i2b2: Informatics for Integrating Biology & the Bedside - A National Center for Biomedical Computing
Collaborations Scientific Collaborations

During the years while i2b2 is finalizing both its prototype Clinical Research Chart (CRC; see Hive Page) and associated stand-alone analytical tools for dissemination to the national community, we are engaged in several ad hoc collaborations that inform our development activities through valuable insights and active input from the wider community.

To give you a flavor of what we're up to with these various intersections which "go beyond" our original scope, we offer the following catalog of activities in which we endeavor to:

Collaborate within the NCBC Network

A project within one of our original DBPs (Brain Structure and Long Term Treatment with SSRIs in Major Depressive Disorder, Dan Iosifescu, MD, MSc, PI; see their i2b2 page) collaborated directly with NAMIC investigator Martha Shenton, PhD, to apply state-of-the-art algorithms developed by her Psychiatry Neuroimagine Laboratory to raw images captured from routine clinical records by i2b2 phenotypic extraction methodologies.

i2b2 continues to actively participate in the NCBC-wide forum for the purpose of enabling the widest possible user access to the collective network's toolkits.  These activities include posting of Biositemaps at each Center's website, a dedicated volume of JAMIA showcasing the network's contributions, and presentations at relevant professional meetings regarding network resources.

Support widespread adoption of the i2b2 software platforms

The advent of the CTSA (Centers for Translation Science Awards) program has provided a unique opportunity to share our model for using existing clinical repositories with academic health centers interested in new approaches to translational research.  

  • i2b2 has worked closely with the main hospitals within the Harvard Catalyst CTSA (Mass General, Brigham and Women's Hospital, Beth Israel Deaconness Medical Center, Children's Hospital Boston, Dana Farber Cancer Center) to facilitate their adoption of i2b2 instances as a vehicle for conducting clinical research at these heterogeneous biomedical research institutions and as a prelude to sharing aggregate data by means of a web-based, federated query system (SHRINE).  

  • i2b2 has actively participated in the design and implementation of the SHRINE (Shared Health Research Informatics Network) system, the software for which is currently available open source (https://open.med.harvard.edu/display/SHRINE/Software) at:

    • Harvard Catalyst CTSA for the sharing of aggregate data counts for demographics, diagnoses, meds and labs.

    • In a collaboration with the West Coast consortium CICTR (U Wash, UCSF, UCDavis) to enable their federation for the sharing of aggregate data totals. 

    • A pan European public-private consortium which is currently evaluating the SHRINE platform to support their drug discovery efforts. 

    •  A supplemental CTSA award to Harvard Catalyst is supporting the deployment of a SHRINE alliance across the i2b2 instances at UTH, UW, UCSF, UMichigan, and Wake Forest Baptist Medical Study for a study on the co-morbities of autism.

  • As an extension of our broad partnership with Morehouse School of Medicine, we were very pleased to support their interest in incorporating the CRC prototype into the clinical infrastructure proposed in their CTSA application.   This effort was facilitated by support from Recombinant Data Corp.

  • OurAcademic Users' Group (clinical champions and informatics implementers at the academic health centers which are installing the i2b2 software) continues to collaborate and synergize on improvements to existing software, to identify new Hive cells of mutual interest, and to make these freely availabe to the user community. Membership currently includes over 275 individual members from 50+ institutions, including over half of the CTSAs and six international sites.  See AUG  for details.

Expand our mandate as a NCBC through new "docking" funding

  • Second round DBP PIs Roy Perlis and Jordan Smoller have built on their i2b2 work in Major Depressive Disorder to win two independent RO1s:

    • R01 MH086026, Building a Risk Stratification Model for Treatment Resistance in Major Depressive Disorder, PI Roy Perlis. See also: International Antidepressant Response Consortium (ARC) www.i2b2.org/work/arc.htm.

    • R01 MH08542, Internaltional Cohort Collection for Bipolar Disorder, PI Jordan Smoller.

  • Susanne Churchill, Ph.D, Anthony Amato, MD. : The i2b2 team has been awarded an industry sponsored DBP to use our data mining methodology to search for unique gene signatures in a characterized disease subphenotype.

