Identifying differential exon splicing using linear models and correlation coefficients
2009

Identifying Differential Exon Splicing Using Linear Models

Sample size: 33 publication 10 minutes Evidence: moderate

Author Information

Author(s): Sonia H. Shah, Jacqueline A. Pallas

Primary Institution: University College London

Hypothesis

In the absence of differential splicing, we would expect a correlation coefficient close to 1.

Conclusion

The study demonstrates that LIMMA can identify differential exon splicing from Affymetrix exon array data, and that the correlation coefficient approach is useful for identifying differentially spliced genes.

Supporting Evidence

  • The LIMMA approach identified several tissue-specific transcripts and splicing events supported by previous studies.
  • Filtering the data reduced the false positive rate significantly.
  • The correlation coefficient approach was better for identifying genes with many differentially spliced exons.

Takeaway

The researchers created a method to find differences in how genes are spliced in different tissues using a special analysis tool, which helps understand gene behavior better.

Methodology

The study used the LIMMA package to analyze gene-normalized exon intensities from Affymetrix exon arrays, applying filtering steps to reduce false positives.

Potential Biases

The analysis may misestimate gene expression due to the presence of cross-hybridizing probes and low-confidence annotations.

Limitations

The study's filtering approach may exclude viable splice candidates, and the lack of a gold standard dataset makes it difficult to benchmark the method.

Participant Demographics

The dataset consists of 33 human tissue samples with three technical replicates per tissue.

Statistical Information

P-Value

p<0.0001

Statistical Significance

p<0.0001

Digital Object Identifier (DOI)

10.1186/1471-2105-10-26

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication