For items like antibiotics and probiotics, I have for a long time been a strong advocated for continuous rotation. The original source for this attitude was Cecil Jadin’s treatment protocol for occult rickettsia (which originated with the Pasteur Institute for Tropical Medicine). This was followed by reading studies finding that rotating or even just pulsing (2 weeks on/ 2 weeks off) was more effective in reducing bacteria than continuous. Probiotics often function via the natural antibiotics they produce (a lot of prescription antibiotics originated with bacteria); hence probiotic rotation became part of my preaching.
If you have microbiome related issues, my soapbox has been “your goal is make the stable dysfunctional microbiome, unstable“. Today I read a study on Nature that further clarifies what may be needed.
Together, these findings suggest that the human gut microbiome’s metabolic potential reflects dietary exposures over preceding days and changes within hours of exposure to a novel nutrient. The dynamics of this ecological memory also highlight the potential for intra-individual microbiome variation to affect the design and interpretation of interventions involving the gut microbiome.
Made from white rice flour, organic amaranth flour (Source)
Made from chickpea flour, organic yellow lentil flour, organic red lentil flour, organic kale powder, organic spinach powder (source)
Every two weeks change dominant proteins source
Fish
Pork
Lamb
Duck
Chicken
Turkey
Change vegetables and fruit too…
Change main spices used….
The key aspect is that every new addition results in a change of the microbiome. If you have microbiome issues, that is what you want to do. You do NOT want to take the same supplements, herbs, spices, vitamin or comfort food – continuously. You want to shake things up!
While the location of SIBO and the results of a stool test are a good physical distance apart, as part of an ongoing series of posts on special studies, I ran SIBO thru this morning — really expecting to find nothing, or perhaps a few overgrowths. To my surprise, the results was significant undergrowth with high statistical significance!
A formal post on the results is in the queue, but I thought that I should make the results available for those who wish to commit SIBO heresy!
SIBO reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. A summary table of various studies has been added there which shows the statistical association is actually pretty strong.
Of course, you are saying “but this is down stream how can it impact upstream?!!?” To me, it is like saying, “I was waiting in a queue and everyone is healthy except the first person in the queue who had COVID, how could I get it?” Bacteria spread, including upstream.
I should point out that these bacteria may not be the cause, rather they may be ‘the canaries in the coal mine’ of the microbiome. A few items of special interest are:
Bacteria
Reference Mean
Study
Shuttleworthia (genus)
273
33
Prevotella stercorea (species)
6378
31
Coprococcus eutactus (species)
7987
1463
Streptococcaceae (family)
3567
1473
Veillonellaceae (family)
17295
10944
Tissierellales (order)
4162
1457
Peptoniphilaceae (family)
4158
1456
In terms of enzymes, the four items that were most significant were all low levels of:
SIBO by definition is OVERGROWTH, the stool samples and statistics says it is an UNDERGROWTH condition. Undergrowth means that substances do not get processed fully….
In the mean while, people with SIBO can try the new algorithm on the changing microbiome page. Feel free to add comments here on experience after trying suggestions for a month.
REMEMBER: The usual test is finding various compounds on the breath. Are the compounds from producing too much(overgrowth) or the compounds are not being consumed enough (undergrowth). Both scenarios produce a positive result. It feels like someone jumped to a conclusion and this arbitrary decision stuck in the name.
This is a common symptom for people that have uploaded. This is reported often in samples, and thus being examined if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links (A summary table of various studies has been added there).
Study Populations:
Symptom
Reference
Study
Bloating
1051
98
Bacteria Detected with z-score > 2.6: found 120 items, highest value was 5.4
Enzymes Detected with z-score > 2.6: found 298 items, highest value was 5.4
Compound Detected with z-score > 2.6: found ZERO items
The highest z-scores above are less than other symptoms with smaller study sizes. The likely cause is a more diverse study population
Interesting Significant Bacteria
All bacteria found significant had too low levels.
The dominant bacteria group seems to be Bifidobacterium, low Bifidobacterium . The latter we know little about. I should point out that these bacteria may not be the cause, rather they may be ‘the canaries in the coal mine’ of the microbiome. These studies’ methodology determines association and not causality.
“probiotics might improve the stool consistency of patients with IBS-C and increase the number of Bifidobacteria “
“B. animalis subspecies lactis supplementation may increase defecation frequency and, in short-term treatment, may reduce CTT in healthy adults and improve stool consistency in individuals without GIS. ” [2022]
“Bifidobacterium adolescentis were significantly less abundant in patients with Functional Abdominal Bloating/Distention, compared with healthy controls.” [2020]
REMEMBER: With your appropriate 16s sample, Dr. Artificial Intelligence on Microbiome Prescription will detail out foods, supplements et cetra to take (and to AVOID). If you do not have a sample, then review Bifidobacterium Summary Page.
This is a common symptom for both ME/CFS and Long COVID. This is reported often in samples, and thus being examined if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links.
Beyond the goal of identifying bacteria involved, I am curious on the intersection of the bacteria with ME/CFS and Long COVID – i.e. bacteria in common and not in common.
Study Populations:
Symptom
Reference
Study
Post-Exertional Malaise (PEM)
1086
62
Bacteria Detected with z-score > 2.6: found 181 items, highest value was 6.2
Enzymes Detected with z-score > 2.6: found 237 items, highest value was 7.0
Compound Detected with z-score > 2.6: found ZERO items
The highest z-scores above are less than other symptoms. There are two possible reasons:
Smaller Study Population
A more varied population in the study group.
Interesting Significant Bacteria
All bacteria found significant had too low levels.
We have two dominant bacteria group, both Bifidobacterium and Sporolactobacillus. The latter we know little about. I should point out that these bacteria may not be the cause, rather they may be ‘the canaries in the coal mine’ of the microbiome. These studies’ methodology determines association and not causality.
hydrogen-sulfide:ferredoxin oxidoreductase (1.8.7.1): This is connected to iron. The blood uses iron to carry oxygen, and thus an absence/low level could [speculation] result in an impact on the blood’s ability to deliver oxygen (thus fatigue).
D-fructose:ubiquinone 5-oxidoreductase (1.1.5.14): This is also connected to iron.
For those wishing to explore more, you may wish to read Oxidoreductase
It does hint at an experiment to try: After exercise, try a dosage of Ubiquinol to see if it influences things.
Common Bacteria Shifts Observed in ME/CFS
We have 45 bacteria in common, they are listed below. A LOT of them are bifidobacterium, and no lactobacillus. This implies that bifidobacterium probiotics may be a good choice for ME/CFS with PEM
Tax_Name
Tax_rank
Thiorhodococcus pfennigii
species
Candidatus Tammella caduceiae
species
Veillonella atypica
species
Tammella
genus
Myxococcales
order
Gemella cuniculi
species
Bifidobacterium catenulatum
species
Nannocystineae
suborder
Olivibacter
genus
Campylobacterales
order
Epsilonproteobacteria
class
Campylobacteraceae
family
Pedobacter kwangyangensis
species
Haemophilus parainfluenzae
species
Haemophilus
genus
Clostridium aestuarii
species
Sterolibacteriaceae
family
Lactococcus fujiensis
species
Bifidobacterium bifidum
species
Atopobium
genus
Balneola
genus
Balneola vulgaris
species
Balneolaceae
family
Thiobacillus
genus
Pigmentiphaga
genus
Thiobacillaceae
family
Balneolia
class
Balneolales
order
Balneolaeota
phylum
Ruminococcus flavefaciens
species
Hydrogenophilalia
class
Hydrogenophilales
order
Hydrogenophilaceae
family
Atopobiaceae
family
Veillonella dispar
species
Veillonella
genus
Clostridium chartatabidum
species
Actinobacillus pleuropneumoniae
species
Sporolactobacillaceae
family
Sporolactobacillus putidus
species
Sporolactobacillus
genus
Bifidobacterium kashiwanohense PV20-2
strain
Bifidobacterium catenulatum subsp. kashiwanohense
subspecies
Bifidobacterium gallicum
species
Bifidobacterium cuniculi
species
Bacteria COMMON to ME/CFS and PEM
Common Bacteria Shifts Observed in Long COVID
We have 42 bacteria in common, they are listed below. We notice some interesting difference from above:
Lactobacillus at the genus level as well as the retail probiotic Lactiplantibacillus plantarum (AKA Lactobacillus plantarum)
Bifidobacterium is still there, but one of them is available as a retail probiotics.
Bifidobacterium animalis
Tax_Name
Tax_rank
Paenibacillus
genus
Veillonella
genus
Actinomycetaceae
family
Lactiplantibacillus plantarum
species
Actinomyces
genus
Flammeovirga
genus
Flammeovirga pacifica
species
Flammeovirgaceae
family
Lactiplantibacillus
genus
Phocaeicola massiliensis
species
Prosthecobacter
genus
Fusobacterium gonidiaformans
species
Candidatus Tammella caduceiae
species
Gammaproteobacteria
class
Tammella
genus
Coriobacteriaceae
family
Fusobacteria
phylum
Fusobacteriia
class
Fusobacteriales
order
Coriobacteriales
order
Bifidobacterium thermophilum
species
Dolichospermum curvum
species
Blautia wexlerae
species
Atopobiaceae
family
Actinobacillus pleuropneumoniae
species
Eggerthella lenta
species
Fusobacteriaceae
family
Atopobium
genus
Schaalia
genus
Bifidobacterium gallicum
species
Eggerthella
genus
Bifidobacterium animalis
species
Aerococcaceae
family
Coriobacteriia
class
Bifidobacterium cuniculi
species
Schaalia naturae
species
Phocaeicola sartorii
species
Leptospira licerasiae
species
Leptospiraceae
family
Leptospira
genus
Leptospirales
order
Alkalibacterium
genus
Bottom Line
There appear to be differences between ME/CFS with PEM and Long COVID with PEM. The main difference is with Long COVID: Lactobacillus probiotics is a suggestion; for ME/CFS it is not.
Remember suggestions that are specific to your unique microbiome are available on the Microbiome Prescription web site.
Tinnitus is not usually viewed as a microbiome issue. It was worth checking if it reaches our threshold for inclusion as defined in A new specialized selection of suggestions links. It did, hence this post
Study Populations:
Symptom
Reference
Study
Neurological-Audio:Tinnitus (ringing in ear)
1075
73
Bacteria Detected with z-score > 2.6: found 129 items, highest value was 6.3
Enzymes Detected with z-score > 2.6: found 493 items, highest value was 7.1
Compound Detected with z-score > 2.6: found ZERO items
Bifidobacterium probiotics: we have 4 in the top group. Unfortunately none of these species are available at the retail level (that I am aware of). Checking interactions for these 4, there was no significant interactions found with common retail bifidobacterium species, just with the general genus.
As above, all levels that were found significant had too little. I will leave it to the reader to go to Kyoto Encyclopedia of Genes and Genomes to learn about these enzymes (a steep learning curve).
gamma-L-glutamyl-L-cysteinyl-glycine:spermidine ligase (ADP-forming) [spermidine is numbered so that atom N-1 is in the amino group of the aminopropyl part of the molecule] (6.3.1.8)
Nothing was found again!!!! In one sense this was a surprise, in another sense, it hints that the results found significant are not random.
Bottom Line
This was an interesting analysis because the dominant deficiencies were in genus that are available as probiotics. In reality, it points to just 3 probiotics: an E.Coli probiotic and two bifidobacterium. Food suggestions will be generated on Microbiome Prescription using an individual’s unique microbiome.
Social Feedback
I went and looked at these two in combinations and got a lot of bacteria in common
Bacteria Detected with z-score > 2.6: found 143 items, highest value was 8.5
Enzymes Detected with z-score > 2.6: found 215 items, highest value was 6.1
Compound Detected with z-score > 2.6: found ZERO items
Interesting Significant Bacteria
One of the top items happens to have probiotics that are known to take up residency – are Symbioflor-2 and Mutaflor. . All of these top items are too low levels
Nothing was found!!!! In one sense this was a surprise, in another sense, it hints that the results found significant are not random.
Bottom Line
Histamine or Mast Cell Issues appears to a condition of deficiency. Common internet thinking is that it is a condition of a surplus of histamine producing bacteria. It is more likely that the normal histamine consumers are being starvedof enzymes that are needed to stop the accumulation of histamine.
The z-scores for bacteria are lower than Long COVID which reflect the diffusion of bacteria over time. Trying to tackle at the compound or enzyme levels becomes excessively complex. Working at the bacteria level appears viable, but do not expect as many bacteria matches to appear.
Going to [Changing Microbiome] page you will see a new box appearing. It may or may not contains links to suggestions. It is scoped to BiomeSight interpretation of microbiome data. If you have Ombre/Thryve samples, do not despair, you can move your raw data to BiomeSight and send it to Microbiome Prescription. It is simple and fast as shown in this video 📹Video on Transferring Data from Ombre/Thryve to Biomrsight and then to Microbiome Prescription.
Actually, if you have done any of those listed below — you can follow the same process
What will appear in this list?
First, why Biomesight? — the reason is very simple, there are more samples (20% more at present and increasing then Ombre despite being in business for much less time). The bigger the sample size, the easier it is to find significant shifts.
The criteria is that there must be strong statistical significance for a good number of bacteria. My current threshold is: z-scores must all be 2.6 or higher, that is p <0.01 or 99% confidence. At least 50 bacteria needs to be identified as significant. I will be going thru the symptom list from most frequency reported to less frequently reported. Make sure that you annotate your samples with your symptoms.
NOTE: The numbers below reflect the statistics when various posts were done. The numbers are recomputed at least bi-weekly.
The bigger the z-score (positive or negative), the more significant
I should point out that these bacteria may not be the cause, rather they may be ‘the canaries in the coal mine’ of the microbiome. These studies’ methodology determines association and not causality.
An additional criteria is that they need to be clear abnormalities with the KEGG Enzymes estimate.
I have done two with acceptable results and have made them available. More will be added over time (each one takes a fair amount of time).
US National Library of Medicine Studies are difficult to use
I have their results available and on the site, including as bacteria filters. The problem is that all of those results are very sensitive to the lab being used and the software processing the results. See The taxonomy nightmare before Christmas… for background. In the absence of better information, they are the best we have — until now. With these suggestions, the lab and the software being used are the same and also the one that your results are done by.
Interesting Observation
For both of the above, lower levels of a large number of bacteria was the common pattern. These bacteria are not present in all samples, and most of the studies seen on the US National Library of Medicine look only at bacteria found in all samples. That approach will exclude the bacteria that we find are significant.
A second item to be aware of is that often those PubMed studies may consist of just 50 people (control and patients). In our analysis, we have 1200+ people with often more then 120 people with a specific condition. Statistically, we are more likely to detect more associations than those studies. It’s a number’s game.
I am still tuning the suggestions engine, so expect reordering of suggestions occasionally. The suggestions pass the reasonableness test.
Quick Cross Validation
I ran the suggestions only for prescription drugs for ME/CFS and the top four suggestions are listed below. Three of the top four are used by ME/CFS physicians such as Dr. Cecile Jadin [Src], Philippe Bottero [src], G. L. Nicolson, M. Y. Nasralla, A. R. Franco, K. De Meirleir, N. L. Nicolson, R. Ngwenya & J. Haier [Src]. These physicians all report various degree of success.
azithromycin,(antibiotic)s
atorvastatin (prescription)
minocycline (antibiotic)s
doxycycline (antibiotic)s
Atorvastatin is an oddity with no studies for its use with ME/CFS. On the flip side, generic statins were high on the avoid list. It would be nice if someone did a clinical trail for this explicit type of statin.
A reader pinged me about new results so I thought it would be good to look at his series of 4 samples from Ombre/Thryve to help people interpret their own results. I am using the tool described in Comparing Samples – Update.
