Related Experiment Video
Updated: Jul 20, 2026
12:56
An Automated Method to Perform The In Vitro Micronucleus Assay using Multispectral Imaging Flow Cytometry
Published on: May 13, 2019
Sources of variability in the human lymphocyte micronucleus assay: a population-based study
K L Radack1, S M Pinney, G K Livingston
1Department of Medicine, University of Cincinnati College of Medicine, OH 45267-0535, USA.
Environmental and Molecular Mutagenesis
|January 1, 1995
Summary
The human lymphocyte micronucleus assay shows significant variability due to sampling error, not inter-individual differences. Researchers suggest focusing on reducing this error for more reliable genotoxicity testing results.
Area of Science:
- Genetics and Genomics
- Biomonitoring
- Toxicology
Background:
- The human lymphocyte micronucleus assay is a key tool for assessing genotoxicity.
- Understanding sources of variability is crucial for accurate interpretation of assay results.
- Previous studies have not fully elucidated the contribution of sampling error versus inter-individual variation.
Purpose of the Study:
- To quantify the intraclass correlation coefficient (ICC) in the micronucleus assay.
- To identify sources of variability, including intra-individual and inter-individual factors.
- To evaluate practical aspects of sample size and its impact on statistical sensitivity.
Main Methods:
- Cytokinesis-block method applied to human lymphocyte cultures.
- Analysis of 20 samples (100 binucleate cells each) per subject (n=112).
- Variance component analysis to determine ICC and sources of variation.
Main Results:
- Intra-individual (sampling) variation exceeded inter-individual variation.
- The ICC was 41.6%, indicating substantial error variance (58.4%).
- A significant gender effect was observed, with higher micronuclei frequency in females.
Conclusions:
- Sampling error is the primary source of variability in this micronucleus assay dataset.
- A strong gender-specific effect favoring females was confirmed.
- Scoring >1,000 cells per subject did not significantly improve statistical sensitivity over 20 determinations, suggesting current sample sizes are adequate if sampling error is minimized.
