal data Observational studies tend to demand flexible identification of patient events Diagnosis Surgery Chemo Timeline of disease progression
- 39. New use cases: sample data Sample order process Biobank Information System CBM-NL Summary data about samples Biobank Information System tranSMART Integration & study workspace Collect sample summary data
- 40. System integration and referenced data Referencing pathology scans based on meta data in tranSMART Automated upload of clinical data from OpenClinica Upload and drill-down into molecular pipelines using tools like R and Galaxy Referencing clinical images based on meta data in tranSMART
- 41. TraIT/tranSMART at the Netherlands Cancer Institute Jelle ten Hoeve
- 42. The Netherlands Cancer Institute • 650 employees • Budget: € 80 million/year • 34 professors • 50 PIs (group leaders) in basic research • 33 PIs in clinical research • distribution among positions in basic research other; 3% group leader; 7% technician; 31% postdoc; 31% PhD student; 29% November 2012 + AvL hospital = Comprehensive Cancer Center
- 43. High Performance Computing at NKI-AvL Infrastructure - 10 High Performance Computers (HPCs) and the Life Science Grid - Each HPC: 32-64 cores, 128-512 GB RAM, 20-40 TB storage - 50 research end users - Linux / Ubuntu, R, Matlab and specialized bioinformatics tools - Support together with IT department Support
- 44. A Research Datawarehouse stores and integrates research data from many data sources across data domains and makes these accessible to researchers. The main challenges for implementing a research datawarehousing are: • • • • • Storage: secure central storage of research data Search and access: govern search of, and data access to, research data Data integration: integrate research data across projects and domains System integration: integrate data from clinical and laboratory software Sustainability: embed into existing IT architecture and into the organization at large To clarify the concept ‘research data’, we define ‘data domains’ and ‘data sources’. Data sources can be categorized into three categories: ‘project’ data sources, ‘registry’ data sources, and ‘workflow’ data sources.
- 45. Translational Research Datawarehouse IT systems and Curated databases Data source Domain Department EZIS (Electronic Hospital Records) Clinical Hospital 8,000 Tumor registry All Dept. of Biometrics PALGA, LMS, MolPA Pathology, Biobanking Dept. of Pathology Pathology, Biobanking Biobanking Core Facility 5,000 Molecular (Clinical) Genomic Core facility 3,000 Clinical and research studies # patients (per year) Array and BAM repositories Many more Ready Domain # patients Kinome Yes Yes BOSOM Yes Clinical, Biobank, Pathology, Molecular Clinical, Biobank, Pathology, Molecular Clinical, Molecular 2,500 NKI295 ART Project MindAct Yes Clinical, Molecular 6,000 80,000 …. Many more 295 8,000 …
- 46. Translational Research Datawarehouse IT systems and Curated databases Data source Domain Department EZIS (Electronic Hospital Records) Clinical Hospital 8,000 Tumor registry All Dept. of Biometrics PALGA, LMS, MolPA Pathology, Biobanking Dept. of Pathology Pathology, Biobanking Biobanking Core Facility 5,000 Molecular (Clinical) Genomic Core facility 3,000 Clinical and research studies # patients (per year) Array and BAM repositories Many more Ready Domain # patients Kinome Yes Yes BOSOM Yes Clinical, Biobank, Pathology, Molecular Clinical, Biobank, Pathology, Molecular Clinical, Molecular 2,500 NKI295 ART Project MindAct Yes Clinical, Molecular 6,000 80,000 End users …. Many more 295 8,000 … TransMart DATA GOVERNANCE - Quality Control Development Support ETLs, ETLs, ETLs Pa ent Selec on Browse / Extract Upload Templates Group leaders (clinical) researchers Researchers Researchers Datamanagers
- 47. What do we expect from our community? • • • • • • A comprehensive Datawarehouse (Clinical + Research data) Active directory and user roles ETL tooling “State of the art” exploration of data and basic analysis Bioinformatician API (TranSMART R/BioC package) Upload support for end users - stepwise data upload Jelle ten Hoeve Project leader NKI Robbert Hardenberg Integration specialist NKI Jan Hudecek Scientific programmer NKI Marco Janssen QQ TraIT WP5 Philips