Apples to Walnuts in GI Tests

A reader sent me a new GI test results as a PDF today. Since I have just converted the XenoGene (Spain) PDF report into an automatic report, my initial thought was — let us convert this one….

Brakes applied…. I noticed the units of measurement: Community Forming Unit/gram

This is not the same unit of measurement as uBiome, American Gut, Thryve, XenoGene and other 16s tests. Those tests are the percentage of the total CFU.

  • In the colon, bacterial concentrations of 10(11)-10(12) cfu/g faeces are found.  [1982] In other words, a 10x variation in CFU/g.

The second item is simple Colony Forming Unit — means counting the bacteria that grows on a sample plate or in a test tube. Not all will grow at the same rate — thus the numbers are post cultivation. 16s Samples do not require cultivation and thus are far closer to the numbers actually in the sample.

So, bottom line: you cannot compare results between these two types of tests — even when done on the same sample! For example, in the above test we see normal for Enterobacteriaceae as being < 9 10(4) CFU/gram. Checking on our uploaded samples for Enterobacteriaceae, we see the average is 10%, but less than 25% of samples are over 5%. Using the lower of the reported ranges: 9×10(4)/10(11) we find that the above test translates to: 0.00009%. Using the higher value.. 0.000009% . That’s a difference of 55,555x!!!!

Are the CFU/gm numbs wrong? NO, they are correct for the process that the lab is using BUT THIS IS NOT THE ACTUAL AMOUNT, it is the count after cultivation.

This is not an apple-to-oranges situation. It is an apple-to-walnuts (perhaps watermelons-to-mustard-seed).

The Other Labs Page ( ) does not capture the same numbers as ubiome, thryve, xenogene, american or british gut. It asks the user to enter green, yellow, red equivalency from the received reports. It also assumes that the labs are correct with their ranges and coding.

Personally, I doubt their ranges, they look so very very much like what you would get using averages and standard deviations. A methodology that I do not think apply to microbiome samples (as shown above for Enterobacteriaceae )

Bottom Line

If we are talking about an overgrowth of one or two bacteria, then the older cultivated lab tests are adequate. If we are talking about a complex shift of microbiome then we really (IMHO) need to use 16s tests (XenoGene, Ubiome, Thryve, American or British Gut). From these tests, we can see the complex relationship between bacteria as shown below (the Why? Citizen Science, and Why? charts)