lassesen 5:56 am on April 20, 2019  

Mixture of Probiotics Ratio

A reader wrote with an interesting question:

I want to create 2 probiotic formulas for myself but need help in determining approximate percentages of each of these strains.
% of: B-Bifidum, B-Breve, L-Salivarious etc

The question is how to answer this!

Ways of Determining Ratios

Looking at this question, I see several logical roots:

  • A proprietary blend (keep secret on the ratio – more below)
  • The percentage seen in appropriate human samples (usually healthy controls)
  • Result of experiments with different ratios and picking the ratio with the best results.

I will deal with the first and last approach first, and then go on to the middle one.

A proprietary blend

Unfortunately, there are no published studies in this area – proprietary means not public. Existing formulation on the market appear to be based on proprietary processes. In better cases it is likely to get past a min level (i.e. reported in some study) for each individual probiotic and then ‘pad’ with the cheaper ones until they hit a BCFU that was determined by marketing studies. In the worst case, it is determined by a marketing study of competitors.

Base on Human Studies with different mixtures

I have never seen such a study done. I recall a very few where there are three groups: {Control, Mixture of A and B, just A}. This does not answer the percentage question at all.

This is further complicated because of some studies reporting no results for a probiotic and another study reporting positive results. Doing a proper valid study is both very time consuming and very expensive.

Based on observed ratios in Human Samples

This is the simplest to do. It has two challenges:

  • Defining the target population
  • Insuring that the testing (usually 16S) methodology measures all of the candidate strains.

I do have a contributed set of human samples and let us see what the results are assuming this is the target population.

Proposed Bifidobacterium Ratios

The numbers below are the result of a little data manipulation (any one proficient in R can repro these numbers

Looking at the common species likely to be available for probiotics

Proposed Lactobacillus Ratios

The numbers produced are very interesting – Lactobacillus acidophilus is far very down the list!

Looking at the common strains, we see a very interesting set of ratios:

Some 114 taxonomy units of Lactobacillus are currently in the sample

Bottom Line

Well, I did create one possible answer to the question that the user asked. A firm like Thryve or Ubiome can likely produce more definitive data. If either firm care to share their numbers with me — I would be glad to share they results here.

Firms like CustomProbiotics or Seed may wish to produce mixtures matching the above ratios (or obtained in the same way elsewhere). Lactobacillus and Bifidobacterium appear to be close to a 1:1 ratio.

What I found especially interesting is that there appear to be some species that IMHO should be more available as probiotics — but are not. Especially the following: