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.

Bacteria (Rank)	Reference Mean	Study Mean	z-score
Sporolactobacillus (genus)	174	60	6.2
Sporolactobacillus putidus (species)	174	60	6.2
Sporolactobacillaceae (family)	173	60	6.2
Bifidobacterium cuniculi (species)	81	24	5.9
Bifidobacterium asteroides (species)	58	23	5.9
[Ruminococcus] gnavus (species)	7421	3336	5.4
Mediterraneibacter (genus)	7870	3717	5.3

Interesting Enzymes

All enzymes found significant had too low levels.

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).

There are some items of special interest appearing which I drill into below.

Enzyme	Z-Score
hydrogen-sulfide:ferredoxin oxidoreductase (1.8.7.1)	7
D-fructose:ubiquinone 5-oxidoreductase (1.1.5.14)	6.2
D-fructosyl-L-lysine 3-epimerase (5.1.3.41)	6.1
L-tryptophan carboxy-lyase (4.1.1.105)	6
aromatic-L-amino-acid carboxy-lyase (4.1.1.28)	6
CTP:N-acylneuraminate cytidylyltransferase (2.7.7.43)	5.9
protein-Npi-phospho-L-histidine:L-ascorbate Npi-phosphotransferase (2.7.1.194)	5.8
propane-1,2-diol hydro-lyase (propanal-forming) (4.2.1.28)	5.7
N-methylhydantoin amidohydrolase (ATP-hydrolysing) (3.5.2.14)	5.5
D-ribopyranose furanomutase (5.4.99.62)	5.5
3-dehydro-L-gulonate:NAD(P)+ 2-oxidoreductase (1.1.1.130)	5.5

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

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

This is similar to Special Study: Histamine or Mast Cell Issues in finding no compounds, but the bacteria factor appears weaker and the enzymes is more.

Interesting Significant Bacteria

We have two dominant items that may be addressed by probiotics: Low Bifidobacterium and Low E.Coli

• E.Coli probiotics are Symbioflor-2 and Mutaflor.
• 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.

Looking at Lactobacillus taiwanensis, there is a solid positive association with Bifidobacterium cuniculi and a weaker association with Bifidobacterium catenulatum subsp. kashiwanohense. There is also a weak association with two retail probiotic species: Bifidobacterium bifidum and Bifidobacterium animalis. It was interesting to note that there was no associations with any retail Lactobacillus species.

Bottom line appears to become: E.Coli probiotics, Bifidobacterium bifidum and Bifidobacterium animalis

All of the significant bacteria has too low levels.

Bacteria	Reference Mean	Study	Z-Score
Bifidobacterium gallicum (species)	3754	534	6.3
Prevotella stercorea (species)	6614	102	6
Bifidobacterium subtile (species)	82	32	5.3
Escherichia coli (species)	716	170	5.3
Lactobacillus taiwanensis (species)	85	12	5.3
Catenibacterium mitsuokai (species)	432	34	5.2
Bifidobacterium cuniculi (species)	81	28	5.1
Enterobacteriaceae (family)	8948	2622	5
Bifidobacterium catenulatum subsp. kashiwanohense (subspecies)	313	71	5

Interesting Enzymes

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).

Enzyme	Reference Mean	Study Mean	Z-Score
[cysteine desulfurase]-S-sulfanyl-L-cysteine:[molybdopterin-synthase sulfur-carrier protein]-Gly-Gly sulfurtransferase (2.8.1.11)	5326	2043	7.1
gamma-L-glutamyl-L-cysteinyl-glycine:spermidine amidase (3.5.1.78)	3212	1066	6.3
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)	3212	1066	6.3
tRNA-uridine13 uracil mutase (5.4.99.27)	3671	1246	6.3
donor:hydrogen-peroxide oxidoreductase (1.11.1.21)	3052	869	6.2
S-adenosyl-L-methionine:tRNA 5-(aminomethyl)-2-thiouridylate N-methyltransferase (2.1.1.61)	3686	1305	6.1
5-oxo-L-proline amidohydrolase (ATP-hydrolysing) (3.5.2.9)	10180	3505	6.1
thioredoxin:protein disulfide oxidoreductase (dithiol-forming) (1.8.4.16)	3812	1301	6.1
acetyl-CoA:N6-hydroxy-L-lysine 6-acetyltransferase (2.3.1.102)	719	143	6.1
UDP-alpha-D-glucose:enterobactin 5′-C-beta-D-glucosyltransferase (configuration-inverting) (2.4.1.369)	819	141	6.1
S-adenosyl-L-methionine:tRNA (uracil54-C5)-methyltransferase (2.1.1.35)	3817	1273	6.1
L-methionine:2-oxo-acid aminotransferase (2.6.1.88)	2771	815	6.1

Interesting Compounds

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 Name	Taxonomy rank
Nostoc	genus
Citrobacter	genus
Anaerococcus lactolyticus	species
Streptomycetales	order
Negativicoccus succinicivorans	species
Sutterella stercoricanis	species
Streptomycetaceae	family
Negativicoccus	genus
Lactococcus fujiensis	species
Actinomycetaceae	family
Prevotella disiens	species
Prosthecobacter	genus
Clostridium chartatabidum	species
Olivibacter	genus
Peptoniphilus asaccharolyticus	species
Butyricimonas synergistica	species
Campylobacterales	order
Epsilonproteobacteria	class
Campylobacteraceae	family
Pedobacter kwangyangensis	species
Staphylococcus pseudolugdunensis	species
Enterobacterales	order
Streptococcus milleri	species
Schaalia	genus
Enterobacteriaceae	family
Mitsuokella	genus
Gammaproteobacteria	class
Prevotellaceae	family
Clostridium cellulovorans	species
Schaalia naturae	species
Enterobacteriaceae incertae sedis	norank
Blochmannia	genus
Bulleidia	genus
Bifidobacterium cuniculi	species
Prevotella	genus
Escherichia	genus
Prevotella copri	species
Bifidobacterium gallicum	species
Escherichia coli	species
Prevotella stercorea	species
Prevotella paludivivens	species

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.

Study Populations:

Symptom	Reference	Study
Histamine or Mast Cell Issues	1092	56

• 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

Bacteria	Reference Mean	Study	Z-Score
Prevotella paludivivens (species)	140	21	8.5
Prevotella stercorea (species)	6451	45	6.1
Escherichia albertii (species)	912	232	5.1
Serratia (genus)	1011	275	4.9
Serratia entomophila (species)	987	263	4.8
Alishewanella (genus)	35	19	4.7
Clostridium cellulovorans (species)	40	17	4.6
Yersiniaceae (family)	1032	317	4.6
Prevotellaceae (family)	81953	28733	4.5
Prevotella copri (species)	65645	13994	4.5
Escherichia (genus)	5617	1664	4.5
Prevotella (genus)	73889	22771	4.5
Staphylococcus pseudolugdunensis (species)	45	20	4.3
Escherichia coli (species)	712	234	4.3
Schaalia naturae (species)	211	37	4.3
Atopobiaceae (family)	132	39	4.2
Rhodovibrionaceae (family)	119	61	4.1
Bulleidia (genus)	188	30	4.1

Interesting Enzymes

As above, too low levels were most significant

Enzyme	Reference Mean	Study Mean	Z-Score
propanoyl-CoA:oxaloacetate C-propanoyltransferase (thioester-hydrolysing, 1-carboxyethyl-forming) (2.3.3.5)	1443	380	6.1
(2S,3R)-3-hydroxybutane-1,2,3-tricarboxylate pyruvate-lyase (succinate-forming) (4.1.3.30)	1395	383	5.8
S-methyl-5′-thioadenosine:phosphate S-methyl-5-thio-alpha-D-ribosyl-transferase (2.4.2.28)	3654	1517	5.4
UDP-N-acetyl-alpha-D-glucosamine:lipopolysaccharide N-acetyl-D-glucosaminyltransferase (2.4.1.56)	1008	264	5.2
(2S,3S)-2-hydroxybutane-1,2,3-tricarboxylate hydro-lyase [(Z)-but-2-ene-1,2,3-tricarboxylate-forming] (4.2.1.79)	1246	401	5.2
n/a (3.4.23.49)	1244	356	5.1
L-carnitinyl-CoA hydro-lyase [(E)-4-(trimethylammonio)but-2-enoyl-CoA-forming] (4.2.1.149)	1380	367	5
acyl-CoA,ferrocytochrome b5:oxygen oxidoreductase (6,7 cis-dehydrogenating) (1.14.19.3)	1013	302	5

Interesting Compounds

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 starved of enzymes that are needed to stop the accumulation of histamine.

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.

Study Populations:

Symptom	Reference	Study
Neurocognitive:Brain Fog	1020	127

• Bacteria Detected with z-score > 2.6: found 97 items, highest value was 5.3
• Enzymes Detected with z-score > 2.6: found 309 items, highest value was 5.6
• Compound Detected with z-score > 2.6: found 571 items, highest value was 5.8

Interesting Significant Bacteria

The top item happens to have probiotics that are known to take up residency – Symbioflor-2 and Mutaflor. All of these top items are too low levels

Bacteria	Reference Mean	Study	Z-Score
Escherichia coli (species)	743	168	5.3
Lactiplantibacillus pentosus (species)	123	25	5.1
Shuttleworthia (genus)	288	95	5.1
Escherichia (genus)	5903	1500	4.5
Veillonella (genus)	4072	2522	4.4
Veillonella dispar (species)	853	127	4.4
Staphylococcus pseudolugdunensis (species)	46	20	4.2
Clostridium cellulovorans (species)	40	18	4.1

Interesting Enzymes

NADH is available as a supplement. It is reported to improve ME/CFS

Enzyme	Reference Mean	Study Mean	Z-Score
3-phenylpropanoate,NADH:oxygen oxidoreductase (2,3-hydroxylating) (1.14.12.19)	611	132	5.6
3-(cis-5,6-dihydroxycyclohexa-1,3-dien-1-yl)propanoate:NAD+ oxidoreductase (1.3.1.87)	598	131	5.6
propanoyl-CoA:oxaloacetate C-propanoyltransferase (thioester-hydrolysing, 1-carboxyethyl-forming) (2.3.3.5)	1514	443	5.2
(2S,3R)-3-hydroxybutane-1,2,3-tricarboxylate pyruvate-lyase (succinate-forming) (4.1.3.30)	1467	418	5.2
[cysteine desulfurase]-S-sulfanyl-L-cysteine:[molybdopterin-synthase sulfur-carrier protein]-Gly-Gly sulfurtransferase (2.8.1.11)	5421	2662	5.1
ATP:D-tagatose 6-phosphotransferase (2.7.1.101)	422	111	5

Interesting Compounds

All of the top compounds had lower levels

Names	Z-Score
2-Phospho-4-(cytidine 5′-diphospho)-2-C-methyl-D-erythritol (C11436)	5.8
O-Acetyl-L-homoserine (C01077)	5.4
Adenosine-GDP-cobinamide (C06510)	5.4
6-Deoxy-L-galactose (C01019)	5.2
[Enzyme]-cysteine (C15811)	5.2
Adenosyl cobyrinate a,c diamide (C06506)	5.1
Xanthosine 5′-phosphate (C00655)	5.1
dTDP-glucose (C00842)	5.1

Bottom Line

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.

Study	Size	Bacteria Z-Score	Enzyme Z-Score	Compound Z-score
Long COVID	157	10.6	-12	-22.7
General ME/CFS	159	6.6	4.5	3.1
ME/CFS with IBS	52	8.4	6.3	n/a
Tinnitus (ringing in ear)	73	6.3	7.1	n/a
Histamine or Mast Cell Issues	56	8.5	6.1	n/a
Neurocognitive: Brain Fog	127	5.3	5.6	5.8
Neurocognitive: Can only focus on one thing at a time	79	5.8	6.7	n/a
Neurocognitive: Difficulty paying attention for along period of time	75	5.2	5.0	n/a
Neurocognitive: Problems remembering things	69	6.3	5.9	n/a
Difficulty finding the right word	67	7.0	5.5	n/a
Post-Exertional Malaise (PEM)	62	6.2	7.0	n/a
Bloating	98	5.4	5.4	n/a
Constipation	83	9.9	5.2	n/a
Unrefreshing Sleep	107	5.5	5.2	n/a
General: Fatigue	130	5.2	5.7	3.9
Worsening of Symptoms with Stress.	92	6.9	5.1	n/a
Irritable Bowel Syndrome	55	6.7	6.1	n/a
Poor Gut Motility	55	8.8	6.4	n/a
Autism	67	8.2	7.6	n/a
Easily Irritated	53	9.9	5.9	n/a
Anxiety/Tension	59	6.4	8.4	n/a
Allergic Rhinitis (Hay Fever)	42	8.6	6.8	n/a
Hypersensitivity to Noise	56	6.9	8.1	n/a
Intolerance of Extremes of Heat and Cold	54	8.8	6.7	n/a
Cold Extremities	78	12.6	5.3	n/a
Small intestinal bacterial overgrowth (SIBO)	35	6.5	7.9	n/a
Depression	53	7.8	8.3	n/a
Allergies And Food Sensitivity	73	9.9	6.8	n/a
Alcohol intolerance + Medication sensitivities	59	8.6	7.9	n/a

• 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.

Enzyme (KEGG Identifier)	Reference Mean	Long COVID Mean	Z-Score
S-adenosyl-L-methionine:16S rRNA (guanine1516-N2)-methyltransferase (2.1.1.242)	17373	10370	7.1
ATP phosphohydrolase (ABC-type, teichoic-acid-exporting) (7.5.2.4)	12291	7250	6.3
(S,S)-butane-2,3-diol:NAD+ oxidoreductase (1.1.1.76)	9666	5525	6.3
(S)-acetoin:NAD+ oxidoreductase (1.1.1.304)	9666	5525	6.3
UDP-N-acetyl-alpha-D-glucosamine:beta-D-mannosyl-glycoprotein 4-beta-N-acetyl-D-glucosaminyltransferase (configuration-inverting) (2.4.1.144)	837	240	6
acetyl-CoA:propanoate CoA-transferase (2.8.3.1)	9772	5569	6

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 reader has prior reviews:

• A review of a ME/CFS Microbiome Jan 2021
• Evidence of ME/CFS improving using Microbiome Data May 2022

Measure	6/15/2022	5/16/2022	3/15/2022	11/21/2021
Lab Read Quality	4.8	4.8	3.6	4.2
Bacteria Reported By Lab	515	669	593	473
Bacteria Over 99%ile	0	1	0	1
Bacteria Over 95%ile	4	22	21	9
Bacteria Over 90%ile	22	40	48	21
Bacteria Under 10%ile	43	112	23	87
Bacteria Under 5%ile	18	60	5	46
Bacteria Under 1%ile	0	9	0	2
Lab: Thryve				NULL
Rarely Seen 1%	3	20	9	0
Rarely Seen 5%	11	67	38	9
Pathogens	37	42	38	33
Outside Range from JasonH	7	7	9	9
Outside Range from Medivere	17	17	20	20
Outside Range from Metagenomics	8	8	7	7
Outside Range from MyBioma	8	8	10	10
Outside Range from Nirvana/CosmosId	23	23	22	22
Outside Range from XenoGene	8	8	10	10
Outside Lab Range (+/- 1.96SD)	2	7	6	2
Outside Box-Plot-Whiskers	42	71	72	25
Outside Kaltoft-Moldrup	113	148	143	91
Condition Est. Over 99%ile	0	0	0	0
Condition Est. Over 95%ile	0	0	0	0
Condition Est. Over 90%ile	0	0	1	0
Enzymes Over 99%ile	91	48	53	5
Enzymes Over 95%ile	599	305	182	147
Enzymes Over 90%ile	719	629	438	458
Enzymes Under 10%ile	70	71	38	169
Enzymes Under 5%ile	19	28	4	78
Enzymes Under 1%ile	2	0	0	0
Compounds Over 99%ile	6	93	143	19
Compounds Over 95%ile	186	277	190	89
Compounds Over 90%ile	303	435	220	118
Compounds Under 10%ile	321	528	333	133
Compounds Under 5%ile	195	378	310	100
Compounds Under 1%ile	0	118	267	42

