Statistical tests to compare motif count exceptionalities
2007
Statistical Tests for Comparing Motif Counts in DNA Sequences
publication
Evidence: moderate
Author Information
Author(s): Robin Stéphane, Schbath Sophie, Vandewalle Vincent
Primary Institution: INA PG/ENGREF/INRA
Hypothesis
Can we statistically compare the exceptionality of a motif in two different DNA sequences?
Conclusion
The exact binomial test is best for small counts, while the likelihood ratio test is recommended for larger counts.
Supporting Evidence
- The exact binomial test is suitable for small counts.
- The likelihood ratio test is recommended for larger counts.
- Both tests can account for sequence compositions.
Takeaway
This study shows how to compare DNA motifs in two sequences to see if one is more special than the other using math.
Methodology
Two statistical tests were developed: an exact binomial test and an asymptotic likelihood ratio test to compare motif counts.
Limitations
The tests are valid only for independent sequences and may not detect skewed oligomers.
Statistical Information
P-Value
pB = 8.6 × 10^-2
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website