Robust regression for periodicity detection in non-uniformly sampled time-course gene expression data
2007

Robust Methods for Detecting Periodicity in Gene Expression Data

Sample size: 7679 publication 10 minutes Evidence: moderate

Author Information

Author(s): Miika Ahdesmäki, Harri Lähdesmäki, Andrew Gracey, Ilya Shmulevich, Olli Yli-Harja

Primary Institution: Institute of Signal Processing, Tampere University of Technology

Hypothesis

Can robust regression methods effectively detect periodicity in non-uniformly sampled time-course gene expression data?

Conclusion

Robust regression methods can effectively handle non-uniform sampling in biological measurements and improve periodicity detection.

Supporting Evidence

  • The robust methods developed are expected to have many uses in biological measurements.
  • M-estimation provides a good compromise between robustness and computational efficiency.
  • The study found no statistically significant connection between circadian rhythm and cell cycle regulated genes.

Takeaway

This study shows that using special math methods can help scientists find patterns in gene data, even when the data isn't collected at regular times.

Methodology

The study developed a framework for robust periodicity detection using regression methods and tested it on simulated and real measurement data.

Potential Biases

Potential biases may arise from the assumptions of the noise distribution and the choice of regression methods.

Limitations

The methods may not perform well with highly irregular sampling or extreme outliers.

Participant Demographics

The study analyzed gene expression data from the mussel Mytilus californianus.

Statistical Information

P-Value

0

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-8-233

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication