Email this pageCTSI Collaborator Newsletter Winter 2008 Edition
Drinking from the Scientific Fire Hose: A Sabbatical at the National Human Genome Research Institute
By Thomas A. Pearson, MD, PhD, MPH
An academic leave of absence (a/k/a sabbatical) is one of the less commonly used privileges of an academic career at the University of Rochester Medical Center. In most circumstances, a six-month leave of absence can be requested every seven years with salary support from the University, and if additional salary support can be procured from an external source, a full-year’s leave can be proposed. This brief report describes the one-year sabbatical leave of Thomas A. Pearson, MD, PhD, Albert D. Kaiser Professor of Community and Preventive Medicine, most of which was spent at the National Human Genome Research Institute (NHGRI) on the NIH campus in Bethesda, Maryland.
The planning for a sabbatical leave was already well in place in November, 2006 when a request was received from NHGRI’s Director of the Office of Population Genomics, Teri Manolio, MD, PhD, for an epidemiologist to assist her for 6-9 months in development of programs in Population Genomics at the NHGRI. Her objectives for this position coincided almost identically with a sabbatical plan previously submitted to Dean Guzick. The NIH support for half of salary support was via an Intergovernmental Personnel Act (IPA). The other major support from the NHGRI was for an apartment just outside the security fence of the NIH campus. This then allowed full-time residence in Bethesda from September 1, 2007 to June 1, 2008.
The focus of the sabbatical leave was the rapidly expanding field of population genomics, with its great promise of applicability to all aspects of clinical and translational research. It was also based on the perceived gap in my knowledge and skills, with my last course on genetics being in 1971. I wished to bring new genomic tools to my research, to advise trainees about incorporating genomic tools into their work, and to be able to write and review grant applications in this area. To prepare for this, July and August of 2007 were spent reading on this topic and participating in a two-week course on human and mouse genomics at the Jackson Laboratory in Bar Harbor, Maine. In addition, the NIH in Bethesda offers a myriad of excellent courses, workshops, etc. that assisted in my upgrade of knowledge and skills in genomics.
To break into this field, it was decided that genome-wide association studies (GWAS) would be an excellent target for initial study. This research has been deemed “Scientific Breakthrough of the Year for 2007” and has provided numerous discoveries into the genetic basis for health and disease. The plan was initially to review the first 100 GWAS (eventually 109 GWAS were reviewed), from an epidemiologist’s perspective. This yielded a manuscript in the March 18 issue of JAMA (Pearson T, Manolio T; How to Interpret a Genome-wide Association Study. JAMA 2008; Vol. 299, No. 11; 1335-1344). Two courses were also developed, each consisting of eight lectures. The first, entitled “Genetics for Epidemiologists” was designed for clinical and population researchers who wish to refresh their knowledge of the basic science aspects of genomics. The eight-lecture course was presented by two of us (T. Manolio and T. Pearson) at Northeastern University in May 2008 and is currently on the NIH website. The second, entitled: “Epidemiology for Investigators Performing Population-wide Genomic Studies,” emphasized aspects of study design, analysis, and interpretation for persons largely performing the basic science aspects of human genome research. This eight-lecture course was presented at the NIH in July, 2008 with over 350 people attending in person or online. In addition, we have a poster for presentation at the American Society of Human Genetics in November, 2008 and have one or two additional manuscripts planned in the area of study design and methods in genome-wide association studies.
Placement at the NIH offered many additional opportunities. I was able to represent the Rochester Clinical and Translational Science Award at a number of meetings and was able to get to know many of the National Center for Research Resources administrators involved with that award. I reestablished many acquaintances with the National Heart Lung and Blood Institute, served as Chair of the Working Group on their new Guideline Development Program, presented a seminar to their Prevention and Population Health branch, chaired a review panel, and participated in a strategic planning exercise for their Women’s Health Initiative Program. Additional meetings with investigators from National Center for Deafness and Other Communication Disorders and the National Eye Institute allowed planning for further projects with our National Center for Deaf Health Research. I was able to submit and have funded a grant application to supplement our CTSA to establish a National CTSA Educational Research Program, centered here at the University of Rochester.
My overwhelming impression is that an academic leave is a critically important tool to upgrade a faculty member’s knowledge and skills or provide time for creative pursuits such as books, grant applications, etc. I would strongly attest that the National Institutes of Health is a superb site to carry this out, with a tradition of excellence and a real sense of excitement for discovery. An NIH sabbatical also provides insights into the process of development and funding of federal research programs. An NIH sabbatical also has two mechanisms to support the additional six months, either with an IPA or by a K Award. As the primary source of research funding for the University, such placements should be encouraged as a University investment in faculty members who seek to expand their involvement with NIH Programs and Awards.