Bottom Line

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
• At 3 hours, take a large dosage of Benfotiamine – an interesting supplement or Thiamine
  • What this is attempting to test is whether lactic acid build up is a factor
• At 3 hours, Huperzine A
  • What this is attempting to test is if acetylcholinesterase build up is a factor.
    See Non-cholinergic effects of huperzine A: beyond inhibition of acetylcholinesterase
• At 3 hours, an antihistamine
  • What this is attempting to test is if histamine build up is a factor.

The goal is to see if any alters the onset at 5 hrs, thus a “tell” for possible causality.

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.

Foreword – and Reminder

I am not a licensed medical professional and there are strict laws where I live about “appearing to practice medicine”.  I am safe when it is “academic models” and I keep to the language of science, especially statistics. I am not safe when the explanations have possible overtones of advising a patient instead of presenting data to be evaluated by a medical professional before implementing.

I cannot tell people what they should take or not take. I can inform people items that have better odds of improving their microbiome as a results on numeric calculations. I am a trained experienced statistician with appropriate degrees and professional memberships. All suggestions should be reviewed by your medical professional before starting.

Dr. A.I. is not a person ,it stands for Artificial Intelligence – a compute program that takes facts (in this case, well over 10 millions) and use logic to come up with suggestions based on probability.

Back Story

I’ve had severe gastric symptoms for about 15 years (unstoppable daily acid reflux, severe bloating and food sensitivities/very reduced diet, upper digestive pain, chronic UTI, chronic tender lymph nodes), and milder related symptoms my whole life (recurrent sore throat, allergies, eczema, psoriasis, vitiligo, depression, anxiety, fatigue ). Since I haven’t been able to fully resolve them, I’ve also developed comorbidities (Hashimoto’s, thyroid microtumor). I’ve found your website and have gotten a bit better and have hope to keep improving. I’m trying to understand your videos and instructions on how to use your AI the best I can.

I did a number of Ubiome tests in its day, unfortunately I only downloaded the raw data for two of them, which I uploaded to your website. I am attaching other gastro tests that I believe I cannot upload since there’s no raw data, they may be useless. I also have uploaded my recent  thryve test, and I plan on having more tests done to monitor progress. I have been going back and forth with microbiome labs to try to get the raw data, we’ll see. I’m “hooked” to herbals, taking bidens for UTI for years, and tied to foods like cabbage and meat, and my microbiome shows signs that it needs a wider variety of foods as well as a rotation of probiotics, herbal antibiotics, and foods. I react very strongly to prebiotics. I can eat cruciferous veggies, meats, fish, low-carb, low oxalate NO: starches, grains, cereals, fruits, sugars, eggs, milk, legumes. Neem helped a bit with oxalate sensitivities. A few probiotics helped with mood. Then I get stuck in needing these things, the flora develops resistance.

My husband and I are both following your sites. Thank you again for all you do.

Addendum

Sending the first draft results in the following information, see Addendum Discussion at the bottom.

I am thankful you have taken my sample to analyze. This is an incredible amount of work and very illuminating on how you approach the tools on your website. I agree, my situation seems to be a bit contradictory and difficult to approach. Here are some comments and reactions to your ideas—I am not suggesting these should be included or modify your post, that is up to you…

Firstly, I would like to say, that I have already been implementing a few of the top probiotic and flavonoid suggestions for a couple of weeks and I do feel that many of my symptoms have either improved some or changed and are changing, which, for at least a decade seemed unmovable. Other are unchanged. I do not think I mentioned this before, but in general I have less hepatic region/upper digestive discomfort and when I do have some, it’s less aggressive and I recover faster. I have related this symptom in the past to oxalate intolerance. My UTI feels much better, though it is not fully gone and my reflux has more frequently better days, though these come and go. My mood has improved and I tend to be more hopeful. I’m a bit less fatigued.

The things I have so far implemented are:

• Before I implemented any of your website suggestions, I used bidens and neem which helped a lot.
• When I took the Ombre sample I had been doing high doses of reuteri, plantarum, longum, casei for a couple of weeks… I wonder if these have anything to do with the high Kegg.
• I have been taking P. acidilactici as a base…
• I did 2 courses of Symbioflor 2 (2 bottles) and Enterogermina, which I think helped. I think some results suggest I may be low in good strep too.
• I am now on a run of B. coagulans and Equilibrium (this last one because I test low in alpha diversity)
• I am planning a run of L. sakei and Enterogermina again next…  The prescript assist folks refuse to share an updated CoA with me.
• I am taking artemisia, planning to alternate with neem, bidens, cryptolepis/sida

Regarding dietary recommendations, these are mostly non-negotiable at this point due to intolerances and allergies. One single pistachio a week ago gave me more bloating and worse symptoms for a couple days, so it is unlikely that nuts will be part of the solution initially. Rice bran is a big question, I have been searching for a brand that has low heavy metal analysis, but generally, I do not tolerate prebiotic fibers well at all. This is subject to change, for instance neem did help minimize a bit oxalate intolerance.

I have been using quercetin and I am waiting for a full-spectrum cannabinoid extract to arrive to try. I have had a complicated relationship with licorice in the past, but I am willing to try again. 

I do not think I can do oats yet, starches and cereals are too challenging.

The foods I eat that contain choline are salmon (once a week), beef daily and cruciferous vegetables which are the bulk of the veggies I eat because I tolerate them well, probably because of low oxalate content.

I think some killing is necessary along with the rest. The Mircobiome FX people recommended their B. subilis strain for this purpose (in their test C. difficile and B. fragilis were high) but it’s not recommended by the AI.

I will follow up with another Ombre test soon.

Analysis

Comparing results from different labs has risks. In this case, I thought it may be worth it. (I was debating about purging ubiome data since the data is stale and they consume a lot of space — I will keep them available).

I should note that I recently revised the code to use lab specific percentiles for ubiome, ombre and biomesight. This means that the numbers are more accurate despite the oddities of different labs.

The ubiome samples had a over representation of bacteria that had extreme values (at 90% or higher). The Ombre sample had a better balance, but still with some overrepresentation of bacteria with extreme value.

Dr. Jason Hawrelak

Jason’s criteria was developed using uBiome data. And all of the samples are good. There were no significant potential medical condition on any samples. The same number of bacteria deemed unhealthy were seen on each report.

• 05-08 – 98.8%ile
• 06-29 – 95.6%ile
• 2022 – 95.6%ile

The new Over and Under Population Experimental Feature

With this feature, we see much more difference between samples. The latest sample suggests that the issue may be related to too much being produced.

The Road Forward

There are a lot of health issues. The site supports targeting of specific issues (because that is how some people wish to proceed). In general, I have been finding the consensus with perhaps 1 or 2 specialized set of suggestions appearing to give good results. We are dealing with fuzzy data everywhere and should resist temptations to over engineer suggestions.

I looked at the Taxon Hierarchy and selected items in Ombre. There was something odd with the data as shown below.

The work around is to keep choices to the family, genus and species level. Picking items with a count of 2000 or more only. The selection is shown below

The result of the hand picked is shown below, I usually truncate the list to 0.4 and higher. Both lists are relatively small!

Doing the pro forma consensus building next

•  Use JasonH (15 Criteria) – 7 bacteria picked
•  Standard Lab Ranges (+/- 2 Std Dev) – 8 bacteria picked
•  Box Plot Whisker – 40 bacteria picked
•  Kaltoft-Moltrup Normal Ranges – 86 bacteria picked
•  Percentile in top or bottom 10 % – 83 bacteria picked

The consensus Report download is below.

The top items on the consensus report are similar to the hand picked results (in decreasing priority)

• rice bran
• cannabinoids
• quercetin
• glycyrrhizic acid (licorice)
• Far infrared Sauna – I still do 45 minutes at least once a week
• cellulose (prebiotic)

The avoid list:

• Slippery Elm
• aspartame (sweetner)
• low-fat diets
• garlic (allium sativum)
• lauric acid(fatty acid in coconut oil,in palm kernel oil,) a.k.a. Monolaurin
• refined wheat breads

Looking at Grains

• Take: Oats (porridge each morning?)
• Avoid: Barley, Rye, Wheat (a bit fuzzy, different types are on either side)

Probiotics

The top ones selected using KEGG may be hard to find in many places: enterogermina, Sun Wave Pharma / Spor Sun. The weight give them is light, so likely not much impact would be expected.

The top ones appears to be:

• bacillus coagulans (probiotics)
• lactobacillus rhamnosus (probiotics)

Again, they appear to be light in potential impact. With too high production of many things found above, probiotics may be adding wood to the fire.

Vitamins

Vitamin b7 biotin (supplement) and Vitamin C (ascorbic acid). A B-Complex has both good and bad items, so be specific to B7.

Bottom Line

This is a sample which I term “on the border of what we have knowledge about”. We can see the issues, but lack studies that we can use to correct the issues. I have encountered one like this before and after doing the first set of suggestions, the microbiome moved into a state where we are able to get a normal number of significant suggestions from. I hope this is the case here. Samples like this are a bit frustrating to me..

• I would suggest stocking up on Spezzatina (perhaps even do what we did, buy a whole kilogram (2.2 lbs) of Amarelli Spezzatina [note: no sugar at all!]) and using them through out the day.
• We had the space and purchased our own infrared sauna a few years ago (on sale at Costco). If you follow that path, remember lots of water and slowly work up duration. For water in the sauna, we use Gerolsteiner Natural Mineral Water (available from Trader Joe’s) because of it’s high mineral content.
• Reduce the foods high in choline in your diet (see the list here)
• Keep nuts (Costco/Trader Joe’s?) handy for snacks. Remember peanuts are NOT nuts.

Addendum Discussion

First, and most important: items are suggestions only. People should do only what works for them. Each item increases the odds of a positive shift a little, key word is a little. Occasionally, I will get emails from people who experienced awesome changes from a single item (Neem and Tulsi are often cited). If something is not mentioned and it helps — keep doing it!.

• “a complicated relationship with licorice in the past” In follow up email, the cause appeared to be licorice oil being used. For licorice, I strongly favor the Italian Spezzatina. This person had no problem with the Spanish equivalent… when there is an item with a suggestion, before discarding entirely, consider the following:
  • If in an oil, you may be reacting to the carrier and not the substance. On occasion, it can be preservatives in it. For example Mountain herbs is 35-45% alcohol
  • The dosage may be too high to start, I always advocate starting low and slowly increasing
  • For capsules, read Study: Herbal Supplements Full Of Contaminants, Substitutes, And Fillers [2013]
    • We rarely buy prepared capsules, instead we buy organic bulk and make our own capsules. It is both cheaper, organic sourced (very few supplement bottles cite organic) and less likely (but not impossible) to have the issues cited above.

Concerning salmon, etc. Remember everything is odds. Nothing is guaranteed. You want to shift your odds where practical.

One item that may be worth trying is Akkermansia Probiotic (link with 30% discount). Both my wife and I have, and are, using it — it has definitely made changes (I had almost the same amount that he has – 52%ile, and there was a change). The link is for by subscription — of course, once you get your first bottle, you can just suspend the subscription.

A reader asked this question about Lactobacillus Salivarius (NCBI 1624). I am doing it for this single strain — for a mixture, just do each one and consolidate the results.

FIRST, login to access more information. Most of the information below do not require a login.

SECOND, look up the species

THIRD, click on that link above, this will show what helps or hurts this probiotic. Click on [Probiotics] for a start — you want to make sure that other probiotics you are taking support it. Note that I have gone thru a few retail mixtures and the probiotics fight each other! Buyer beware of mixtures! In this case we have 26 listed. 24 are friendly and 2 are hostile (bacillus laterosporus (probiotic), bifidobacterium bifidum (probiotics)

FOURTH, There is one more source of information, by looking at the associations in microbiome report. On the same page above click the red interactions button.

This will show a page of bacteria that are associated positively or negatively with this bacteria (i.e. probiotic). You will need to type in the family names:

Looking for enterococcus, pediococcus and E.Coli — we find nothing, looking at clostridium butyricum (probiotics), we do not find it, but it’s siblings appear to have a positive impact.

FIFTH, check prebiotics (they are often bunddled in probiotics) — 18 are friendly and two are hostile

SIXTH Check Herbs and Spices, the results here may impact the rotation of herb and spices. You want to take friendly when you are doing the probiotics. Although they will likely not take up residency, the chemicals that they produce while in transit are good… you do not want to reduce the production of those chemicals.

Bottom Line

Doing your homework will get much better results than just tossing them in your mouth with something that will inhibit it!

This is using data from the study being done with BiomeSight. We will only use their samples. After the first review, a z-score of 6.4 or higher (or a lots of items) was set as a cutoff point. The following ignore False Detection Rate.

• Conclusion: the ENZYME production of the microbiome is by far the strongest indicator.
• The reference set consists of 1037 heterogenous samples (i.e. no Long COVID, but a variety of medical conditions) and 154 samples with Long COVID

Taxon Patterns

Care needs to be taken with these numbers because the frequency of reporting on a bacteria is a factor that impacts the z-score. The data for this table is available at Citizen Science site and independent analysis is strongly recommended. This table is a simplified view of very complex data.

tax_name	tax_rank	No Symptom Mean	Symptom Mean	Z-Score	Change
Terrabacteria group	clade	715040	520885	10.4	73%
Firmicutes	phylum	652452	502830	9.0	77%
Tenericutes	phylum	2562	6362	-7.9	248%
Eubacteriales	order	609888	482468	7.9	79%
Mollicutes	class	2562	6362	-7.9	248%
Clostridia	class	613743	487719	7.8	79%
Emticicia oligotrophica	species	769	2553	-6.8	332%
Faecalibacterium prausnitzii	species	100292	142415	-6.7	142%

End Product Patterns

End products only had a single item above our 6.3 z-score threshold with a very small shift.

EndProduct	No Symptom Mean	Symptom Mean	No Symptom StdDev	Change
H2	1329	1307	6.6	98%

KEGG Enzyme Patterns

This is where we see a massive number of patterns(182!!) with very high z-scores (i.e. 6.4 or higher). This hints that the bacteria associated with these enzymes may be a good target to modify.

