Download My NIH Biosketch - Department of Biostatistics

BIOGRAPHICAL SKETCH
Provide the following information for the Senior/key personnel and other significant contributors.
Follow this format for each person. DO NOT EXCEED FIVE PAGES.
NAME: Li, Yisheng
eRA COMMONS USER NAME (agency login): YISHENG
POSITION TITLE: Associate Professor
EDUCATION/TRAINING (Begin with baccalaureate or other initial professional education, such as nursing,
include postdoctoral training and residency training if applicable.)
INSTITUTION AND LOCATION
Peking University, Beijing, China
Sun Yat-sen University, Guangzhou, China
University of Toledo, Toledo, OH
University of Michigan, Ann Arbor, MI
DEGREE
(if applicable)
BS
MS
MS
PHD
Completion Date
MM/YYYY
7/1991
6/1994
8/1998
12/2003
FIELD OF STUDY
Probability & Statistics
Probability & Statistics
Statistics
Biostatistics
A. Personal Statement
I have a broad background in biostatistics, with specific training and expertise in longitudinal data analysis and
study design, Bayesian adaptive clinical trial design, Bayesian nonparametrics, missing data methods, and
mediation analysis. I have served as a co-investigator or primary statistician on over 30 externally-funded (in
particular, NIH-funded) grants in behavioral science, health disparities research, and clinical trials, including as
director of the biostatistics and data management core of a P60 Center of Excellence Grant, and a key
statistical collaborator in a Stand Up To Cancer Dream Team Translational Research Grant. I have been an
ex-officio of the Institutional Animal Care and Use Committee (IACUC) at MDACC since 2010, responsible for
reviewing biostatistical aspects of the Animal Care and Use Forms (ACUFs). I currently serve as a
collaborator on the PI’s ongoing NIH-funded PREVENT Cancer Program. I have overseen the statistical
design, conduct and analysis, and assisted in the interpretation of analysis results and preparation of
manuscripts, in the above mentioned collaborative projects.
1. Cofta-Woerpel L, McClure J, Li Y, Urbauer D, Cinciripini PM, Wetter DW. Early cessation success among
women attempting to quit smoking: trajectories and volatility of urge and negative mood during the first
post-cessation week. Journal of Abnormal Psychology 2011; 120(3):596-606. PMCID: PMC3153568.
2. Basen-Engquist K, Carmack CL, Li Y, Brown J, Jhingran A, Hughes DC, Perkins HY, Scruggs S, Harrison
C, Baum G, Bodurka DC, Waters A. Social cognitive theory predictors of exercise behavior in endometrial
cancer survivors. Health Psychol 32(11):1137-48, 11/2013. PMCID: PMC4057057.
3. Mayer IA, Abramson VG, Isakoff SJ, Forero-Torres A, Balko JM, Kuba MG, Sanders ME, Yap JT, Van den
Abbeele AD, Li Y, Cantley LC, Winer E, Arteaga CL. A SU2C phase Ib study of pan-PI3K inhibitor
buparlisib with letrozole in ER+/HER2-negative metastatic breast cancer. Journal of Clinical Oncology
2014; 32(12):1202-1209. PMCID: PMC3986383.
4. Vidrine JI, Spears CA, Heppner W, Reitzel LR, Marcus MT, Cinciripini PM, Waters AJ, Li Y, Nguyen NT,
Cao Y, Tindle HA, Fine M, Safranek LV, Wetter DW. Efficacy of mindfulness based addiction treatment
(MBAT) for smoking cessation: A randomized clinical trial. Journal of Consulting and Clinical Psychology.
2016 May 23. PMID: 27213492 PMCID: PMC5061584.
B. Positions and Honors
Positions and Employment
1998-1999
2002
2002
2004-2010
2008
Parke-Davis Pharmaceutical Research (Pfizer Inc.), Ann Arbor, MI, Research Assistant
Wyeth Research, Collegeville, PA, Summer Intern
Boron, Lepore & Associates, Cardinal Health, San Diego, CA, Consultant
Assistant Professor, Department of Biostatistics, Division of Quantitative Sciences, The
University of Texas MD Anderson Cancer Center, Houston, TX
Visiting Assistant Professor, Department of Statistics, Rice University, Houston, TX
2010
Visiting Assistant Professor, Department of Statistics, Rice University, Houston, TX
2010-present Associate Professor, Department of Biostatistics, Division of Quantitative Sciences, The
University of Texas MD Anderson Cancer Center, Houston, TX
Other Experience and Professional Memberships
1999-2002
2003
2006-2007
2008
2009
Research Assistant, University of Michigan, Ann Arbor, MI
Research Assistant, University of Michigan, Ann Arbor, MI
President, Houston Area Chapter, American Statistical Association
Program Committee Member, Joint Statistical Meetings, American Statistical Association
Center for Scientific Review Cancer Biomarkers Study Section -- Special Emphasis Panel #10,
NIH, Internet Assisted Reviewer
2010-2016
Member of Editorial Board, Journal of Biometrics and Biostatistics
2011-present Associate Editor, BMC Medical Research Methodology
2012-present Associate Editor, The American Statistician
2015-present Mathematics and Statistics Discovery Grant, NSERC, External Reviewer
Honors
1988
1992
1994
Merit-award recipient, Peking University, Department of Probability and Statistics
Merit-award (first prize) recipient, Sun Yat-sen University, Department of Mathematics
Title winner of "Outstanding Graduate Student," Sun Yat-sen University, Department of
Mathematics
2003
Fellowship, University of Michigan, Department of Biostatistics
2007
Travel Award, Current and Future Trends in Nonparametrics conference, Columbia, SC
2008
Early Career Researcher Travel Award, 9th World Meeting of the International Society for
Bayesian Analysis (ISBA), Hamilton Island, Australia, ISBA
2009-present Acceptance into Cambridge Who's Who among Executives, Professionals and Entrepreneurs,
Cambridge Who's Who
2009
Best Poster Presentation, The International Workshop on Objective Bayes Methodology,
Philadelphia, PA, The International Society for Bayesian Analysis
2009
Young Researcher Travel Award, The 11th Annual Winter Workshop on Semiparametric
Methodology, Gainesville, FL
2010-2011
Ranked #3 for a few months in the 10 most accessed Biometrics papers, The International
Biometric Society
C. Contribution to Science
1. The uniform shrinkage priors (USPs) for the random-effect variance or covariance matrix in Bayesian linear
or generalized linear mixed models were developed in 1999 and 2000. This approach provided an
excellent alternative to the commonly used diffuse proper priors, which are known to have issues of
posterior near-impropriety and sensitivity of the inference to the prior specification. Despite the satisfactory
performance of the USPs, this approach has not been widely used because it was only developed for
simple two-stage hierarchical models, due to both conceptual and technical difficulties in extending the
method to more general hierarchical models. I extended the definition of the USP to semiparametric mixedeffects models and showed desirable properties of the new prior both analytically and empirically via
simulations. Further extension of this prior specification in a general class of mixed-effects models is
currently underway. In all the work, I developed the original idea of the extension and carried out most of
the modeling and implementation work that has led to the publication of the following paper.
a. Li Y, Lin X, Müller P. Bayesian inference in semiparametric mixed models for longitudinal data.
Biometrics 2010; 66(1), 70-78. PMID: 16984320. PMCID: PMC3081790.
2. Nonparametric modeling of an unknown distribution, in particular, using the Dirichlet process prior, is at the
center of robust inferences in Bayesian hierarchical models. When this approach is used in random-effects
model, an identifiability issue arises which, if not addressed, will result in incorrect and seriously misleading
inferences for the fixed effects that are often of scientific interest. I formally pointed out the issue and
proposed a simple yet effective adjustment to inference based on analytic evaluation of the posterior
moments of the fixed effects that addressed the identifiability issue. This work has solved a long-standing
identifiability problem in nonparametric Bayesian inference using the Dirichlet process prior, thus allowing
its wide use for valid inferences in Bayesian hierarchical models.
a. Li Y, Müller P, Lin X. Center-adjusted inference for a nonparametric Bayesian random effect
distribution. Statistica Sinica 2011; 21(3), 1201-1223. PMCID: PMC3870168.
