I mean I read it in every paper that "Rat of weight xxx-yyy" were taken and.... What is so important about choosing a rat of particular weight for microarray studies though?
I would not narrow myself down to a certain weight range, but please make sure that your test groups have that same weight distribution. Please note that weight is a lognormal distributed variable and it differs by gender as well. As a general suggestion it is always best to balance all your study parameters as best as you can and vary only the study parameters that you want compare your expression array results e.g. same weight, age, same diet ….
As Mark pointed out correctly unless you want to address an obesity related question the weight should be play not too much of a role in your study design as long as keep it the same for your test groups.
Study design is no rocket science, but this is an important step on the way towards biomarker discovery or disappointment. Biomarkers are rare and usually not black and white, but they come in different shades of grey. Therefore it is important to design your study and experiments in robust way and ensure that you minimize all environmental effects and record as much information about the experiment as possible.
Getting professional help at this stage can save you time and money and is always faster and cheaper than to repeat the study with improved design or to write it off.
Here are some general suggestions
> randomize the study groups e.g. don’t give all case samples to one operator and all control samples to another operator
> calculate the sample size that is required for a study
> create SOP’s and perform experiments accordingly
> train staff to perform experiments accordingly
> create a database or spread sheets to record important experimental variables such as buffer batches, operator, instruments, calibrate instruments etc.
> identify best data normalization method if that is required before entering into real life study
> perform a small scale pilot study to test the steps above
> perform repeat assays during the study and compare the reproducibility of the assays
www.biomarker-discovery.com
Feel free to have a look at the link below. It is about study design.
I don't think it is important to choose a specific weight, but rather you want to show that all the rats in your study are similar in size. For example, say you are conducting a study on cholesterol synthesis, and you choose one group of obese rats and one group of very skinny rats, people will think you are skewing the results. It is much easier to simply state that all rats are between x and y grams.
I would not narrow myself down to a certain weight range, but please make sure that your test groups have that same weight distribution. Please note that weight is a lognormal distributed variable and it differs by gender as well. As a general suggestion it is always best to balance all your study parameters as best as you can and vary only the study parameters that you want compare your expression array results e.g. same weight, age, same diet ….
As Mark pointed out correctly unless you want to address an obesity related question the weight should be play not too much of a role in your study design as long as keep it the same for your test groups.
Study design is no rocket science, but this is an important step on the way towards biomarker discovery or disappointment. Biomarkers are rare and usually not black and white, but they come in different shades of grey. Therefore it is important to design your study and experiments in robust way and ensure that you minimize all environmental effects and record as much information about the experiment as possible.
Getting professional help at this stage can save you time and money and is always faster and cheaper than to repeat the study with improved design or to write it off.
Here are some general suggestions
> randomize the study groups e.g. don’t give all case samples to one operator and all control samples to another operator
> calculate the sample size that is required for a study
> create SOP’s and perform experiments accordingly
> train staff to perform experiments accordingly
> create a database or spread sheets to record important experimental variables such as buffer batches, operator, instruments, calibrate instruments etc.
> identify best data normalization method if that is required before entering into real life study
> perform a small scale pilot study to test the steps above
> perform repeat assays during the study and compare the reproducibility of the assays
www.biomarker-discovery.com
Feel free to have a look at the link below. It is about study design.