Tools & Resources

Methodologic Studies

Sample Size Calculations for Studies of Gene-Environment Interactions

We are carrying out methodological work to evaluate the sample size needed to study gene-environment interactions, as well as the impact of exposure and genotype misclassification on bias and sample size. We have shown that even small errors in measuring environmental or genetic factors could have a very strong impact on sample size requirements. Therefore, high quality exposure and genotype assessments are crucial, not only to obtaining unbiased estimates of interaction effects, but also to being able to assess interactions with a manageable study size. In collaboration with Dr. Jay Lubin from the Biostatistics Branch we have developed a statistical software to perform power and sample size calculations for studies of gene-environment interactions, as well as to assess the impact of misclassification in a particular study. For More Information Contact: Montserrat García-Closas

Buccal Cell Sample Collection and Processing to Evaluate Genetic Polymorphisms

We are conducting a study to compare DNA yields and quality obtained from different buccal cell collection protocols (cytobrushes and mouthwash) and different DNA extraction techniques. In this study we are also exploring the use of whole genome amplification techniques to increase the number of assays we can perform from each DNA sample. For More Information Contact: Montserrat García-Closas

Reproducibility and Validity of Hormone Assays

The branch has been involved in an effort to develop approaches to ensure high quality of biochemical probes in large-scale epidemiologic studies, particularly in the area of hormonal carcinogenesis. An extensive methodologic study designed to assess laboratory reproducibility and validity of a large number of hormone assays has demonstrated that current laboratory assays for most circulating estrogens and some but not all androgens are adequate for large research efforts provided quality control procedures are carefully implemented and monitored. Current assays for urinary estrogens and the 2- and 16 -hydroxyestrone metabolites are also reliable, even in postmenopausal women, where serum estrogen concentrations may be at or below assay sensitivity. For More Information Contact: Roni Falk