How many real time-PCR reactions for statistics and how? - (Jun/11/2012 )

Hello.

The task is to follow expression of four genes (incl. the housekeeping) in six experimental groups (three individuals each).

This makes 76 single ractions (all samples (3) from all groups (6) with all primers (4) + 4NTC).

I was told that two real time plates can not be compared but I wonder how the statistical evaluation would be done?

Which variant would be better:

1. One plate with 76 single reactions and some.....calbration between them (I heard some trick with the treshold line) - three-four times;

OR

2. One plate with 1 individual from each group with all primers, in triplicate - three times, for all individuals;

OR

3. One plate with all samples fom all groups (18) and only one primer, in triplicate - four times for the four primers?

Thank you

Hi, imho variant 1 and 3 is ok. you also don't need calibration between the plates then. you need calibration (inter-run calibration, IRC) if you would have samples on different plates analyzed with the same primer pair and you want to compare them, but this is not the case in v1 and v3 but in variant 2!

Thank you!

I would do it like variant 3: all samples in triplicates with 1 primer per plate.

Thank you once again, Tea-test - your advise makes me more confident