Reference to paper I have following excerpt.
100,000 random counts were generated from a Poisson distribution with a mean of 20, simulating an RNA-Seq experiment consisting of 10,000 transcripts, each with 10 replicate measurements. A Fisher exact test was performed on the sum of all counts from the first five replicates versus the sum of all counts from the second five replicates. The percentage of transcripts with a p-value less than or equal to 0.05 was taken to be the false positive rate. A t-test was also performed for the first five replicates versus the second five replicates of each simulated transcript, and the corresponding false positive rate was again calculated at the 0.05 p-value level. Next, five replicate values were generated from a Poisson distribution with mean 25 for each of the 10,000 transcripts, corresponding to a true fold change of 1.25. The differential expression analysis was repeated on this data set using the Fisher exact test and the t-test to evaluate the sensitivity of each to detect the relatively low 1.25 fold change at the 0.05 p-value level
Can somebody guide me in python how can I perform such an experiment?