In default mode, GREAT only displays the top 20 terms for each category which pass two statistical tests (p-value < = 0.05): A binomial test which accounts for over-representation in genomic regions, and a hypergeometric test which accounts for over-representation in functionally associated gene sets.
The obvious tool to analyse the result is a binomial test, which is trivial to anyone with a basic grasp of statistics, but exotic to those who work in the humanities and so is briefly described below.
Interchangeability of goniometers was also demonstrated by the binomial tests, which involved assessment of the inter-goniometer ≤ 5-degree agreement.
We asked how likely these counts were under a binomial distribution in which the probability of success is one minus the conversion rate, and we corrected this probability value for multiple testing (Benjamini and Hochberg 1995).
The P-values were calculated according to the binomial test integrated in the PANTHER online tool.
This conclusion results from the χ2 goodness-of-fit test [22], in which we considered a wide range of hypothesized null distributions (i.e., Poisson, binomial, negative binomial, normal, gamma, lognormal, Weibull, beta; geometric, exponential, chi-squared, uniform).
This excess was statistically significant at the pre-specified level of 0.10 (binomial probability test p=0.05) which may indicate bias in trials with published results.
Here, the obtained p value from a binomial test is not identified in a traditional way which a significance level at 0.05 or 0.01 are usually defined.
Differences in seroprevalence rates were tested using binomial test for differences in proportions.
