What does "cell type composition" mean?

Question

I am reading an article about detection of differential methylation. They use the term "cell type composition" in the following context:

"Researchers need to carefully design studies that associate phenotypes with DNA methylation. Some aspects, such as cell type composition, cannot be readily controlled by design; patients and therefore individual DNA samples simply differ in their cell type composition." (Robinson, MD., 2014)

I would like to know to what "cell type composition" refers?

Does it refer to the composition of a specific cell-type in terms of its methylation pattern (or can it refer to something more general outside this context)? It makes sense to me that different cell-types should have different methylation patterns, that's why there are different cell-types in the first place. But is it that even within one specific cell-type, the composition of methylation pattern is different across different individuals?

Answer 1

The problem is that for most sequencing operations you nowadays still need more than one cell - even though this will change soon.

Accordingly, you have to sample cells from a tissue which often, if not usually, contains more than one cell type, see e.g. this picture from a lung tissue from the Wikipedia article on tissues linked above:

---

---

It is not always easy to pick cells of one type only (think of trying it with blood, which is a tissue as well). So what you then get is a mixture of different cell types which probably differ in methylation patterns.

Does it refer to the composition of a specific cell-type in terms of its methylation pattern (or can it refer to something more general outside this context)?

No, it means that you have cells of different types within your sample.

It makes sense to me that different cell-types should have different methylation patterns, that's why there are different cell-types in the first place.

At least one of the reasons, there are more, e.g. other modifications and mutations.

But is it that even within one specific cell-type, the composition of methylation pattern is different across different individuals?

That is a different problem and may also be true. But this is not what your quotation, as I understand it, refers to. They refer to the problem that you might sample the same tissue across patients but still get a different composition of cells from different types due to sampling error.