Threshold-free high-power methods for the ontological analysis of genome-wide gene-expression studies
2007

New Methods for Analyzing Gene Expression Data

Sample size: 10000 publication 10 minutes Evidence: high

Author Information

Author(s): Nilsson Björn, Håkansson Petra, Johansson Mikael, Nelander Sven, Fioretos Thoas

Primary Institution: Lund University Hospital

Hypothesis

Can threshold-free methods improve the ontological analysis of genome-wide gene-expression studies?

Conclusion

The study introduces new threshold-free methods for ontological analysis that are statistically powerful and provide better detection of relevant gene categories.

Supporting Evidence

  • New methods were evaluated using simulations with 10,000 gene scores.
  • Threshold-free methods detected more categories than traditional discrete methods.
  • Results showed that different methods focus on different types of gene categories.

Takeaway

This study shows new ways to look at gene data that don't rely on arbitrary cutoffs, making it easier to find important patterns in how genes behave.

Methodology

The study uses extensive simulations and real microarray datasets to evaluate the performance of new threshold-free methods for detecting gene categories.

Potential Biases

The study acknowledges that the choice of method can significantly influence results, which may introduce bias.

Limitations

The methods do not account for dependencies between genes, which may lead to underestimated p-values.

Statistical Information

P-Value

p<0.001

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1186/gb-2007-8-5-r74

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication