Dear all.
In paper which named as "**Initial sequence of the chimpanzee genome and comparison with the human genome", published on nature (Vol 437|1 September 2005|doi:10.1038/nature04072)**, the author described a new method about how to get rapidly evolution go categories, the correspondent information as listed here:

**The binomial probability of observing X or more non-synonymous substitutions, given a total of X þ Y substitutions and the expected proportion x from all orthologues, was calculated by summing substitutions across the orthologues in each GO category. For the absolute rate test, Y ¼ the number of synonymous substitutions in orthologues in the same category.
For the relative rate tests, Y ¼ the number of non-synonymous substitutions on the opposite lineage. Note that this binomial probability is simply a metric designed to identify potentially accelerated categories, it is not a P-value that can be used to reject directly the null hypothesis of no acceleration in that particular category.
For each test, the observed number of categories with a binomial probability less than 0.001 was compared to the expected distribution of such outliers by repeating the procedure 10,000 times on randomly permuted GO annotations. The significance of the number of observed outliers n was estimated as the proportion of random trials yielding n or more outliers.**

Now, my question is how to perform random permuted GO annotations, and which content can be random permuted in this step? If you have some suggestions, please tell me. Thanks for your help. Best Regards