Preceptorships – Potential Projects

Below you will find a list of preceptorship projects that faculty have offered as potential project ideas. Note that this list is not exhaustive and you may approach faculty members for other project ideas.

Preceptorship Projects Available for 2023-24

(revised September 5, 2023)

  1. Assessing longitudinal surrogate markers (Emily Roberts)
    In randomized clinical trials, an intermediate marker may serve as surrogate for a clinical outcome with the goal of making the trial run more efficiently. This proposed project will use causal inference approaches to determine if a surrogate outcome is valid to use as the primary endpoint in a future trial. Specifically, we will assess how well an extension of methods using linear or generalized linear mixed models performs when the marker is measured longitudinally (as it is in a diabetes clinical trial motivating this work). This project could also involve comparison of existing methods via simulation or other surrogate marker questions.
  2. Electroencephalography (EEG) Data Analysis (Ryan Cho)
    Electroencephalography (EEG) is a noninvasive and widely available brain imaging modality which measures spontaneous electrical activity across brain regions. EEG data created in event-related potential (ERP) experiments possess a complex high-dimensional structure: (1) each trial generates an ERP waveform which is an instance of functional data, (2) the experiments are made up of sequences of multiple trials, resulting in longitudinal functional data, and (3) responses are recorded at multiple electrodes. EEG studies produce high-dimensional data with regional, functional, and longitudinal dimensions, which would provide an unique opportunity that analyzes EEG data in a variety of ways.
  3. Measures of diversity and equity   (Jeff Dawson)
    Dr. Dawson is interested in supervising a preceptorship focusing on statistical methods in justice, equity, diversity, and inclusion (JEDI).  This project would investigate existing and novel metrics of diversity, representativeness, and/or income disparities, through simulations and analysis of publicly-available data.  The project will extend recent work to be presented by JEDI experts at the Joint Statistical Meetings in August of 2023, where Dr. Dawson was an official discussant.
  4. Stillbirth prevention data (Jeff Dawson)
    Healthy Birthday Inc., a non-profit in Des Moines, Iowa, that focuses on stillbirth prevention in expectant families, is seeking statistical help to assess CDC Wonder data for stillbirth measure to examine trends and with health equity in mind.  Both CDC Wonder data and data from other state partners would be used for the purpose of this preceptorship.
  5. Two potential projects with Kai Wang
    Please contact Dr. Wang to discuss the data use.
    • 5.1  Causal discovery using omics data
    • 5.2  Mendelian randomization analysis of exposures and outcomes of interest
  6. Simulation Study for the Sample Size Reassessment in Ongoing Cluster Randomized Clinical Trials (Emine Bayman)
    Sample size calculations for the cluster randomized clinical trials should make assumptions about intraclass correlation coefficient (ICC), which may not be available at the time of study design. We would like to conduct a simulation study for the sample size reassessment in ongoing cluster randomized clinical trials.
  7. Interactive web portal for exploring miRNA binding sites (Patrick Breheny)
    Modern deep sequencing methods have enabled researchers at the University of Iowa to assemble a genome-wide map of binding interactions between proteins and micro RNAs (miRNA). This is a potentially valuable resource, but the challenge is how to share it with the research community. In particular, outside researchers need a way to navigate through the size and complexity of the data to find the information that is relevant to them.  Interactive web portals (Shiny apps) provide an attractive solution to this problem, and would allow researchers around the world to explore this resource, visualize it, find interesting patterns, and download relevant portions.
  8. Improving power for genome-wide study of heart failure in patients with arrhythmia (Patrick Breheny)
    Researchers at the University of Iowa led a multi-center study called Genetic Risk Assessment of Defibrillator Events (GRADE) in the hopes of identifying genetic risk factors that would indicate which arrhythmia patients are at elevated risk of heart failure. One challenge of genome-wide association studies (GWAS) such as this is that there is a heavy correction for multiple testing, and indeed, initial analyses failed to identify risk factors that could survive the multiple testing correction. A potential remedy would be to reanalyze this data incorporating information from other cardiac GWAS in order to focus power on a smaller set of risk factors that have prior support for affecting cardiac function.
  9. R package for tumor growth experiments (Patrick Breheny)
    A common experiment in cancer research is to implant tumors in two groups of mice, then give an experimental drug to one group in order to see if it slows the growth of the tumor. The experiment is relatively simple, and yet not trivial to analyze, as it involves repeated longitudinal measurements, nonlinear trends (tumor growth tends to be exponential), and censoring (if the tumor is too small, it cannot be detected). An R package (which I personally would call “tumr”) to facilitate these analyses would be very helpful — to cancer researchers here at the University of Iowa Cancer Center but also other centers.
  10. Enteric pathogens are a major source of morbidity and mortality in infants and young children living in low to middle income countries.  The pathways through which disease is transmitted are myriad and varied. Potential projects with Daniel Sewell
  • PROJECT 1: We wish to apply machine learning approaches to predict 2-week prevalence of a variety of pathogen infections in infants using structured observational data on infant and caregiver behaviors as well as environmental factors.   
  • PROJECT 2: (Predicated on someone completing PROJECT 1) We will build off of PROJECT 1 by determining the predictive capacity of each variable using Bayesian bootstrapping.
  • PROJECT 3: We have both behavioral observations and microbiological samples on hands, objects, food, soil, water, etc.  We can combine these data along with prior studies on, e.g., hand-to-mouth transfer efficiency, to determine the distribution of pathogen dose through multiple pathways, thereby allowing direct comparison of their importance.  This will be critical in designing effective interventions aimed at reducing enteric disease.
  • PROJECT 4: Often pathogens will appear in infant stools, but it may not indicate an infection.  We wish to determine the non-linear relationship between pathogen concentration and symptomology.