Correcting Probe Signals in Methylation Studies
Author Information
Author(s): Dustin P. Potter, Pearlly Yan, Tim H. M. Huang, Shili Lin
Primary Institution: The Ohio State University
Hypothesis
Can we develop models to correct for non-biological signal in differential methylation hybridization experiments?
Conclusion
The proposed models effectively correct for non-biologically relevant probe signals, with most parameters being statistically significant.
Supporting Evidence
- The majority of estimated parameters were statistically significant in all considered models.
- The quadratic model showed more biologically relevant results compared to the full model.
- Both models effectively corrected for signal effects associated with probe sequence.
Takeaway
This study created two models to help fix errors in measuring DNA signals, making it easier to understand how genes are affected by methylation.
Methodology
Two model-based approaches were developed to correct for probe-sequence effects and dye-bias in DMH data.
Potential Biases
Potential biases due to probe sequence composition and dye interactions were noted.
Limitations
The biological significance of the estimated values in the models is questionable despite their statistical significance.
Participant Demographics
The study used data from 9 randomly selected breast cancer cell lines.
Statistical Information
P-Value
p<0.01
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website