Mixture Modeling of Microarray Gene Expression Data
Author Information
Author(s): Yang Yang, Tashman Adam P, Lee Jung Yeon, Yoon Seungtai, Mao Wenyang, Ahn Kwangmi, Kim Wonkuk, Mendell Nancy R, Gordon Derek, Finch Stephen J
Primary Institution: Stony Brook University
Hypothesis
Do certain genes exhibit a mixture distribution in their expression patterns?
Conclusion
About 28% of genes analyzed show evidence of following a mixture distribution, indicating potential biological mechanisms at play.
Supporting Evidence
- 28% of genes analyzed showed a mixture distribution.
- 18 trios and 35 quartets of genes were identified with high concordance rates.
- Factor analysis revealed three groups of genes with varying concordance rates.
Takeaway
Scientists looked at how some genes behave in groups and found that many of them act similarly, which might mean they are connected in some way.
Methodology
The study used mixture analysis, partial correlation coefficients, and factor analysis to identify gene groups and their expression patterns.
Potential Biases
The study's reliance on familial data may introduce bias due to non-random sampling.
Limitations
The data were collected from 14 pedigrees, which may not represent a random sample, and the effect of familial dependence is unclear.
Participant Demographics
The study involved 194 subjects from 14 pedigrees.
Statistical Information
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website