This is about design and analysis of BSP.
How many clones per group do you usually analyze from one set of experiment (I have three groups)?
And how many independent repeats do you perform?
What is the currently most accepted and statistically relevant way of design and analysis?
Thank you in advance!
Designing and analyzing BSP
1 reply to this topic
Posted 22 April 2009 - 06:58 PM
After BSP and cloning, at least 10 clones of plasmid from each PCR SAMPLE should be sequenced. Two independent experiments should be carried out (for such large scaled experiment, I guess many just do once). In terms of statistic analysis, the situation is a bit complicated because you have groups, samples in groups, 10 clones for each sample, the status of several to over a dozen CpG sites within each clone. Both parametric and nonparametric stats can be applied. For example, for each sample, you can obtain the percent of methylated clones (a clone that show methylation at any CpG), you can also calculate how many CpG sites are methylated for each sample. There is no most accepted methods, you can decide which method to use based on your preference and your data.