In any case, it is inescapable that conclusions drawn from null-effect studies that contain significant biological flaws reveal essentially nothing about nutrient efficacy.
Accordingly the discussion focuses on the positive results observed, since few conclusions can be drawn from the null results.
Substantial overlap in 95% confidence intervals (Fig. 4) is suggestive of indifference, but clearer conclusions cannot be drawn from a null result alone.
The p-value is the probability that a matrix drawn from the null distribution will be more nested than the input matrix.
For each window, a binomial trial (P = 0.01) was used to determine whether the phenotype should be drawn from the null or alternate.
We were able to reject the null hypothesis that the asthma GWAS P-values for cis eQTL SNPs paired with the 117 lymphocyte count-associated genes were drawn from a null distribution (by permutation, P = 0.02; Fig. 5).
The DABG P-value describes the probability that an intensity value at least as extreme as the observed could have been drawn from the null distribution, in this case the background distribution of intensity values.
Summary: Q vality is a C++ program for estimating two types of standard statistical confidence measures: the q-value, which is an analog of the p-value that incorporates multiple testing correction, and the posterior error probability (PEP, also known as the local false discovery rate), which corresponds to the probability that a given observation is drawn from the null distribution.
We also calculated the FDR (Storey 2002), the expected ratio of the number of false positives to the total number of positives, or the expected fraction of results drawn from the null distribution at a given significance level (Benjamini and Hochberg 1995).
Following the same approach as that outlined for the analysis of the GABRIEL data set, we were again able to reject the null hypothesis that asthma GWAS P-values for variants classified as cis eQTL for the 117 candidate genes were drawn from the genome-wide null distribution (by permutation, P = 0.027; Fig. 5).
In well-behaved situations where Wilks theorem [ 5] applies, the asymptotic distribution of LRT is known, provided the data was drawn from a distribution which satisfies the null hypothesis H0.