3. Dose finding and dose-schedule finding are critical early steps in the development of drugs that treat
cancer. Along with colleagues, I have developed efficient and robust designs for dose-finding and doseschedule-finding trials that have improved the drug development process. Examples of our proposed
designs include the modified toxicity probability interval-based designs, which have been routinely used by
practitioners including major pharmaceutical companies in their dose-finding trials. I have also mentored
trainees (marked by a * below) in developing other novel and efficient designs for dose- and doseschedule-finding trials.
a. Li Y, Bekele BN, Ji Y, Cook JD. Dose-schedule finding in phase I/II clinical trials using a Bayesian
isotonic transformation. Statistics in Medicine 2008; 27(24), 4895-4913. PMCID: PMC4562497.
b. *Guo B, Li Y. Bayesian designs of phase II oncology trials to select maximum effective dose
assuming monotonic dose-response relationship. BMC Medical Research Methodology 14:95, 2014.
PMCID: PMC4119476.
c. *Guo B, Li Y. Bayesian dose-finding designs for combination of molecularly targeted agents
assuming partial stochastic ordering. Statistics in Medicine 2015; 34(5):859-875. PMCID:
PMC4359011.
d. *Guo B, Li Y, Yuan Y. A dose-schedule-finding design for phase I/II clinical trials. Journal of the
Royal Statistical Society, Series C 2016; 65(2):259-272. PMCID: PMC4747255.
4. Along with colleagues, I have developed missing data methods, in particular, nonparametric multiple
imputation methods for ignorable missing data. These methods are widely applicable in biomedical studies,
where missing data commonly arise.
a. Hsu C-H, Li Y, Long Q, Zhao Q, Lance P. Estimation of recurrence of colorectal adenomas with
dependent censoring using weighted logistic regression. PLoS One 6(10):e25141, 10/2011. PMCID:
PMC3204965.
b. Long Q, Hsu CH, Li Y. Doubly robust nonparametric multiple imputation for ignorable missing data.
Statistica Sinica 22(1):149-172, 2012. PMCID: PMC3280694.
c. Hsu C-H, Long Q, Li Y, Jacobs B. A nonparametric multiple imputation approach for data with
missing covariate values with application to colorectal adenoma data. Journal of Biopharmaceutical
Statistics 24(3):634-648, 4/2014. PMCID: PMC4353564.
d. Hsu C-H, He Y, Li Y, Long Q, Friese R. Doubly robust multiple imputation using kernel-based
techniques. Biometrical Journal 2016; 58(3):588-606. [PubMed - in process]
Complete List of Published Work in MyBibliography:
http://www.ncbi.nlm.nih.gov/sites/myncbi/yisheng.li.1/bibliography/45331928/public/?sort=date&direction=asce
nding.
D. Research Support
Ongoing Research Support
9/4/1988-6/30/2018
4 P30 CA016672 41 DePinho (PI)
NIH/NCI
Cancer Center Support Grant - Biostatistics Shared Resource (BGR)
The overarching goal is to improve the standard of patient care, as researchers and clinicians work to eliminate
cancer. The Biostatistics Resource Group provides biostatistical expertise and quantitative research resources
in support of all CCSG programs at MDACC. PID838041
Role: Statistician
127952 Milbury (PI)
7/1/2015-6/30/2017
American Cancer Society (ACS)
Couple-based meditation program for lung cancer patients and partners
To improve quality of life in cancer patients. PID3238
Role: Co-Investigator
1U01CA199218-01 Berry (PI)
9/1/2015-8/31/2020
Georgetown University Medical Center
Comparative Modeling: Informing Breast Cancer Control Practice and Policy
To develop model inputs related to treatment benefits and harms that have been developed in recent years.
