The parents provided rich information:

• OmbreLab results
• OmbreLab results processed thru BiomeSight (preferred because of Special Studies)
• OATS

Both children are in the same household, so diet will be similar.

The Son

 Our son is 14 and weighs 90 pounds and is Autistic.  It’s amazing that he sleeps well (10 hours) although the minute he gets up he is not stop moving, spinning, stimming, etc.  He can be redirected and can follow directions and can slowly progress is school although if we could calm him down I think he could learn a ton.  He understands everything we say and can ask for things if he slows down although he is not conversation.  Every 6-8 weeks he seems to have face flair (although hasn’t been that bad the last 6-10 months although we don’t know if it’s yeast, bacteria, etc.  Once he has it his OCD kicks in and he can’t stop putting lotions and creams on it.  He’s a skinny kid with not much muscle although can ride a bike, run a mile or two, etc.  

The Daughter

Our daughter is almost 16 and is not verbal although uses a device.  She has the autism diagnosis as well although also has a genetic mutation of the WDR45 gene (BPAN) which can create iron accumulation in the brain.  We spend a lot of time with supplements, organic nutrition, gluten free, time on the treadmill and sauna and more to make sure her body is detoxing the best it can.  She understands everything you say just can’t speak back.  Where my son is not stop she would cuddle all day and be 100% content.  She also  has fine and gross motor delays and in general is just cautious in her physical movement.  We’ve treated for SIBO several times, etc. although we’ve obviously never solved the problem.  She isn’t necessarily constipated although 95% of the time needs assistance with a small suppository for her to go the bathroom.  The stool is always there it’s just like she doesn’t have the urge to release it. 

High Level Comparisons

Below are summaries of the two children for comparison

	Daughter	Son
Biomesight Gut Wellness	77.37	85.89
Potential Medical Conditions Detected	none	a few fuzzy issues
Bacteria deemed Unhealthy	7	7
Dr. Jason Hawrelak Recommendations	98.8 %ile	99.7 %ile
KEGG Enzymes – Low Levels	Over reported (18.5)	Over reported (158)
Percentile Representation	High in rare bacteria	High in rare bacteria
Special Studies: Autism	49 matches	63 matches
Special Studies: Constipation	28 matches	19 matches
Probable Symptoms	Nothing	High Anxiety 85 Depression 85 ME/CFS with IBS 72

Comments on the above

Going thru the above, we see little difference between the two on may ways to evaluate the microbiome. However, there were a few where we see a great contrast:

• Enzymes at both extremes are very much higher for the son (4x higher at both top and bottom 1%iles)
• Bacteria under 1%ile is much higher with the son
• Box-Whisker and Kaltoft-Moldrup are much higher with the daughter
• Having profiles matching predicted symptoms, son only has significant matches

Daughter Action Plan

For the daughter, build up a consensus report from:

• Special Studies Autism (49 all low)
• Special Studies Constipation (28 all low)
• PubMed Autism (6)
• Box-Whisker (all were high)
• Kaltoft-Moldrup (17 low, 72 high)

Obtain Probiotics from:

• OATS
• From KEGG Computation
• From Consensus

Son Action Plan

For the son, build up a consensus report from:

• Special Studies Autism (49 all low)
• Special Studies Anxiety (30 all low)
• Special Studies Depression (39 all low)
• Special Studies ME/CFS with IBS (49 all low)
• PubMed Autism (6)
• Box-Whisker (5 too low, 44 too high)
• Kaltoft-Moldrup (36 low, 39 high)

Obtain Probiotics from:

• OATS
• From KEGG Computation
• From Consensus

This looks like a lot of work!

Actually, it’s not — I have done a video doing the son to illustrate the steps better.

Son Consensus

My person picks from the list are below. I was not overly happy with the To Take suggestions.

• gluten-free diet
• carboxymethyl cellulose (prebiotic)
• navy bean
• beef
• blueberry
• glycyrrhizic acid (licorice)
• iron supplements
• sesame cake/meal
• Conjugated Linoleic Acid

Looking at the avoid items, I would be inclined to try shifting these items out of his diet (if it is in it)

• walnuts
• brown rice
• apples
• inulin (prebiotic)
• chicken (beef is a to take)
• all probiotics except those below
• Barley

Son Probiotics

See video for details, the key items that occured at the

• Escherichia coli – symbioflor 2 e.coli probiotics or Mutaflor
• bacillus subtilis (many retail products have it)

Daughter Consensus

Some of the items may please the daughter.

• Top item is Cacao — trying shifting her to 85% or higher
• lactobacillus casei (probiotics) – I would suggest Yakult probiotic drink (it is well studied – lactobacillus casei shirota (probiotics))
• pea (fiber, protein) – perhaps pea soup occasionally
• Human milk oligosaccharides (prebiotic, Holigos, Stachyose)
• trametes versicolor(Turkey tail mushroom)
• mediterranean diet

On the avoid side, we see also:

• barley
• grapes, wine, and grape seed extract (which is interesting because aspirin is a to take)
• cannabinoids – which you likely approve of, as a parent
• inulin (prebiotic)
• saccharomyces cerevisiae (probiotics)

Daughter Probiotics

The consensus probiotics were much different than her brother

Looking at the OATS results, we see Bifidobacterium dominating it. This is not a surprise because Human milk oligosaccharides usually shows up when Bifidobacterium needs good encouragement. Please note: the list below did not use any microbiome data — just the OATS results.

