A full Bayesian hierarchical mixture model for the variance of gene differential expression
2007

Bayesian Model for Gene Expression Variance

Sample size: 9216 publication Evidence: moderate

Author Information

Author(s): Manda Samuel, Rebecca E Walls, Mark S Gilthorpe

Primary Institution: Biostatistics Unit, Centre for Epidemiology and Biostatistics, Leeds, UK

Hypothesis

Can a Bayesian mixture model provide more accurate estimates of gene expression variance under limited replicates?

Conclusion

The Bayesian mixture variance model offers a more realistic estimate for gene expression variance, leading to more robust p-values.

Supporting Evidence

  • The model was tested on a dataset of 9216 genes with four replicates per condition.
  • The mixture variance model showed improved robustness in p-value estimation compared to traditional methods.
  • The study highlights the importance of accurate variance estimation in identifying differentially expressed genes.

Takeaway

This study created a new way to look at gene data that helps scientists understand which genes are different from each other, even when they don't have a lot of samples to compare.

Methodology

A Bayesian hierarchical mixture model was used to classify genes based on variance similarity, utilizing data from multiple replicates.

Potential Biases

Potential biases may arise from the assumptions made in the model regarding gene variance.

Limitations

The model's performance may vary based on the assumptions of the prior distributions for mixture weights and scales.

Digital Object Identifier (DOI)

10.1186/1471-2105-8-124

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication