I am helping out with an experiment with multiple bacterial cultures (>20 species). I need to a very specific concentration (cells/mL or CFUs*/mL) for this purpose. Unfortunately it appears that standard growth curves (OD** vs. CFU/mL) are not available for these cultures. I am not in a position to generate them afresh either given the sheer number of cultures I have to work with and how laborious the growth curve experiments are.

Given these constraints, are there other reliable (and needless to say fast and accurate) ways through which I can find out how many cells/mL (or CFUs/mL) I have in a given bacterial culture at a given OD? All the more better if the method is parallelizable.

I came across this study (Robust estimation of bacterial cell count from optical density.) where they suggest using silicon microspheres with dimensions similar to E. coli to reliably estimate ODs, but I don't know what the cell sizes for the bacteria I am working with are. In this case, what are some easy and accurate ways in which I can estimate the average size of these bacterial species?

* - CFU - colony forming unit

** - OD - optical density

  • 1
    $\begingroup$ Do you have access to a flow cytometer? That would be one way of making direct counts of a culture. And, of course, there's always the good ol' hemocytometer. $\endgroup$
    – MattDMo
    Commented Oct 19, 2022 at 21:33
  • 2
    $\begingroup$ You really need something tangible to tie back to the OD, whether it's CFU, MPN, or direct cell counts. Without some kind of standard curve, you're basically just guessing. If you have a plate reader, you could do your OD dilution series on that, and follow up with a quick drop plate assay (basically a spot titer for CFU) to get you in the ball park (probably within half a log or so). Should be easy to blow through 20 bugs in a day or so (three or four days if you want a few replicate measurements). Just realize your plate reader OD will not be identical to your cuvette OD. $\endgroup$
    – MikeyC
    Commented Oct 19, 2022 at 22:49

2 Answers 2


I would argue that performing a full growth curve isn't strictly necessary to obtain a standard curve to convert between OD and CFU. As long as you're collecting cells in a standardized way, a standard curve generated from a simple dilution series of cultured cells can prove quite reliable.

For example, you can harvest cells from your culture in early stationary phase and generate a 2-fold dilution series, collecting OD measurements for each dilution until you get below the threshold for which your OD reader can generate accurate OD measurements. Then switch to 10-fold dilutions and plate those for CFU counts. You can then back-calculate the CFU/mL present at each of the OD measurements for your standard curve.

The same concept can be applied to overnight colonies suspended in normal saline or PBS. Pick colonies from an overnight streak plate and suspend them in saline until it's turbid, with an OD reading >1.0. Do the 2-fold dilutions to get multiple OD readings until you bottom out, and plate for CFU. This is generally how innocula are standardized for determining MIC values in broth dilution methods, and the range of acceptable CFU concentrations for those methods is often quite narrow.

Whatever method you use, replicate it several times to make sure you're getting consistent conversions, stick with that method for your experiment, and always verify your experimental CFU.


You can estimate cell size through a microscope. However cell size can vary significantly in different stages of growth, or growth under different conditions.

  • $\begingroup$ @markur I actually did ask about this also. $\endgroup$
    – Dunois
    Commented Oct 20, 2022 at 15:44
  • $\begingroup$ ah I see, I deleted my comment $\endgroup$
    – markur
    Commented Oct 20, 2022 at 16:47

You must log in to answer this question.

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