A long time ago there were only three kinds - red, white, and rose platelets. Advances in cell sorters and more recently in genetic techniques and differentiation studies have led to the identification of a wide variety and complex lineage of cells that comprise human blood.

Roughly at least, how many types of cells are currently identified?

I understand that "types" is a non-technical term, and I'll try to better define what I mean. For example, I do not mean to enumerate each antigen specificity separately because that would make the number fairly large and probably dependent on the individual's history.

One possible definition of a "type" would be that it has been given a name.

I don't mean to include cells that are present due to injury, clean-up, trash collection, or foreign cells including bacteria, viruses, parasites etc.

Just for an example, here is an overview of types of hematopoietic stem cells. There may be further recognized subdivisions, I don't know how far it goes. There are certainly other cells that comprise human blood. (From here.)

enter image description here

  • $\begingroup$ The short answer is that there are still three "kinds" of blood cells: red (RBCs), white (WBCs) and platelets. The graph above names pretty much all of them. $\endgroup$ Apr 9, 2017 at 3:32
  • $\begingroup$ @anongoodnurse I always liked the sound of T-helper cell - are they not a well-defined type? $\endgroup$
    – uhoh
    Apr 9, 2017 at 3:35
  • $\begingroup$ Maybe discussing an application or disease could help. Let's take WBCs as an example. The journal of immunology, which is dedicated to WBCs and immune response, prefers to make sections as innate immunology, allergy, antigen recognition, immune regulation, immune system development, etc. Same story can happen for thrombocytes and erythrocytes. The point is discussing an application can help to answer about each cell "kind/type" more precise. $\endgroup$
    – SoFar67
    Apr 9, 2017 at 6:07
  • 1
    $\begingroup$ @uhoh Here's a chart I cited in a different question. I think that's among the better estimations of what you might find, and in what quantity, ignoring unlimited subsetting. $\endgroup$
    – CKM
    Apr 19, 2017 at 20:33
  • 1
    $\begingroup$ I think any answer will be subjective, depending on what you consider to be distinct cell types --- how detailed should the classification be. As you already mentioned, if you consider all cell populations that can be separated by surface antigens, that's already a pretty large number. Also, what previously was thought to be terminally differentiated cell types are increasingly found to be more like cell "states" that can change over time, so-called "plasticity" (see for example nature.com/ni/journal/v11/n8/abs/ni.1899.html). This makes it hard to even uniquely define a cell type. $\endgroup$
    – Roland
    Jun 11, 2017 at 10:41

1 Answer 1


I think any answer to this question will be subjective, depending on what you consider to be distinct cell types --- how detailed the classification should be.

As you mentioned, if you consider all immune cell populations that can be separated by surface antigens, that's already a pretty large number. In 2013, the Immunological Genome (Immgen) project reported characterization of 249 immune cell types, and I think they are still working on expanding their catalogue (see immgen.org). One can argue about which of these are really different enough to be called "cell types", but I think it gives a pretty good ballpark number.

At some level of resolution though, the notion of "cell type" itself becomes slippery. While previously it was thought that many cell types were terminally differentiated (unable to change back to more primitive forms), we now increasingly find that these "cell types" are more like temporary "cell states", or phenotypes, that can change over time. For immune cells, this phenomenon is often called "plasticity"; for more information, see for example this paper. The advent of single-cell genomics has further strengthened this view, demonstrating that individual cells can dynamically change their gene expression pattens. These issues makes it hard to uniquely define and count cell types, and it might be more useful to reason quantitatively in terms of the phenotypes and behavior we can measure --- expression patterns, signalling molecules, metabolism and so on.

  • $\begingroup$ Nice answer!... $\endgroup$
    – canadianer
    Jun 16, 2017 at 20:18
  • $\begingroup$ When I started to write this question, I chose blood instead of a solid organ (e.g. liver, heart, kidney) because its natural state is individual cells in fluid and so it lends itself more naturally to some types of cell characterization. I'd thought that a 3D organ would have to be sectioned or destructively disturbed before individual cells could be separated and "typed" and that would still be difficult to do without bias. Now I'm wondering if I should ask a new, similar question using one of those solid organs where the "slippery slope" is not so steep. $\endgroup$
    – uhoh
    Jun 16, 2017 at 21:47
  • $\begingroup$ The abstract of your linked Genome Research paper Trapnell 2015 is really well written and works hard to explain the "landscape" well. It turns out "how many kinds of cells are there" in the human body is a 60 year old question, and the historical answer used to be "210". I wonder if a block quote or two from that paper in your answer might be informative to future readers. I certainly found it helpful. Thank you very much for taking the time to post a well-written answer my question and to use the space to teach some important points! $\endgroup$
    – uhoh
    Jun 16, 2017 at 21:51

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .