Skip Navigation
  Print Page

Biostatistics & Bioinformatics Branch

Skip sharing on social media links
Share this:
Paul S. Albert, Ph.D.

Paul S. Albert, Ph.D.

Longitudinal and Correlated Data Analysis
Analyzing Time-to-Event Data
Analysis of Biomarker Data
Analysis of Genetic Data

Branch Chief:  Paul S. Albert, Ph.D.

The mission of the Biostatistics and Bioinformatics Branch (BBB) is to: 1) conduct both collaborative and methodological research that is important to the mission of the Division and Institute, 2) provide statistical training in areas of statistical research that will advance the Division's and Institute's research programs, and 3) serve as a resource for the Division, the Institute, the NIH, and other professional and government organizations.  The research component of the BBB's mission is multifaceted.  First, providing first-rate statistical collaboration requires understanding of the scientific issues and state-of-the-art statistical methodology relevant to the scientific problem.  Therefore, investigators within the Branch play a role in all aspects of the study.  Second, the Branch develops new statistical methodology for designing and for analyzing data.  Analytical issues encountered in collaborative research directly motivate much of the Branch's independent research.

A majority of the Division's studies are longitudinal and involve sampling frameworks such as schools, families (parent-child triads), couples, maternal/fetal pairs, and individuals.  Particular methodological problems that have been addressed include: 1) the joint modeling of longitudinal data and time to event or understanding the association of longitudinal profiles and an outcome of interest; 2) the characterization of longitudinal menstrual cycle and circadian rhythm patterns; and 3) the development of new approaches for designing and analyzing correlated data subject to informative cluster size, where the number of measurements is related to the underlying process of interest.

An important analytical issue for many Division studies is the characterization of the time to an event.  In many studies, correlated event times are measured (e.g., repeated time-to-pregnancy and gestation at birth in consecutive pregnancies) and interest is on identifying environmental, genetic, or behavioral factors that influence these durations.  A major research focus during 2012 has been on statistical modeling of time-to-pregnancy, which poses new analytic challenges since, unlike with traditional survival analysis, it must account for the fact that there is no risk of pregnancy without intercourse during a particular window in time.

BBB investigators have developed new statistical methods for analyzing biomarker data.  For example, in 2012 we have developed optimal pooling strategies for reducing the expense of assay measurements in longitudinal studies and supervised latent-class models for examining the effects of a large number of biomarkers on the incidence of disease.  BBB investigators have developed new methodology for assessing agreement from longitudinally collected ratings and scores.  In addition to assessing agreement, BBB investigators have new approaches for assessing the accuracy of ratings or tests when no gold standard test is available.

During 2012, BBB investigators have developed new statistical methodology for analyzing quantitative traits when the outcomes are longitudinal and have developed entropy-based methods for detecting gene-gene and gene-environmental interactions of complex diseases.

Staff

  • Paul S. Albert, Ph.D., Senior Investigator and Chief
  • Aiyi Liu, Ph.D., Senior Investigator
  • Zhen Chen, Ph.D., Investigator
  • Ruzong Fan, Ph.D., Investigator
  • Danping Liu, Ph.D., Investigator
  • Sundaram Rajeshwari, Ph.D., Investigator
  • SungDuk Kim, Ph.D., Staff Scientist
  • Yunlong Xie, Ph.D., Postdoctoral Fellow
  • Kirsten J. Lum, M.S., Predoctoral Fellow
  • Kara Fulton, B.S., Postbaccalaureate Fellow

Awards and Accomplishments

  • Aiyi Liu, Ph.D., Elected Fellow of the American Statistical Association, 2012

Longitudinal and Correlated Data Analysis

SungDuk Kim, Ph.D.

SungDuk Kim, Ph.D.

Danping Liu, Ph.D.

Danping Liu, Ph.D.

A majority of the Division's studies are longitudinal and involve sampling frameworks such as schools, families (parent-child triads), couples, maternal/fetal pairs, and individuals. Longitudinal studies have inherent methodological challenges over time, including the problem of attrition, difficulties in making statistical inference when data are correlated, and difficulties in characterizing complex longitudinal patterns.  Many of the Branch's independent research projects address one or more of these issues in the context of substantive problems related to one or more of the Division's studies.  Particular methodological problems that have been addressed include: 1) the joint modeling of longitudinal data and time-to-event for understanding the association of longitudinal profiles and an outcome of interest. Branch Investigators have proposed approaches for inference and prediction with applications to the Longitudinal Investigation of Fertility and the Environment (LIFE) Study as well as to the NICHD Fetal Growth Study; 2) characterizing longitudinal menstrual cycle and circadian rhythm patterns in longitudinal data with applications to the BioCycle Study and the NEXT Study, and 3) the development of new approaches for designing and analyzing correlated data subject to informative cluster size, where the number of measurements is related to the underlying process of interest.

2012 Publications

  1. Albert PS. A linear mixed model for predicting a binary event under random effects misspecification. Statistics in Medicine 31(2):143-54, 2012.
  2. Bhadra D, Daniels M J, Kim SD, Ghosh M, Mukherjee B. A Bayesian semiparametric approach for incorporating longitudinal information on exposure history for inference in case-control studies. Biometrics68:361-370, 2012.
  3. Cheon K, Albert PS, Zhang Z. The impact of random effect misspecification on percentile estimation for longitudinal growth data. Statistics in Medicine 31:3708-3718, 2012.
  4. Malinovsky Y, Albert PS, Schisterman EF. Pooling designs for outcomes under a Gaussian random effects distribution. Biometrics 68:45-52, 2012.
  5. Ogbagaber S, Albert PS, Lewin D, Iannotti R. Summer activity patterns among teenage girls: Harmonic shape invariant modeling to estimate circadian cycles. Journal of Circadian Rhythms10(1):2, 2012.
  6. Roy A, Danaher M, Mumford S, Chen Z. A Bayesian order restricted model for hormonal dynamics during biocycles in healthy women. Statistics in Medicine 31:2428-2840, 2012.
  7. Wu MX, Yu KF, Liu A, Ma TF. Simultaneous optimal estimation in linear mixed effects models. Metrika 75:471–489, 2012.
  8. Zhang Z, Albert PS, Simons-Morton B. Marginal analysis of longitudinal count data in long sequences: methods and applications to a driving study. Annals of Applied Statistics 6:27-54, 2012.
  9. Albert PS, Shih JH. Modeling batched Gaussian longitudinal data subject to informative dropout. Statistical Methods in Medical Research (In press).
  10. Jackson J, Albert PS, Zhang Z, Simons-Morton B. Ordinal latent variable models and their application in the study of newly licensed teenage drivers.  Journal of the Royal Statistical Society: Series C (In press).
  11. McLain AC, Sundaram R, Louis GMB. Modeling time to pregnancy in presence of sterile fraction using transformation survival model. Statistical Methods in Medical Research(In press).
  12. Sun W, McLain AC. Multiple testing of composite null hypotheses in heteroscedastic models. Journal of the American Statistical Association, Theory and Methods (In press).
  13. Zhang B, Chen Z, Albert PS. Estimating diagnostic accuracy of raters without a gold standard by exploiting a group of experts. Biometrics (In press).
  14. Zhang Z, Chen Z, Troendle J, Zhang J. Causal inference on quantiles withapplication to safe labor. Biometrics (In press).

Analyzing Time-to-Event Data

Rajeshwari Sundaram, Ph.D.

Rajeshwari
Sundaram, Ph.D.

An important analytical issue for many Division studies is the characterization of time to an event.  In many studies correlated event-times are measured (e.g., repeated time-to pregnancy, gestation at birth in consecutive pregnancies, gap times between accidents in teenage driving) and interest is on identifying environmental or behavioral factors that influence these durations.

There are many new analytic challenges for appropriate analysis of such data. For example, time to pregnancy and other outcomes related to maternal and child health poses new analytic challenges since, unlike with traditional survival analysis, time-to-pregnancy analysis must account for the fact that there is no risk of pregnancy without intercourse during a particular window in time.  Statistical modeling of human fecundity has been an important area of Branch research in this area.  Other areas include developing new approaches for modeling consecutive pregnancy outcomes subject to competing risks (e.g., incidence of pre-term birth due to preeclampsia) and modeling the gap times between pregnancies.

2012 Publications

  1. McLain AC, Sundaram R, Louis GMB. Modeling time to pregnancy in presence of sterile fraction using transformation survival model. Statistical Methods in Medical Research(In press).
  2. Sundaram R, McLain A, Louis Buck G. A survival analysis approach to modeling human fecundity. Biostatistics 13:4-17, 2012.

Analysis of Biomarker Data

Zhen Chen, Ph.D.

Zhen Chen, Ph.D.

Aiyi Liu, Ph.D.

Aiyi Liu, Ph.D.

Most of the studies within the Division collect biomarkers as either measures of exposure or outcome, with these biomarker measurements often being measured repeatedly. Often, these biomarkers are subject to large biological and technical errors as well as detection limits.  BBB investigators have developed optimal design strategies for reducing measurement error when multiple assays are subject to detection limits, and on optimal pooling strategies for reducing the expense of assay measurements in large studies. BBB investigators have also developed supervised latent-class models for examining the effects of a large number of biomarkers on the incidence of disease, an area of research which will have increased importance as the number of assays that can be examined with a single biospecimen will increase substantially.

We have an active research program in assessing inter-rater agreement and diagnostic accuracy.  BBB investigators have developed new methodology for assessing agreement from longitudinally collected ratings and scores.  In addition to assessing agreement, researchers are often interested in assessing the accuracy of ratings or tests when there is no gold standard test available.  Many of the methods developed for assessing agreement and diagnostic accuracy were developed from collaborative research in the Endometriosis: National History, Diagnosis, and Outcome (ENDO) Study, which is focused on comparing and evaluating different measures for diagnosing endometriosis in the absence of a gold standard.

2012 Publications

  1. Albert PS, Schisterman EF. Novel Statistical methodology for analyzing longitudinal biomarker data. Statistics in Medicine 31:2457-2460, 2012.
  2. Buck Louis GM, Sundaram R. Exposome: time for transformative research: invited commentary. Statistics in Medicine 31:2569-2575, 2012.
  3. Liu A, Liu C, Zhang Z, Albert PS. Optimality of group testing in the presence of misclassification. Biometrika 99: 245-251, 2012.
  4. Malinovsky Y, Albert PS, Schisterman EF. Pooling designs for outcomes under a Gaussian random effects distribution. Biometrics 68:45-52, 2012.
  5. Roy A, Danaher M, Mumford S, Chen Z. A Bayesian order restricted model for hormonal dynamics during biocycles in healthy women. Statistics in Medicine 31:2428-2840, 2012.
  6. Schisterman EF, Albert PS. The biomarker revolution. Statistics in Medicine 31:2565-2568, 2012.
  7. Tang LL, Liu A, Schisterman EF, Zhou X, Liu CL.  Homogeneity tests of clustered diagnostic markers with applications to the BioCycle Study. Statistics in Medicine 31:3638-3648, 2012.
  8. Zhang B, Chen Z, Albert PS. A supervised latent class model for high dimensional biomarker data. Biostatistics 13:74-88, 2012.
  9. Zhang Z, Albert PS, Simons-Morton B.  Marginal analysis of longitudinal count data in long sequences: methods and applications to a driving study. Annals of Applied Statistics 6:27-54, 2012.
  10. Zhang Z, Liu A, Lyles RH, Mukherjee B. Logistic regression analysis of biomarker data subject to pooling and dichotomization. Statistics in Medicine 31:2485-2497, 2012.
  11. Jin M, Liu, A, Chen Z, Li ZH. Group sequential design in inter-rater reliability study. Statitica Sinica (In press).
  12. Sun W, McClain AC. Multiple testing of composite null hypotheses in herteroscedastic models. Journal of the American Statistical Association, Theory and Methods (In press).
  13. Tang LL, Liu A, Chen Z, Schisterman, EF, Zhang B, Miao Z. Nonparametric ROC summary statistics for diagnostic marker data. Statistics in Medicine (In press).
  14. Zhang B, Chen Z, Albert PS. Estimating diagnostic accuracy of raters without a gold standard by exploiting a group of experts. Biometrics (In press).
  15. Zhang Z, Chen Z, Troendle J, Zhang J. Causal inference on quantiles withapplication to safe labor. Biometrics (In press).

Analysis of Genetic Data

Ruzong Fan, Ph.D.

Ruzong Fan, Ph.D.

The analysis of genetics data is an active area of biostatistics research and presents unique opportunities and statistical challenges, especially when dealing with data related to birth defects. For example, Division studies often have genetic information on a particular child as well as on both parents (triads), resulting in difficult analytic and design issues that are being addressed by BBB investigators.  BBB investigators are also developing new methodology for analyzing quantities and qualitative traits when the outcomes are longitudinal and developing entropy-based methods for detecting gene-gene and gene-environmental interactions of complex diseases.  Of particular interest in the Division is accounting for measurement error in environmental exposures.  BBB investigators have developed methodology for estimating gene-environment interactions in the presence of measurement error in the environmental factors.

2012 Publications

  1. Danaher MR, Schisterman EF, Roy A, Albert PS.  Estimation of gene-environment interaction by pooling biospecimens.  Statistics in Medicine31:3241-3252, 2012.
  2. Fan R, Albert PS, Schisterman.  A discussion of gene-gene environment interaction and longitudinal genetic analysis of complex traits.  Statistics in Medicine31:2565-2568, 2012.
  3. Fan R, Zhang Y, Albert PS, Liu A, Wang Y, Xiong, M. Longitudinal genetic analysis of quantitative traits. Genetic Epidemiology 36:856-869, 2012.

Collaborative Research

BBB investigators are essential members of the research team on all major projects in the Epidemiology Branch (EB) and Prevention Research Branch (PRB), with a primary and a secondary statistical investigator being on most projects.  The investigators are involved in all aspects of the study from its earliest concept, including study design, implementation and data quality, and analysis.  We are also involved in collaborations with Division of Intramural Research (DIR) investigators as well as with extramural staff in and outside NICHD.  Further, we serve on important NIH and external committees such as the NICHD IRB, the NIH Biometry and Epidemiology Tenure Committee, and numerous NIH DSMBs.

2012 Publications

  1. Bobe G, Murphy G, Albert PS, Sansbury LB, Lanza E, Schatzkin A, Cross AJ. Dietary Lignan and proanthocyanidin consumption and colorectal adenoma recurrence in the Polyp Prevention Trial.International Journal of Cancer 130:1649-1659, 2012.
  2. Brown DA, Hance KW, Rogers CJ, Sansbury LB, Albert PS, Murphy G, Laiyemo A, Wang Z, Cross AJ, Schatzkin A, Danta M, Srasuebkul P, Amin J, Law M, Breit SN, Lanza E.  Serum macrophage inhibitory cytokine-1 (MIC-1jGDF15) for screening and prevention of colon cancer.  Cancer Epidemiology Prevention and Biomarkers 21:337-348, 2012.
  3. Buck Louis GM, Sundaram R, Schisterman EF, Sweeney AM, Lynch CD, Gore-Langton RE, Chen Z, Kim S, Caldwell K, Barr DB. Heavy Metals and Couple Fecundity, the LIFE Study. Chemosphere 87:1201-1207, 2012.
  4. Carter TC, Kay DM, Browne ML, Liu A, Romitti PA, Kuehn D, Conley MR, Caggana M, Druschel CM, Brody LC, Mills JM.  Hirschsprung's disease and variants in genes that regulate enteric neural crest cell proliferation, migration, and differentiation.  Journal of Human Genetics 57(8):485-93, 2012.
  5. Hartman TJ, Mahabir S, Stevens RG, Albert PS, Dorgan JF, Kesner J, Meadows JS, Shields R, Taylor PR. Moderate alcohol consumption and 24-hour urinary levels of melatonin in postmenopausal women. Journal of Clinical Endocrinology and Metabolism 97:E65-68, 2012.
  6. Kunisue T, Chen Z, Louis GMB, Sundaram R, Hediger ML, Sun L, Kannan K. Urinary concentrations of benzophenone-type UV filters in the US women from two cities and their association with endometriosis. Environmental Science and Technology46:4624-4632, 2012.
  7. Laughon SK, Zhang J, Grewal J, Sundaram R, Beaver J, Reddy U. Induction of labor in contemporary obstetric practice. American Journal of Obstetrics & Gynecology 206:486.e1-9, 2012.
  8. Buck Louis GM, Chen Z, Peterson CM, Hediger ML, Croughan MS, Sundaram R, Stanford JB, Varner MW, Fujimoto VY, Giudice LY, Trumble A, Parsons PJ, Kannan K. Persistent lipophilic environmental chemicals and endometriosis: The ENDO Study, 2007-2009. Environmental Health Perspectives 120:811-816, 2012.
  9. Lynch CD, Sundaram R, Buck Louis GM, Lum KJ, Pyper C. Are increased levels of self-reported psychosocial stress, anxiety, and depression associated with fecundity? Fertility and Sterility 98:453-458, 2012.
  10. Mechanic LE, Chen HS, Amos CI, Chatterjee N, Cox NJ, Divi RL, Fan R, Harris EL, Jacobs K, Kraft P, Leal SM, McAllister K, Moore JH, Paltoo DN, Province MA, Ramos EM, Ritchie MD, Roeder K, Schaid DJ, Stephens M, Thomas DC, Weinberg CR, Witte JS, Zhang S, Zöllner S, Feuer EJ, Gillanders EM.  Next generation analytic tools for large scale genetic epidemiology studies of complex diseases. Genetic Epidemiology 36:22-35, 2012.
  11. Mills JL, Carter TC, Kay DM, Browne ML, Brody LC, Liu A, Romitti PA, Caggana M, Druschel CM. Folate and vitamin B12 related genes and risk for omphalocele. Human Genetics131:739–746, 2012.
  12. Nansel TR, Iannotti RJ, Liu A. Clinic-integrated behavioral intervention for families of youth with type 1 diabetes: A randomized clinical trial. Pediatrics129: 866-873, 2012.
  13. Simons-Morton BG, Cheon K, Guo F, Albert P. Trajectories of kinematic risky driving among novice teenagers. Accident Analysis and Prevention 51:27-32, 2012.
  14. Thoma M, Hediger M, Sundaram R, Stanford J, Peterson CM, Croughan M, Chen Z, Buck Louis GM. Comparing "Apples and Pears": How well do women perceive their body size and shape? Journal of Women's Health 21(10):1074-1081, 2012.
  15. Winer KK, Zhang B, Shrader J, Peterson D, Sinaii N, Smith M, Albert PS, Cutler G. Synthetic human parathyroid hormone 1-34 replacement therapy: a randomized crossover trial comparing pump versus injections in the treatment of chronic hypoparathyroidism. The Journal of Clinical Endocrinology & Metabolism97:391-399, 2012.