New Translational Technology opens to Faculty on December 15th
By Mark Plessinger, PhD
The CTSI Translational Technology Key Function (Dr. Steve Welle, Director) and the CTSI Laboratory Support Center (Dr. Mark Plessinger, Director) announce the availability of new equipment to the University of Rochester faculty. The newly renamed Confocal and Conventional Microscopy (CCM) Core is home for a new, highly sophisticated and state-of-the-art microscope: the Olympus FluoView™ FV1000 Laser Scanning Confocal Microscope (figure left below). This instrument is capable of both standard confocal imaging of fixed, as well as advanced live imaging capabilities. The microscope is currently undergoing its “shakedown cruise” and will be available to all University users by December 15, 2008. Dr. Linda Callahan, Director of the CCM Core, states that the new FV1000 instrument is operational and is equipped with high power objectives, motorized stage, and is capable of generating microscopic images that contain up to four colors, obtained simultaneously (see figure below). This sensitive instrument enables the user to visualize and image multiple cellular structures, cellular proteins, as well as time-dependent live cell changes and responses. The FV1000, purchased by the University, is also capable to utilize a synchronized laser scanning system: while one laser stimulates, the second laser simultaneously provides high-resolution imaging. This coordination of laser stimulation and imaging makes the URMC FV1000 an ideal choice for FRET, FRAP, FLIP and photoactivation. A stage incubator will soon be added to the instrument to promote live cell imaging over significant time periods under optimal conditions. To facilitate use of this new instrument, the CCM Core will sponsor workshops and meetings with Olympus representatives to familiarize Faculty, fellows, and students in the features and operation of the FV1000.
Also available in the CCM Core is StereoInvestigator, advanced state-of the art software for performing designed and user interfaced stereological analysis and features tools for anatomic mapping, cell distributions, and cell counting. Other services include a workstation equipped with identical software as the FV1000, a Leica SP1 confocal microscope with SP1 workstation and standard brightfield and fluorescent microscopy utilizing a MicroFire camera and PictureFrame for image capture. Information resources, including consultations for histologic processing, imaging, imaging processing, and networking among microscopy users at the University of Rochester are available. The CCM Core is housed in two facilities: Facility I is located in the fourth floor S-wing in 4-7546, and houses the FV1000; Facility II is located in the 1-2100 area of the Hospital and houses all other CCM Core equipment. For more information contact Linda Callahan email: Linda_Callahan@urmc.rochester.edu or x5-1317 / x3-3163.
CTSI Biostatistics Group Develops New Methodologies
By Sally Thurston, PhD
Under the auspices of the CTSI, the Department of Biostatistics and Computational Biology has developed original software for several novel methodologies, which may be useful to CTSI more broadly. The software developed is available to University of Rochester investigators through the University of Rochester CTSI consulting service. Investigators who would like more information or are potentially interested in applying one of the methods to his or her application are encouraged to contact Susan Messing at the Biostatistics consulting service, Susan_Messing@urmc.rochester.edu . A brief description of the methodologies is given below.
Constructing Multivariate Prognostic Gene Signatures- The development of novel methods and algorithms to construct multivariate prognostic signatures based on high-dimensional data, including gene arrays, has resulted in more than one operational algorithm, along with associated specialized software. The initial software implementations fit Cox proportional hazards models to right-censored time-to-event survival data when the number of candidate predictors is very large (hundreds, thousands, or more), but the same ideas can also be implemented in other regression settings, including logistic regression for discriminating between 2 groups (e.g., healthy vs. diseased). Investigators should be aware that the required sample sizes are larger than those for investigating simple, unadjusted relationships of each predictor with the outcome; for logistic regression, at least 25-50 observations per group is recommended, and performance improves with 100-200 per group. Simulation studies suggest that these multivariate methods can identify significantly more true predictors, with fewer false discoveries, compared with univariate testing or a stepwise search. A related Technical Report is on file in the Department of Biostatistics and Computational Biology (Peterson, 2006).
• Peterson, D. (2006). Constructing Multivariate Prognostic Gene Signatures with Censored Survival Data. (Technical Report 06/09).