As you can see, the bacteria count do bounce around, with the change from the last sample being a definite improvement
There is a ongoing shift towards overproduction of Enzymes, but this does not cascade into increased compound (Produced – Consumed)
Several of the External Criteria measures showed improvement (2 less) and less showed more (1 more)
The microbiome is a dynamic system and as shown in the image below, it is not a simple straight path. There is no single measure that indicate the current status of the gut, rather a variety of measures.
The latest sample showed (compare to prior):
Drop in compound extremes
Increase in high Enzyme production — this may hint at the body stocking up on supplies for an attack of troublesome bacteria
A drop of bacteria out of range by both external criteria and Microbiome Prescription Criteria.
Reader’s Comments
This round I restarted the b.lactis by custom probiotics. I did it last test and took a month break and it came up again. 1st round doing it , the herx was kicking my butt and I lasted 10 days. This time around , only did 1 scoop at night and I was still herxing and sleeping 8 hours hard. After day 10 I did half dose mornings and half at mid day or night. Much better tolerance and body seems to respond well. Calmer, better sleep, allergies seem better.
Now also taking rice bran, blue berries, Lactoferrin for iron, and a lot of the other high priority items. I try playing basketball 1 time a week to check my progress. I have plenty of energy to play, but if I play too hard I crash still exactly 5 hours later (impaired sleep, brain fog, fatigue , headache, inflamed stomach lasting exactly 24 hours) . Then back to normal the same day. Tried the symptom /handpick bacteria tool and focused on picking PEM/brain fog/sleep with associations. The top modifier for that was glycemic (licorice root tincture). Added that to the mix this week. Still crash, but notice some weight loss and less belly inflammation after I play. Overall improvement in sleep and anxiety. I am still a work in progress and could be better, but seeing improvement. My post exertion crashes are the thorn in my side I haven’t been able to dent much, no clue why I crash exactly 5 hours after every time but can play and feel find right after. Couple more days of b lactis and I switch to my next probiotic L.gasseri. Then will test again in about 2 weeks!
Mystery of Crash after 5 hours lasting exactly 24 hours
I have a few suggestions to try — looking at possible mechanisms..
At 3 hrs (2 hrs) before crash — take 3 regular aspirins
What this is attempting to test is whether coagulation is a factor
A reader asked me to compared her latest sample to her prior samples. Comparing samples can be time consuming and complex, so I revisited my past comparisons posts and created a table that saves time and add clarity.
Bacteria Report
Lab Read Quality indicates the numbers of bacteria that the lab obtained. In this case, the quality of the samples were similar
This impacts Bacteria Reported. In this case 1/14 less, and the bacteria count is about 1/14 less. If there was a severe change, then we wish to understand why
Bacteria Over/Under: Between samples we would like them to be reduced (less extremes). If lab quality goes down or up a lot, then the numbers may need to be adjusted for comparison.
We see more than a 1/14 drop, which indicates a better microbiome.
Lab Relatives
If both labs were done with the same provider, then this section will appear. In this case we see that rarely seen bacteria goes down – a good sign.
Foreign Criteria
This applies various foreign criteria to the samples. The increase of pathogen is a potential concern. I tend not to focus on pathogens because the source data usually report “higher” or “lower” only.
Microbiome Prescription Criteria
This reports on the main approaches used for generating consensus suggestions. All of them indicate little change except KM which suggests improvement.
PubMed Conditions
This uses the literature from the US National Library of Medicine to estimate likelihood of having various conditions. It is a fuzzy estimate due to the what the studies report. In this case we see good improvement across the board.
KEGG Criteria
As above, we want to reduce extreme values (represented by high and low percentiles). What we see below is very significant improvement across the board.
Bottom Line
The above are the list of my first go-to items comparing samples. If someone has followed the suggestions from Dr.A.I. between samples (or done other things), it give a quantification of the changes that occur across multiple dimensions.
In this case, we see significant improvement over a few months! It should be noted that the various external criteria show no apparent changes, the deeper dive that Microbiome Prescription does show significant positive changes.
Recent Comments