This new publication in Genetics comes from a collaboration between the Bank and Jensen labs; the statistical analyses were performed by Sebastian Matuszewski and Marcel Hildebrandt. The authors present a statistical guide (and web tool, implemented by Hermina) that can be used to design and improve your future experimental setup of a deep mutational scanning study. Deep mutational scanning is an approach in which hundreds or thousands of pre-engineered mutants are grown together and their relative abundance, directly related to their fitness, is assessed through deep sequencing at several time points.
The main requirements for the statistical model of the paper to be applicable are: 1) Large copy number of each mutant in the initial library; 2) Mutants need to grow exponentially during the experiment; 3) Both population and sample sizes need to be large; 4) You should to collect your data at two or more time points.
With the interactive tool that comes with the publication, it is possible to introduce and manipulate several experimental parameters, and it allows you to quantify and maximize not only your experimental resolution, but also to get estimates of the time and expenses for a particular experiment.