What Is A Well-Dish?

Question

[Layla]

Hi: I have read that a batch us a treatment and control performed together with the following issue: - if its performed in 6 well-dishes we will have 4 treatments and 1 calibrator - if its performed in 96 well-dishes, we have 41 treatments, 1 calibrator and 6 vehicle controls

Can somebody explain me why in the first model there is no vehicle control? and why in the other one there is only 41 treatments, does it mean that the remaining ones are controls? how I can decide those numbers? The information is obtained from the help of the CMAP tool that says:

batch: The set of treatments and controls performed together. Each batch is uniquely identified by a batch number. Batches produced in six-well dishes contain four treatments, one calibrator compound and one vehicle control (and have batch numbers less than 500 or greater than 999). Batches produced in ninety-six-well dishes contain forty-one treatments, one calibrator compound and six vehicle controls.

and the link is: http://www.broadinstitute.org/cmap/help_topics_linkified.jsp#batch

I am a newbie in this stuff so it would helpful if somebody could give me some hint Thanks

PS I see that this is also a bioinformatics related question, because when one tries to tweak the information contained in CMAP one should go to the lower level of Microarray analysis.

[Obi Griffith]

Its a little hard to follow this question. But, from what I can tell it is not a bioinformatics question. I suggest you post this to a forum that is geared towards biological/experimental questions. You might try the biology stack exchange. When you do so, maybe explain where you are getting the numbers for these plate layouts from. There is practically an infinite number of ways of setting up such an experiment. Without more details/clarification it will be impossible to comment. Closing as off topic.

[seidel]

If the question is closed, can the asker still edit it to be better? I find questions related to experimental design directly related to bioinformatics, as it is directly related to evaluating the information that comes from the experiment. Explaining controls for experiments, and evaluating statistics, is of interest to me as someone who engages in bioinformatics. How experiments are controlled can be confusing for everyone and in my opinion can't be discussed enough. I agree that the question could be better asked, but I hope shutting the question down doesn't exclude the possibility of improving it.

[Obi Griffith]

That is a good question. I believe that the questioner can still edit the question even though it has been closed. But, I am not sure. Hopefully an admin can jump in and confirm? If it is improved to be more understandable and have a bioinformatics angle then I would support re-opening it. I agree that questions of experimental design are very important but this particular question didn't seem to tie in to informatics. At least as far as I could tell.

[Istvan Albert]

Yes a user can edit their own question. Any user can also close or delete their own questions. They cannot however reopen or undelete a question even if they were the ones closing/deleting it.

[Layla]

It was just a question that it was found in a purely bioinformatics paper, I think that it was not so nice that they put -3 points just for this question

[Obi Griffith]

Yes. This can seem like a very harsh welcome to the forum. Why don't you edit your question to include a reference to that bioinformatics paper and try to clarify a bit. Then, we can open this question up again. I have no doubt you will then get some answers to your question and those negative votes will be replaced by positives.

[Layla]

well it cannot be edited, because the post is closed, by the way this was described in the CMAP tool; so it seems really harsh to put negative points for trying to know more about the inner caveats of this software

[Obi Griffith]

If you read the thread above you can see that an admin in the forum has said that the post can still be edited by the user even if closed. Is that not the case? Details like where this experimental design was described (which you have just provided) are what we are trying to get from you so that we can understand and help you with your question.

[Obi Griffith]

I have tried to find the documentation you are referring to for CMAP tool (is it here? http://cmap.ihmc.us/). I see many publications and references associated with that software but not one describing the experiment you are referring to. Please edit your question with links or quotes of relevant text.

[Istvan Albert]

you can edit it - but I'll just open the post - your original question seemed off topic but this discussion revealed that there could be reason to reopen -

[Layla]

thanks for your comprehension in the topic and for reopening the post

[Istvan Albert]

well really it was Obi Griffith that worked most on it, best

Answer 1

[Obi Griffith]

Thanks for clarifying your original question and providing links. I will make a first attempt at answering your questions.

First, "What is a well-dish?". You are quoting from help documentation for the connectivity map project where they are describing various experiments using different multiwell cell culture plates. They refer to "six-well" and "ninety-six-well" plates (aka dishes). These are two common formats of cell culture plates (the other two most common formats are 24- and 384-well).

Second, "Can somebody explain why in the first model there is no vehicle control?". The answer is that they do have a vehicle control. Quoting from their documentation "Batches produced in six-well dishes contain four treatments, one calibrator compound and one vehicle control".

Third, "why in the other one there is only 41 treatments, does it mean that the remaining ones are controls?". On the 96-well plate they use 41 wells for treatments, 1 for the calibrator compound and 6 for vehicle controls for a total of 48 wells. This is exactly half of the plate. So my guess is that they are putting two batches of 48 to completely fill each 96-well plate for each round of treatments. Another possibility is that they are just leaving many wells empty. This is a common practice in cell culture experiments on 96-well plates. For example, if optical density measurements are being taken it is common to leave all 36 wells at the periphery empty. But, this doesn't quite match their description in this case.

Finally, "how I can decide those numbers?". It is likely that they chose these designs for primarily practical reasons in order to use the 6-well and 96-well plates most efficiently while still including a minimum number of controls. The connectivity map experiment was a very ambitious project where they profiled expression of several cell lines against many perturbations. They had to balance the advantages of more replicates and controls against their desire to profile many perturbations. If you are planning a similar but smaller scale experiment you might have different criteria for deciding numbers of vehicle controls and so on.

[Layla]

thank you for your help, that clarification was great!