search

llrs / experDesign

Design experiments distributed in several batches
https://experDesign.llrs.dev
Other
10 stars 1 forks source link

Random subsample for QC #2

Closed llrs closed 5 years ago

llrs commented 6 years ago

Sometimes it is advised to make some checks on a subsets (like running a gel to see if the primers bind correctly).

It would be nice to have a function to select a representative sample or arbitrary size it could be either of all the samples or some from each batch

llrs commented 6 years ago

The subset for QC should try the opposite of design: maximise the difference, so that it has a representative sample of each group.

llrs commented 6 years ago

Also the random subsample could be distributed in each batch (ie: 10 samples of each batch are checked )

llrs commented 6 years ago

qcSubset select the samples randomly

llrs commented 6 years ago

To select the number of samples required to get 50% or more an iterative process could be done. Proposed method: 1) Find each factor 2) Calculate how much of each factor there is 3) See what happens with the probabilities each time we add a new sample, (the interval of probabilities for each category)

llrs commented 5 years ago

This is also doable with the extreme_cases function