My project is genetics based and looking at links between a particular disease and possible genetic links. The project plan is to screen a bunch of people with the disease and without the disease, look at some of the DNA sequences and analyse data to see if there is any connection.

So I'm looking into collection blood samples and possibly saliva sample or swabs to get the required genomic DNA as it's the easiest option. However my supervisor is adamant that the samples must come directly from the diseased tissue, which makes is much more difficult to collect, not to mention get ethics approval and funding for.

As this is genomic DNA I don't see the difference where it comes on the body, but can't seem to get my supervisor to understand. It's getting to the point where things are actually getting a bit heated and I don't want to pursue it further.

I'd appreciate any advice on this.

I don't know anything about DNA but... is it the case that there are multiple ways of doing this and the easier option is adequate but the harder option would be optimal? I am just wondering why your supervisor is keen for you to do the harder option. Perhaps it would get a stronger publication as it is seldom done that way?

From what you've said - you both want different things - both ways are acceptable but one is harder to achieve (or impossible)... so how about you make a plan of data collection for the more difficult to obtain type of samples, and follow through with it. Alongside you have a back up plan of saliva samples or swabs. You present this to your supervisor and show that you are going to try and get the more difficult samples (presumably there is some rationale for this?), but as you perceive that there will be difficulties you also want to have a Plan B in place. Then you go ahead try and get both.

Could this work?

Hard to advise when I don't fully understand the situation - but in a nutshell I mean try and approach it diplomatically. It isn't helpful if you go in with the attitude that your supervisor is "wrong" and you are "right". Surely he has some rationale for wanting the other kind of sample? And if he doesn't and it is utter nonsense, you can still deal with it diplomatically...

Hope this helps.

Maybe it depends on the disease you're looking at? If it's something like cancer then mutations might only be in the cancerous cells, so there's not much point in looking at DNA from elsewhere

If not, then yeah, any genomic DNA will do. Are you sure it's not that your supervisor wants you to look at the transcriptome ie the RNA or something?

Mmm... I think your supervisor might be right. It depends on what you are doing. Are you doing Rna-Seq or analysing chromatin remodelling? (Or any other technique to check expression levels)

Different cells express different genes depending on function (differentiation). The genetic code will be the same for blood or tissue sample, but to truly compare what is going on you need to know which ones are being expressed in tissue (if healthy patient has gene A and disease patient has gene B instead , but these genes are not being expressed in your target tissue, it has no value for the question you are asking).

But, if you have:
Genes 1, 2 and 3 expressed in healthy samples
Genes 1,2, 3 and 4 in disease ones
You can make an assumption that gene 4 causes disease.

Does your supervisor have a background in molecular biology or genetics? If not, then you need to educate them a bit because you're correct, if they do, then you need to get them to explain to you why it makes a difference what cell type the DNA is extracted from.

You are probably not the first guys to look at a disease using a sequencing approach. If you are absolutely sure that it does not matter, then maybe you can find some papers to support your point. It always helps me when my supervisors propose something that I don't agree with ;)