Related Experiment Video
Updated: May 3, 2026
Optimized Analysis of DNA Methylation and Gene Expression from Small, Anatomically-defined Areas of the Brain
Published on: July 12, 2012
Analysis of gene expression data using a linear mixed model/finite mixture model approach: application to regional
Daniah Trabzuni1, , Peter C Thomson2
1Department of Molecular Neuroscience, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK, Department of Genetics, King Faisal Specialist Hospital and Research Centre, Riyadh 11211, Saudi Arabia and ReproGen - Animal Bioscience Group, Faculty of Veterinary Science, The University of Sydney, 425 Werombi Road, Camden, NSW 2570, AustraliaDepartment of Molecular Neuroscience, UCL Institute of Neurology, Queen Square, London WC1N 3BG, UK, Department of Genetics, King Faisal Specialist Hospital and Research Centre, Riyadh 11211, Saudi Arabia and ReproGen - Animal Bioscience Group, Faculty of Veterinary Science, The University of Sydney, 425 Werombi Road, Camden, NSW 2570, Australia.
This study introduces a whole-genome analysis method for gene expression data using linear mixed models and finite mixture models. The approach simplifies differential gene expression analysis for researchers, revealing co-regional expression patterns in human brain tissue.
Area of Science:
- Genomics
- Bioinformatics
- Statistical Genetics
Background:
- Gene expression data contains genome-wide information.
- Existing analysis methods require significant bioinformatics and statistical expertise.
- A novel, accessible whole-genome approach is needed.
Purpose of the Study:
- To develop and demonstrate an efficient whole-genome analysis method for gene expression data.
- To enable users with limited bioinformatics expertise to analyze gene expression data.
- To identify differentially expressed (DE) transcripts for further functional analysis.
Main Methods:
- Application of large-scale linear mixed models to genome-wide expression data.
- Utilizing finite mixture models to distinguish DE from non-DE transcripts.
- Illustration using UK Brain Expression Consortium data from 12 human brain regions.
Main Results:
- Linear mixed models effectively investigated gene expression variation across biological states (brain regions, gender, age).
- The method identified striking patterns of co-regional gene expression.
- Finite mixture models provided a straightforward method for filtering and extracting DE transcripts.
Conclusions:
- The developed method offers an efficient whole-genome perspective for gene expression analysis.
- This approach is accessible to researchers with limited statistical and bioinformatics expertise.
- The identified DE transcripts are suitable for advanced functional analysis.
