jump to content  sitemap
The Ohio State University
Center for Biostatistics and Bioinformatics

Home > Methods Research

Methods Research

Some current methods research by our biostatisticians:

  • Adaptive clinical trial design
  • Sample size re-estimation for laboratory experiments
  • Causal inference using propensity score methods
  • Moderated t-statistics for small sample size gene expression arrays
  • Mice-human correlations based on gene expression data
  • Missing data sensitivity analysis
  • Bias with propensity scores used as a regression covariate
  • Sample size re-estimation for clinical trials
  • Robust mixed model hypothesis testing
  • Microarray/microRNA normalization and filtering methods



Moderated t-statistics for small sample size gene expression arrays

Gene expression microarray experiments with few replications lead to great variability in estimates of gene variances. We extend the existing methods by allowing the CV to vary with gene expression, which we refer to as the fully moderated t-statistic, and compared to three other methods (ordinary t, and two moderated t predecessors). Our CV varying method had higher power in a spike-in dataset, and in a real dataset better identified higher expressing genes that were consistent with functional pathways associated with the experiments.

Yu, Lianbo; Gulati, Parul; Fernandez, Soledad; Pennell, Michael; Kirschner, Lawrence; and Jarjoura, David (2011) "Fully Moderated T-statistic for Small Sample Size Gene Expression Arrays," Statistical Applications in Genetics and Molecular Biology: Vol. 10: Iss. 1, Article 42.

Download software