PID3477
Role: Co-Investigator
Satcher (PI)
9/1/2015-8/31/2017
Cancer Survivorship Research Seed Money
Using telemonitoring to optimize the mobility of cancer survivors with skeletal metastases after surgery to
preserve limb function
To develop customized APPS and imaging technology for mobile devices for remote surveillance of late effects
and functional status of cancer survivors after surgery. To evaluate how well the current face to face follow up
visit can be adapted to the innovative remote surveillance format. To perform limited efficacy testing of remote
surveillance. (no salary support, 1% unpaid)
Role: Co-Investigator
1R21CA191711-01A1 Milbury (PI)
12/28/2015-11/30/2017
NIH/NCI
Couple-based meditation program for patients with metastatic lung cancer and their partners
This project will demonstrate the feasibility and initial effectiveness of a couple-based meditation intervention to
improve quality of life and symptom burden in patients with metastatic lung cancer.PID3883
Role: Co-Investigator
16634 Garcia (PI)
2/23/2016-2/22/2017
Gateway for Cancer Research
Randomized-controlled trial of acupuncture for post-mastectomy pain syndrome
To evaluate the use of acupuncture for reducing chronic pain, improving physical function, and improving QOL
between three groups of participants experiencing PMPS after breast cancer surgery: active
electroacupuncture (EA), sham electroacupuncture (SEA), and waitlist controls (WLC). PID4291
Role: Co-Investigator
Proposal # 00015311 Shen (PI)
8/1/2016-1/31/2018
NIH/NCI PREVENT Cancer Program
Examination of HJC0152, a putative modulator of glucose and energy metabolism, for mammary cancer
prevention
Examination of HJC0152, a putative modulator of glucose and energy metabolism, for mammary cancer
prevention To characterize the activity of HJC0152 as a cancer chemopreventive agent for potential application
in future human cancer prevention trials. This task order addresses this objective with respect to ER-negative
and TN mammary carcinogenesis. The tasks will involve assessment of cancer prevention efficacy in animals,
investigating the mechanistic basis of HJC0152 activity and its pharmacokinetic and pharmacodynamic profile.
Role: Collaborator
Completed Research Support
4/1/2010-3/31/2016
5 R01CA138800-04 Cohen (PI)
NIH/NCI
Yoga for women with breast cancer undergoing radiotherapy
To examine efficacy of incorporating a Yoga program alongside radiotherapy for women with breast cancer.
Role: Co-Investigator
5 U01 CA152958 05 Mandelblatt (PI)
9/1/2010-8/31/2015
NIH/NCI (Subcontract from Georgetown University) CISNET
Comparative Modeling: Informing Breast Cancer Control Practice & Policy
To develop model inputs related to treatment benefits and harms that have been developed in recent years.
Role: Co-Investigator
CP15990 Hortobagyi (PI)
6/1/2015-5/31/2016
Breast Cancer Foundation
Revising AJCC TNM Breast Cancer Staging: Analyzing National Datasets
To provide support for analysis of three national datasets that include robust biomarker data and complete
follow‐up to support incorporation of biomarker status into AJCC breast cancer staging. PID3504
Role: Co-Investigator
U01 CA152958 Mandelblatt (PI)
9/1/2015-8/31/2016
Subaward from Georgetown University (NCI)
Pilot CISNET - BOLD Task Force Supplement
To provide support for analysis of three national datasets that include robust biomarker data and complete
follow‐up to support incorporation of biomarker status into AJCC breast cancer staging. PID3361
Role: Co-Investigator
710177-80-107891 Cohen (PI)
MDACC - Servan Schrieber Cohen
Servan Schrieber Cohen
Increase funding for cancer research
Role: Statistician
9/1/2015-8/31/2016
172100-30-111870-46 Prokhorov (PI)
6/15/2013-08/31/2016
Centers for Medicare and Medicaid Services (CMS)
An evidence-based smoking cessation program for underserved persons living with HIV/AIDS
To reduce the morbidity and mortality due to tobacco-related cancers among persons living with HIV/AIDS.
FUND111870
Role: Co-Investigator
1 R01 HL127260-01 Fagundes (PI)
8/1/2015-8/31/2016
Subaward from Rice University /NHLBI
Project Heart: Biobehavioral effects on Cardiovascular Risk for Bereaved Spouses
To provide insight into the links between early adversity, social support, depressive symptoms and
inflammation in sample of breast cancer survivors. PID3522
Role: Co-Investigator
DFI Seed Funding Research Program McNeill (PI)
11/1/2014-10/31/2016
UTMDACC Duncan Family Institute Seed Funding Research Program
Food Deserts in Houston? Increasing Fruit and Vegetable Consumption to Reduce Cancer Risk
To determine the feasibility of recruiting African American families and implementing Brighter Bites program in
a Houston area food desert-located church. To pilot test and evaluate the efficacy of Brighter Bites in
increasing healthy diet and physical activity, and reducing obesity at the end of the study compared to wait-list
control group.
Role: Co-Investigator