InfluentialPoints.com Biology, images, analysis, design... 

"It has long been an axiom of mine that the little things are infinitely the most important" 

Just a note
The red cross on the first graph indicates the coefficient of variance we estimate for our observed result of All of these results were obtained by repeatedly performing permutation tests  randomly classifying 5 observations as 'controls' and the remaining 10 as 'treated'. Each combination of test location and number of randomizations were tested 500 times (so, for the 12 points on our first graph, we performed a total of 6000 tests and 22,200,000 randomizations). The resulting Pvalues are the proportion of the randomization distribution, excluding the location under test, that were more extreme than that location  thus making zero the lowest possible Pvalue. If instead you include the observed value, as is customary, the lowest obtainable 2tailed midP value is 1/R.