  16. Zhang J, Kim SD, Grewal U, Albert PS. Predicting large fetuses at birth: Do multiple ultrasound examinations and longitudinal statistical modeling improve prediction? Paediatric and Perinatal Epidemiology 26:199-207, 2012.
  17. Buck Louis GM, Sundaram R, Schisterman EF, Sweeney AM, Lynch CD, Gore-Langton RE, Maisog J, Kim S, Chen Z, Barr DB. Persistent environmental pollutants and couple fecundity, The LIFE Study. Environmental Health Perspectives (In press).
  18. Carter TC, Kay DM, Browne ML, Liu A, Romitti PA, Kuehn D, Conley MR, Caggana M, Druschel CM, Brody LC, Mills JM. Anorectal atresia and variants at predicted regulatory sites in candidate genes. Journal of Human Genetics (In press).
  19. Erickson HS, Canales JR, Albert PS, Yala K, Mukherjee S, Hu N, Goldstein AM, Chuaqui RF, Hewitt SA, Taylor PR, Emmert O, Buck MR. Interrogation of chromosome 13q12-14 esophageal squamous cell carcinoma. The Open Pathology Journal (In press).
  20. Buck Louis GM, Chen Z, Peterson CM, Hediger ML, Croughan MS, Sundaram R, Stanford JB, Varner MW, Fujimoto VY, Giudice LY, Trumble A, Parsons P J, Kannan K. Perfluorochemicals and endometriosis: The ENDO Study. Epidemiology 23:799-805, 2012.
  21. Mumford, SL, Steiner, AZ, Pollack, AZ, Perkins, NJ, Filiberto, AC, Albert, PS, Mattison, DR, Wactawski-Wende J, Schisterman EF. The hormonal profile and its effect on menstrual cycle length. The Journal of Clinical Endocrinology & Metabolism (In press).
  22. Mitchell  JB, Anvers MR, Sowers AL, Rosenberg PS, Figueroa M, Thetford A, Albert PS, Cook JA. The antioxidant tempol reduces carcinogenesis and enhances survival in mice when administered after non-lethal total body radiation. Cancer Research (In press).
  23. Schliep KC, Stanford JB, Zhang B, Chen Z, Dorals JK, Johnstone EB, Hammoud AO, Varner MW, Louis BG. Inter- and intra-reliability in the diagnosis and staging of endometroiosis:  the ENDO Study. Obstetrics and Gynecology120:104-112, 2012.
  24. Thoma ME, McLain AC, Louis JF, King RB, Trumble AC, Sundaram R, Buck Louis GM. The prevalence of infertility in the United States as estimated by the current duration approach and a traditional constructed approach. Fertility and Sterility 2013 Jan 3. doi: 10.1016/j.fertnstert.2012.11.037.
  25. Wolff EF, Hediger ML, Sundaram R, Peterson CM, Chen Z, Buck Louis GM. In utero exposures and endometriosis, the ENDO Study. Fertility and Sterility 2012 Dec 1. doi: 10.1016/j.fertnstert.2012.11.013.
  26. Yeung EH, Zhang C, Albert PS, Ye A, Mumford SL, Perkins NJ, Hediger ML, Wactawski-Wende J, Schisterman EF. The influence of adiposity on menstrual cycle patterns of sex hormones: the BioCycle Study. The International Journal of Obesity (In press).
  27. Waldman TA, Colon KC, Steward DW, Worthy TA, Janik JE, Fleischer TA, Albert, PS, Figgg WD, Spencer SD, Decker JR, Goldman CK, Bryant BR, Petrus MN, Creekmore SP, Morris JC.  Phase I clinical trial of blockade of IL-15 transpresentation using humanized Mik-Beta-1 monoclonal antibody directed toward IL-2/IL-15R beta in patients with T-cell large granular lymphocytic leukemia.  Blood (In press).
Last Updated Date: 02/05/2013
Last Reviewed Date: 02/05/2013

Contact Information

Name: Dr Paul Albert
Chief and Senior Investigator
Biostatistics and Bioinformatics Branch
Phone: 301-496-5582
E-mail: albertp@mail.nih.gov

Staff Directory
Vision National Institutes of Health Home BOND National Institues of Health Home Home Storz Lab: Section on Environmental Gene Regulation Home Machner Lab: Unit on Microbial Pathogenesis Home Division of Intramural Population Health Research Home Bonifacino Lab: Section on Intracellular Protein Trafficking Home Lilly Lab: Section on Gamete Development Home Lippincott-Schwartz Lab: Section on Organelle Biology