EnzymeName	No Symptom Mean	Symptom Mean	No Symptom StdDev	Change
dihydrourocanate:acceptor oxidoreductase	58562	147222	-18.2	251%
(S)-3-hydroxy-3-methylglutaryl-CoA acetoacetate-lyase (acetyl-CoA-forming)	55210	142006	-18	257%
(1->4)-alpha-D-galacturonan reducing-end-disaccharide-lyase	54601	139740	-17.7	256%
acetyl-CoA:kanamycin-B N6′-acetyltransferase	55382	140425	-17.7	254%
acetyl-CoA:2-deoxystreptamine-antibiotic N3-acetyltransferase	56590	141511	-17.6	250%
poly(deoxyribonucleotide)-3′-hydroxyl:5′-phospho-poly(deoxyribonucleotide) ligase (ATP or NAD+)	55562	141080	-17.6	254%
D-serine ammonia-lyase (pyruvate-forming)	55931	140065	-17.6	250%
poly(deoxyribonucleotide)-3′-hydroxyl:5′-phospho-poly(deoxyribonucleotide) ligase (ATP, ADP or GTP)	55562	141080	-17.6	254%
alpha-maltose-6′-phosphate 6-phosphoglucohydrolase	57944	142024	-17.5	245%
ATP phosphohydrolase (ABC-type, iron(III) enterobactin-importing)	57953	141331	-17.4	244%
protein-Npi-phospho-L-histidine:D-mannose Npi-phosphotransferase	66964	152717	-17.4	228%
ATP phosphohydrolase (ABC-type, Fe3+-transporting)	68676	154113	-17.4	224%
D-psicose 3-epimerase	70754	155871	-17.2	220%
D-tagatose 3-epimerase	70754	155871	-17.2	220%
2′-(5-triphosphoribosyl)-3′-dephospho-CoA:apo-[citrate (pro-3S)-lyase] 2′-(5-phosphoribosyl)-3′-dephospho-CoA-transferase	77143	161549	-17.1	209%
ATP:3′-dephospho-CoA 5-triphospho-alpha-D-ribosyltransferase	78363	162298	-17	207%
2,4,6/3,5-pentahydroxycyclohexanone 2-isomerase	75196	158863	-16.9	211%
ATP:[protein]-L-tyrosine O-phosphotransferase (non-specific)	60964	143510	-16.9	235%
acetyl-CoA:citrate CoA-transferase	79352	162680	-16.7	205%
L-aspartate:tRNAAsx ligase (AMP-forming)	63596	144560	-16.7	227%
poly(deoxyribonucleotide)-3′-hydroxyl:5′-phospho-poly(deoxyribonucleotide) ligase (ATP)	69642	156282	-16.7	224%
penicillin amidohydrolase	69734	151011	-16.6	217%
protein-Npi-phospho-L-histidine:D-mannitol Npi-phosphotransferase	57950	140690	-16.5	243%
ATP:D-erythronate 4-phosphotransferase	65433	145262	-16.4	222%
acetate:holo-[citrate-(pro-3S)-lyase] ligase (AMP-forming)	90668	176404	-16.4	195%
ATP:D-threonate 4-phosphotransferase	65433	145262	-16.4	222%
D-aspartate:[beta-GlcNAc-(1->4)-Mur2Ac(oyl-L-Ala-gamma-D-Glu-L-Lys-D-Ala-D-Ala)]n ligase (ADP-forming)	73487	157884	-16.4	215%
4-phospho-D-erythronate:NAD+ 3-oxidoreductase	65773	145502	-16.3	221%
4-phospho-D-threonate:NAD+ 3-oxidoreductase	65773	145502	-16.3	221%
nucleoside-triphosphate diphosphohydrolase	69217	153915	-16.2	222%
4-amino-5-aminomethyl-2-methylpyrimidine aminohydrolase	75806	165018	-15.7	218%
ATP:D-glycero-alpha-D-manno-heptose 7-phosphate 1-phosphotransferase	81281	169414	-15.7	208%
aryl-ester hydrolase	77314	159122	-15.6	206%
palmitoyl-CoA hydrolase	76772	157265	-15.4	205%
UDP-alpha-D-glucose:1,2-diacyl-sn-glycerol 3-alpha-D-glucosyltransferase	91112	172382	-15.4	189%
D-tagatose 1,6-bisphosphate D-glyceraldehyde-3-phosphate-lyase (glycerone-phosphate-forming)	75959	152459	-15.2	201%
ADP-alpha-D-glucose:alpha-D-glucose-1-phosphate 4-alpha-D-glucosyltransferase (configuration-retaining)	63077	146386	-15.1	232%
L-glutamate:tRNAGlx ligase (AMP-forming)	97313	177576	-14.5	182%
oligosaccharide 6-alpha-glucohydrolase	96720	174292	-14.3	180%
S-adenosyl-L-methionine:tRNA (adenine22-N1)-methyltransferase	96117	168859	-13.9	176%
alkylated-DNA glycohydrolase (releasing methyladenine and methylguanine)	93342	182716	-13.7	196%
sn-glycerol 3-phosphate:quinone oxidoreductase	113940	189562	-13.6	166%
L-iditol:NAD+ 2-oxidoreductase	113731	190510	-13.4	168%
(3S)-citryl-CoA oxaloacetate-lyase (acetyl-CoA-forming)	108775	197009	-13.3	181%
N-succinyl-LL-2,6-diaminoheptanedioate amidohydrolase	88237	163157	-13	185%

KEGG Product

Products are the output of enzymes. Various enzymes may produce the same product. Our starting assumption was that products would have stronger association than enzymes. That was not shown in the data.

