For a non rigorous filter there is always a trade off between sensitivity (that means in a way correctness) and filtration ratio (that means speed). Depending on the nature of the task one might want to have a very fast program that maybe miss some hits or an exhaustive program with the disadvantage of a hugh runtime.

We ran our RNAsifter with several testsequences to get an insight of its performance. Below you see the results of this evaluation. For more details please take a look at the "Results and Discussion" section of our paper.

If you move your mouse over the various datapoints, the parameter configuration for the related result will show up in the upper left corner of the graph. We changed the names of the parameters a little bit: kfamily of the paper is called firstXdb, kquery is firstXquery and �� is the energyTolerance.

The parameter sets we recommend for average usage are marked in magenta.