Sensitivity and Specificity with a Time-varying Reference Standard within a Longitudinal Setting- Diagnostic tests are used in a wide range of behavioral, medical, psychosocial, and health-care related research. Test sensitivity and specificity are the most popular measures of accuracy for diagnostic tests. Available methods for longitudinal study designs assume fixed gold or reference standards and as such do not apply to studies with dynamically changing reference standards, which are especially popular in psychosocial research. We have developed a novel approach to address missing data and other related issues for modeling sensitivity and specificity within such a time-varying reference standard setting. The approach is illustrated with real data in sexual health research (Yu et al. 2008).
• Qin Yu, Wan Tang, Sue Marcus, Yan Ma and Xin Tu (2008). Modeling Sensitivity and Specificity with a Time-varying Reference Standard within a Longitudinal Setting. Under revision for Applied Statistics.
Mixture modeling of microarray data- We have developed a set of statistical algorithms and associated software for the analysis of microarray data, based on Bayesian hierarchical models. The algorithms permit reporting of standard error measures such as the False Discovery Rate or Family-Wise Error Rate, but allow further types of analysis such as sample size estimation based on pilot data. Computational fitting methods are generally simpler than is usually the case for such models, and can easily be incorporated into, for example, bootstrap algorithms to permit formal inference statements. The algorithms are implemented in an R library.
Exposure effects on multiple outcomes nested in domains- There is a need for more powerful methodologies to detect small but potentially important effects of low-dose exposure on multiple correlated outcomes, such as those that characterize childhood cognition and behavior. This task is complicated by the fact that the multiple outcomes may be measured on different scales, and may be manifestations of a smaller number of broad outcome classes, or “domains”. For example a neurodevelopmental test battery typically contains multiple tests, some of which measure motor function, while others measure cognition or other aspects of human behavior. Estimation of the exposure effect on each domain may also be of interest. Our Bayesian model and associated software incorporates data on all outcomes, exposure, and covariates in a single model that allows the exposure effect to differ across domains and across outcomes within domains. Covariate effects are also allowed to differ across outcomes and domains. Our model may also be useful for applications that do not involve exposure to toxicants, and can result in a sizeable increase in power as compared to separate models. The model assumes that all outcomes are continuous, that linear regression would be appropriate for modeling the covariate-adjusted exposure effect on each outcome separately, and requires at least 3 outcomes per domain. Missing outcome data can be handled in our model, with certain restrictions, thereby allowing exposure effect comparisons on the same subjects (Thurston et al., 2008).
• Sally W. Thurston, David Ruppert and Philip W. Davidson, “Bayesian models for multiple outcomes nested in domains”, accepted for Biometrics.
Emergency Medicine Network Holds Annual Retreat
By Dr. Thomas Richardson
The Emergency Research Network in the Empire State (ERNES) held its second annual retreat on November 3rd at Upstate Medical University in Syracuse. Department chairs, research directors, researchers, and research staff from Albany Medical Center, Bassett Healthcare, Erie Country Medical Center, Guthrie Health, Upstate Medical University and the University of Rochester met to review the research networks progress and plan for the coming year. Development of the ERNES was stimulated by the need for high quality and efficient emergency medicine relevant epidemiologic, health services, clinical and translational research in order to improve the care of the diverse patient populations served.
During the past year the ERNES has developed the new collaboration, created it “bylaws” and operational procedures, and invested in infrastructure development to conduct multicenter studies. Efforts also included the site’s Institutional Review Boards working together to drastically reduce the time it takes to obtain study approval in the Network. Creating unique patient enroller programs at each site to meets the recruitment needs in each ED has also been an important step in developing the research infrastructure. Projects that the ERNES is involved in includes recruitment of patients with mild traumatic brain injury, and developing a knowledge translation project that evaluates the effects of an educational campaign on provider attitudes and behaviors.
The remainder of the retreat was spent on developing plans for continued development of the research infrastructure and developing plans for sustainability of the Network. A core component of the plan is to leverage information technology into the research infrastructure in to order to maintain a competitive advantage of being a nimble and efficient research network. The Network aims to respond quickly to research proposals and rapidly implement research protocols that involve primary data collection, use of secondary data sources across institutions, and provider surveys, all critical elements to conducting high quality translational research. Because the Network sees nearly 1,500 ED patients a day, the recruitment time for studies is also reduced and the diversity of subjects is enhanced.