CompoundName	No Symptom Mean	Symptom Mean	No Symptom StdDev	Change
Acetoacetate	37878	55442	-8.1	146%
Reduced electron-transferring flavoprotein	106971	149551	-6.9	140%
Dialkyl phosphate	773	2553	-6.8	330%
Indole-3-acetate	773	2553	-6.8	330%
Pseudouridine 5′-phosphate	109418	150579	-6.7	138%
3-Hydroxy-3-(methylthio)propanoyl-CoA	758	2494	-6.7	329%
3-Oxopropionyl-CoA	758	2494	-6.7	329%
N-Acetyl-beta-D-glucosaminylamine	760	2473	-6.7	325%
(2E,4Z)-2,4-Dienoyl-CoA	68809	95899	-6.6	139%
Short-chain trans-2,3-dehydroacyl-CoA	103627	144711	-6.6	140%
(2E,4E)-2,4-Dienoyl-CoA	68809	95899	-6.6	139%
4-(4-Deoxy-alpha-D-gluc-4-enuronosyl)-D-galacturonate	33961	47705	-6.6	140%
4-Hydroxyphenylglyoxylate	113250	152269	-6.5	134%
Oleoyl-[acyl-carrier protein]	735	2376	-6.5	323%
(4Z)-Hexadec-4-enoyl-[acyl-carrier protein]	735	2376	-6.5	323%
N6′-Acetylkanamycin-B	34762	48298	-6.5	139%
(6Z)-Hexadec-6-enoyl-[acyl-carrier protein]	735	2376	-6.5	323%
Pyocyanine	751	2381	-6.5	317%
(1E,3E)-4-Hydroxybuta-1,3-diene-1,2,4-tricarboxylate	1430	4549	-6.5	318%
Aldose	764	2429	-6.4	318%
Molybdoenzyme molybdenum cofactor	119172	159672	-6.4	134%
N3-Acetyl-2-deoxystreptamine antibiotic	35939	49415	-6.4	137%

KEGG Substrate

Subtrate are the fuel for enzymes reaction. Various enzymes may consume the same compound. Our starting assumption was that substrate would have stronger association than enzymes. That was not shown in the data.

CompoundName	No Symptom Mean	Symptom Mean	No Symptom StdDev	Change
Dihydrourocanate	38026	55395	-7.8	146%
(S)-3-Hydroxy-3-methylglutaryl-CoA	34820	50269	-7.4	144%
Electron-transferring flavoprotein	106880	149551	-6.9	140%
threo-3-Hydroxy-D-aspartate	760	2532	-6.9	333%
3-(Methylthio)acryloyl-CoA	757	2494	-6.8	329%
3-Hydroxy-3-(methylthio)propanoyl-CoA	757	2494	-6.8	329%
3-Oxopropionyl-CoA	757	2494	-6.8	329%
ADP-sugar	772	2553	-6.8	331%
Aryl dialkyl phosphate	772	2553	-6.8	331%
beta-D-Mannose	772	2553	-6.8	331%
D-erythro-3-Hydroxyaspartate	761	2532	-6.8	333%
Pseudouridine	105195	147050	-6.8	140%
N4-(Acetyl-beta-D-glucosaminyl)asparagine	759	2473	-6.7	326%
Short-chain acyl-CoA	103536	144711	-6.7	140%
(2-Amino-1-hydroxyethyl)phosphonate	752	2431	-6.6	323%
trans-2,3-Dehydroacyl-CoA	68745	95899	-6.6	140%
(S)-4-Hydroxymandelate	113176	152269	-6.5	135%
5-Methylphenazine-1-carboxylate	750	2381	-6.5	317%
Hexadecanoyl-[acp]	1468	4753	-6.5	324%
Kanamycin B	34729	48298	-6.5	139%
Octadecanoyl-[acyl-carrier protein]	734	2376	-6.5	324%
(1E)-4-Oxobut-1-ene-1,2,4-tricarboxylate	739	2338	-6.4	316%
2-Deoxystreptamine antibiotic	35916	49415	-6.4	138%
Adenylated molybdopterin	119083	159672	-6.4	134%
Alditol	763	2429	-6.4	318%
beta-Carotene	720	2314	-6.4	321%
Molybdate	119083	159672	-6.4	134%

Bottom Line

Several years ago, I hypothesized that a symptom or condition is the result of a coming together of many small deviations in individual bacteria representation. There may be 10 different combination of bacteria with none overlapping causing a symptom. The inspiration for this was observing the literature and experience of people with Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) — a sibling condition to Long COVID. This model is contrary to the common belief that there is a single or small number of items that is the cause. My looking at Brain fog (using same technique as above Brain Fog: Microbiome scents…) came up with nothing. That was not desired, but almost expected because that population is very heterogenous for cause with a long time since the triggering event for the microbiome to diverge from each other (often treatment attempts would be a factor). With long COVID, we have a short time since the triggering event and the people tend to be treatment naïve, This makes finding patterns a lot easier (when you look under the right rocks!).

Almost everything is overproduction. This may be caused by the immune system ramping up to provide fuel to fight COVID. The microbiome is stuck in an on-state, likely with cross talk between enzymes keeping it stuck on. The term of the Pasteur Institute for Tropical Medicine, “an occult infection” describes the behavior seen nicely.

Addressing the few microbiome shifts is one approach — but the enzymes dominate in both statistical significance and number of items, It is likely the best path to address the enzymes instead of individual bacteria.

Leave a Reply

Logged in as Admin. Log out? Required fields are marked *

Comment *