Calculating Confidence Intervals for DNA Transformation Experiments

[azheid]

So I have a research problem and I am not very good at statistics. I need to amplify a library of dna molecules with a set number of dna species and transform them into a organism. Basically, I need to know how many moles of the library I need in order to be confident that I am getting at least one transformant of each of the dna species when I know the efficiency of the transformation step.

If I assume that each dna species is equally likely to be transformed, then this problem sounds exactly like some of the statistics homework questions that I was never very good at. Math comes back to haunt me. I want to know for a high confidence level, how many transformants I need to ensure that I get at least one transformant of every different dna species. So if I want 99% confidence, and I have X dna species in a pool, how many times must I pick from that pool to ensure I get every one?

 

[Stephen Tashi]

I want to know for a high confidence level, how many transformants I need to ensure that I get at least one transformant of every different dna species. So if I want 99% confidence, and I have X dna species in a pool, how many times must I pick from that pool to ensure I get every one?

Only someone familiar with your laboratory procedues would understand your description of the problem. Try describing the problem using specific numbers or letters to represent each different quantity involved.

You mention:
"a libarary of DNA molecules"
"a set number of DNA species"

What is the relation between these things. Are there, for example, 20 DNA species per mole of the molecules? For given DNA species, is there more than one representative of that species in a mole of the molecules? If so, how many?

You say "If I assume that each dna species is equally likely to be transformed...", but it isn't clear whether a "tranformation" can only transform one species at a time. Does a "transformation" act on a given quantity of molecules? If it transforms one speces (or one representative of that species) is it unable to tranform any different species or different representatives of the same species?