The ERNES is funded under the pilot collaborative research grant program from the CTSI and the Foundation for Healthy Living. The Institutions are part of the Upstate New York Consortium for Healthcare Research and Quality (UNYCHRQ) and the Upstate New York Translational Research Network (UNYTRN). For more information on UNYCHRQ / UNYTRN please visit the Foundation for Healthy Living’s website at
http://www.foundationforhealthyliving.org/unychrq.php.
CTSI Calendar of Events
December 2008
• December 2nd– RCTRC Seminar Series, 12:15-1:15 pm, Whipple Auditorium (2-6424), “Deconstructing Carcinogenesis– Rational System Approaches Towards Interventions” presented by Dr. Hucky Land
• December 9th– RCTRC Seminar Series, 12:15-1:15 pm, Whipple Auditorium (2-6424), “The Nasonia Genome: Applications to Understanding the Genetics of Complex Traits” presented by Dr. Jack Werren
• December 9th– Faculty Development Workshop, 2:00-4:00pm, Louise Slaughter Room (1-9555), “Career Development for Medical Educators” presented by Dr. Ralph Jozefowicz
• December 10th– Leadership Series, 5:30-7:30 pm, Natapow Room (1-9545), “Leadership of Major Academic Teaching Hospitals” presented by Dr. Steven Goldstein, President & CEO of Strong Memorial and Highland Hospitals
January 2009
• January 13th– Faculty Development Workshop, 2:00-4:00 pm, Louise Slaughter Room (1-9555), “Giving Feedback that Works” presented by Dr. Denham Ward
• January 20th– RCTRC Seminar Series, 12:15-1:15 pm, Whipple Auditorium (2-6424), Presenter TBD, Topic: Cardiovascular IDP Session 1
• January 21st– Leadership Series, 5:30-7:30 pm, Natapow Room (1-9545), “Managing and Leading Professional People” presented by Dr. Henry Tung
• January 27th– RCTRC Seminar Series, 12:15-1:15 pm, K-307 Auditorium (3-6408), Presenter TBD, Topic: Cardiovascular IDP Session 2.
February 2009
• February 3rd– RCTRC Seminar Series, 12:15-1:15 pm, Whipple Auditorium (2-6424), Presenter TBD, Topic: Nanomedicine ISP Session 1.
• February 10th– RCTRC Seminar Series, 12:15-1:15 pm, Class of 62 Auditorium (G-9425), Presenter TBD, Topic: Human Subject’s Protection Program Presentation
• February 10th– Faculty Development Workshop, 2:00-4:00 pm, Louise Slaughter Room (1-9555), “Small Group Teaching” presented by Dr. Denham Ward and Dr. Barbara Davis
• February 11th– Leadership Series, 5:30-7:30 pm, Natapow Room (1-9545), “Leadership– A Chair’s Perspective” presented by Dr. Nina Schor
• February 17th– RCTRC Seminar Series, 12:15-1:15 pm, LeChase Auditorium (G-9576), Presenter TBD, Topic: Nanomedicine ISP Session 2.
• February 24th– RCTRC Seminar Series, 12:15-1:15 pm, Whipple Auditorium (2-6424), Presenter TBD, Topic: Infectious Disease / Immunology IDP
• February 24th– Faculty Development Workshop, 2:00-4:00 pm, Louise Slaughter Room (1-9555), “The Teachable Moment: Teaching in the clinical setting” presented by Dr. Scott Tripler, Dr. Kathryn Markakis, and Dr. Karl Illig
In Recognition of Excellence
Susan Gross Fisher, M.S., Ph.D., chair of the Department of Community and Preventive Medicine and co-director of the CTSI Design, Biostatistics and Clinical Research Ethics Key Function, recently received a Presidential Citation for her role in a collaborative, national study of head and neck cancers. She is an accomplished researcher in caner epidemiology, and currently serves on the editorial board at two scientific journals, Journal of Clinical Oncology and Leukemia Research.
Stephen Dewhurst, Ph.D., professor of Microbiology and Immunology, senior associate dean for Basic Research, and member of the CTSI Executive Committee has received the University’s William H. Riker Award for Excellence in Graduate Teaching. Given annually, the award is the University’s highest honor in graduate education. Well-known for his research contributions to HIV vaccine development and neuroAIDS, Dewhurst is also recognized for his passion as an educator. He has served as Ph.D. thesis mentor to 20 graduate students, as well as many students in master’s degree programs.
CTSI KL2 Scholar, Jennifer Carroll, MD, PhD, recently won career development awards from the National Institutes of Health and the American Cancer Society. Her research goal is to study primary care based communication interventions to promote physical activity in underserved populations.