  • Shawn Muphy, M.D., Ph.D.: 3UL1RR025758-02S Subcontract, Sharing Clinical Images using i2b2 and SHRINE, supports the development and dissemination of additions to the i2b2-based bioinformatics infrastructure that will allow clinical imaging data from sophisticated medical imaging modalities to be used for secondary reserarch purposes.

  • Gary Gibbons, M.D., Morehouse School of Medicine: R01ES017656 Vascular Epigenomic Dynamics in African-American Hypertensives.

  • Shawn Murphy: MURPHY090MOPO Distributed research partner collaboration with the observational medical outcomes parthership (OMOP).

  • Katherine Liao, MD: Am. College of Rheumatology Sci. Dev. Award, Genetic Risk Factors for Coronary Artery Disease in Rheumatoid Arthritis, studied whether genetic risk factors associated with traditional risk factors of CAD in the general population are present in RA cases.

  • Elizabeth Karlson, MD: K24AR052403 CR, Genetic Predictors of Early Atherosclerosis in SLE, provides protected time for the PI's patient oriented research training program that leverages the i2b2 RA DBP to extract clinical data or diagnostic features through database mining and NLP.

  • Gordon Williams, MD: R01 HL094452 Aldosterone, Histone Demethylase and Cardiovascular Disease

  • Isaac S. Kohane, M.D., Ph.D., Shawn N. Murphy, M.D., Ph.D. et al: HHS-2010-ONC-005 ( SHARP Healthcare Application and Network Platform Architectures): SMArt (Substitutable Medical Applications, reusable technologies) is developing web services that support analytic tools to maximize the value of electronic medical record data as well as personal health records for studies related to comparative effectiveness, quality reporting, evidence-based views for patient research, and medical outcomes.

  • Robert Plenge, MD, PhD, Elizabeth Karlson, M.D.: U01 GM092691, Genetic Predictors of Response to Anti-TNF Therapy in Rheumatoid Arthritis, is part of the PGRN network and focuses on genotyping of new samples and developing new statistical methodologies for analysis of response to anti-TNF therapy in patients with RA.  The selection algorithm devleoped in the original RA DBP has been applied to patient databases at Northwestern and Vanderbilt for access to additional samples.

  • Elizabeth Karlson, M.D.: R01 AR04880, Clinical Risk Prediction Modeling in Rheumatoid Arthritis, is developing a model to predict who is at risk for developing RA based on information on environmental, behaviorl and clinical risk factors, as well as genetic factors, that will lead to risk factor modification and earlier introduction of effective therapies in RA.  Dr. Karlson is working with the i2b2 NLP team to develop standards for a patient functional status classifier.

  • Katherine Liao, M.D.: K08 AR060257, Genetic Risk Factors for Coronary Artery Disease In Rheumatoid Arthritis, is conducting a genetic association study to (1) determine how genetic markers associated with traditional risk factors and CAD risk in the general population related to CAD risk in RA, (2) study whether genetic risk factors for developing RA markers of immune dysregulation also increase risk for CAD in RA, and (3) test whether inflammatory mediators thought to accelerate atherosclerosis in RA, increase risk for CAD.

  • While not directly docking with i2b2, many new awards have been made to academic health centers to expand or apply their i2b2 instances, including:

    • Laura Schanberg, MD CARRA: Accelerating toward an evidence based culture in pediatric rheumatology RC2AR058934 60 site i2b2-SHRINE network to study pediatric rheumatology
    • Elmer Bernstam RC1RR028254-0110Concept-Level Methods for Comparative Effectiveness Research
    • Nancy Kressin, MD & Bill Adams, MD RC2HL101628-01 Massachusetts Health Disparities Monitoring System
    • Jay Moskowitz, MD RC2L010796-0110 An open source research permissions framework for South Carolina
  • Ross Lazarus, Ph.D., co-PI of our original Asthma DBP is PI of a NHGRI RO1 (HG003646) "A Genetic Association Research Statistical Framework".  As proposed in the application, i2b2 faculty member Dr. Lazarus will disseminate his software tools (for importing experimental data and genomic annotation methods; for statistical power calculation and for selecting maximally informative subsets of markers during experimental design; for visualizing and summarizing experimental results for established and recently developed methods supporting statistical inference on single markers, multiple markers and epistatic and gene by environment interactions) through the i2b2 portal as Hive cells.

  • Marco Ramoni, Ph.D., was awarded an NHGRI RO1 (HG003354) "Decoding Gene Expression Control Using Conditional Clustering by Dynamics" to develop computational methods to identify cellular control mechanisms from genome-wide experiments and disseminate these through the i2b2 portal.

  • Core 1 colleague Shamil Sunyaev, Ph.D., is PI of a NIGMS award (GM078598) "New Methods and Enhanced Software for Predicting Functional SNPs".  Dr. Sunyaev is working with Dr. Inna Dubchak  (photo, left) at VISTA to develop improved SNP targeting methods which will be made available via the i2b2 portal as both a stand alone program and as a Hive cell.  We were able to work directly with Dr. Dubchak to expand the list of target genes in our Hypertension DBP and, as a further result, were pleased to collaborate in the submission of a newly funded second RO1, "Comparative Genomics of Non-coding Regions to Facilitate Translational Research" (RO1HL091495).

     
  • Diabetes DBP leader Mary Elizabeth Patti, M.D., participated on an subcontract with James Lyons-Weiler, Ph.D., at the University of Pittsburgh to develop robust proteomic profiles that are predictive of various insulin resistant states leading up to diabetes mellitus.

Explore innovative ways to acquire tissue samples for linking genomic analyses to clinical records w/o direct consent

i2b2 continues to actively work with a novel system under development at Brigham and Women's Hospital, Boston, to use discarded clinical blood samples for genetic evaluation and linkage to anonymized clinical records.  Because both records and samples are anonymized (but linked through the anonymized id's), patient privacy is strictly maintained.  This concept, which has been instrumental to all of our DBPs, has led to an award from the NLM "Crimson-i2b2 integration for high-throughput, scalable sample collection" (R01 LM010100)  by Lynn Bry, M.D, Ph.D, at BWH to develop a workflow Hive Cell for the CRC.  i2b2 Team Member Brian Wilson is actively supporting this project.

Develop Relevance Network Software that will enable healthcare systems to identify highly significant and novel associations across the entire spectrum of healthcare data.  Zak Kohane has been collaborating with Vlad Valtchinov, Ph.D. (right, with daughter Yvana) to develop a new Hive Cell that will test the hundreds of millions of potential associations across clinical diagnoses and laboratory findings (including clinical genetics) for statistical significance at a level which merits additional follow-up.  Preliminary studies have suggested several novel correlations between clinical findings and metabolic pathways that may suggest new disease classifications and therapeutic targets.  This work was initially supported by funding from Partners HealthCare IT.
Expand our Education and Outreach Program

  • In order to stabilize funding for – and grow - our collegiate Summer Institute in Bioinformatics and Integrative Genomics, we have combined forces with a Ph.D. program under the direction of Zak Kohane which is administered by the Harvard-MIT Division of Health Sciences and Technology and funded by NHGRI.  In the recent competing renewal to the NHGRI, we received six slots for URM summer students.  Coupled with i2b2 funding this gives us a total of 12-16 slots in our undergrad program.

  • Formation of the Boston Area Diversity Action Plan (DAP) Working Group – as a spinoff of NHGRI's focus on the importance of DAPs for entities working at the genomics interface, several of the local programs which sponsor undergraduate, post baccalaureate, and postdoctoral training that emphasize inclusion of URMs and first generation college students have come together to share approaches, planned lectures and seminars, social activities and, where appropriate, applicants.  Current membership includes i2b2, the Harvard-MIT Health Sciences and Technology Bioinformatics and Integrative Genomics Ph.D. T32, the Diversity Initiative for Scientific Research at the Broad Institute, Model Organism Databases (Harvard University, Jackson Lab, Stanford University, Cal Tech, and University of Oregon Consortium), the CEGS at Harvard Medical School, and the CEGS at the Dana Farber Cancer Institute.

  • We have collaborated with NHGRI to sponsor our i2b2 Minority Postdoctoral Fellowship Program, which has supported two fellows. One of these, Dr. Adam Davis, subsequently was awarded a K99 from the NHLBI and is now completing the R00 portion of the K99 as an Assistant Professor of Medicine at Morehouse School of Medicine.  He also continues to serve as the Teaching Assistant for our Summer Institute.

[ back to top ]
Home | Contact | Sitemap | Search
©2017 Partners Healthcare