The revised KEGG Probiotic had two items that tend to be frequent:

• Bacillus subtilis (which the brother also had suggested in multiple ways)
• Escherichia coli (symbioflor 2 e.coli probiotics was a positive take on the consensus report)

If there are any Lactobacillus probiotics being taken, I would eliminate them.

Bottom Line

Everything in this post is created by modelling data and nothing has been validated clinically. The advantage with modelling is that it is usually better than than a MD tossing an idea out in frustration.

All suggestions should be reviewed by a qualified medical professional before starting.

This is an old drug with most of the studies in 1970-1990. It is very different in action from antihistamines (i.e. it is NOT an antihistamine) with an awesome safety record. A reader asked me to assemble and review the literature. Mast cell and histamine issues are often reported with other conditions; the domination of other symptoms may mask the mast cell issues.

Conditions Often Comorbid With Mast Cell Diseases

• Chronic inflammatory response syndrome (CIRS)

• Fibromyalgia syndrome (FMS)

• Ehlers-Danlos Syndrome (EDS)

• Gulf War Illness (GWI)

• Interstitial cystitis/bladder pain syndrome (IC/BPS)

• Irritable bowel syndrome (IBS)

• Kounis syndrome

• Multiple chemical sensitivity syndrome (MCSS)

• Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS)

• Post-Lyme syndrome

• Postural orthostatic tachycardia syndrome (POTS)

• Post-traumatic stress disorder (PTSD)

Cromolyn sodium is a medication used to manage bronchial asthma, allergic rhinitis, and certain allergic eye conditions such as vernal conjunctivitis, keratitis, and keratoconjunctivitis. It is a mast cell stabilizer. 

Cromolyn Sodium – Continuing Education Activity – StatPearls [NIH site – 2022]

Notes from the above:

• Derived from Mediterranean herb called Ammi visnaga
• Cromolyn sodium metered spray is used to prevent and relieve nasal symptoms of hay fever and other nasal allergies like runny or itchy nose, sneezing, and allergic stuffy nose.[7]
• Cromolyn sodium is off-labeled used for preventing serious reactions to foods.[8]
• The effect of cromolyn sodium on mast cells lasts for approximately 6 hours
• Cromolyn sodium inhibits immediate and late reactions.[15][12] 
• Adverse reactions reported with the inhalation solution were throat irritation and hoarseness, esophagitis, laryngeal and pharyngeal edema, drowsiness, dizziness, bronchial irritation, pulmonary infiltrates, and cough.[19]
• Given its safety profile and current evidence, its use is likely most favorable for those who cannot tolerate intranasal corticosteroids or require short-term prevention before known exposures.^6,7 

I was unable to find any generic clinical studies for Cromolyn with Myalgic encephalomyelitis/chronic fatigue syndrome or ASD (despite mast cell issues being associated with inflammation of the brain.

Availability

In the US available at retail (Walmart etc) from several brand,

• NasalCrom Nasal Spray Allergy Symptom Controller, 200 Sprays, – $16.26 – i.e. 1 cent per spray
• HealthGuard Cromolyn Sodium Nasal Spray, 200 Sprays – $12.99

The Herb Ammi Visnaga (Visnaga daucoides )

Used for over 2000 years

• “it is used in modern medicine to treat many aliments such as renal colic and coronary insufficiency, and is used as an antioxidant, antifungal, and antibacterial, with a larvicidal effect on mosquito larvae…. A.K.A. Khella Baldi or toothpick weed, In Arabic countries A. visnaga has many common names such as Khella baladi, Khella, Khellah, Khellakl, Chellah, Kella, Gazar sheitani, Kammon habashi, Bizer Al-Khilla, Kulla, and Swak Al-Nabi. In Turkey, it is known as disotu, kilir, and hiltan.” [2020]

Headaches — likely a spray symptom

A study of this spray against a placebo was found “The most common adverse events in both groups were headache and rhinitis, and there was no significant difference in the rates of such events between groups.” [2002]. This suggests that the spraying is a more likely cause, not the cromolyn nasal spray.

Literature

• Ways to Address Perinatal Mast Cell Activation and Focal Brain Inflammation, including Response to SARS-CoV-2, in Autism Spectrum Disorder. [2021]
• Mast Cells, Stress, Fear and Autism Spectrum Disorder. [2019]

Bottom Line

Given its awesome safety record, I would suggest a discussion with your medical professional on doing a course of over-the-counter nasal spray (done before each meal and at bed time – because it effects lasts only 6 hours). A single bottle should last for 50 days.

With regular testing and following suggestions, Hawrelak’s criteria out of range items went from 9 , to 7, to 4. Suggestions appear to work (slowly). This is a follow up of prior posts, the bottom line is that with

• Jan 25,2022 A review of a ME/CFS Microbiome
• Jul 16,2022 Comparing 4 ME/CFS Samples with New Tool

I am going to start with the same comparison as I did in the prior post above, and then move on to do an analysis using the latest tools available.

Why Follow Up Posts are important

The first item is simple, does the model and suggestion appear to work. Everything is theoretically computed, not based on clinical practice or clinical studies. The second item is that these posts encourages people to try suggestions, or to do “self-serve” with the site.

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 appears to 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.

Comparisons between Samples

We have the latest test processed through both Ombre and Biomesight. This analysis will use Ombre because that is what was used in prior; latest one will use Biomesight so we can use the new special studies for suggestions

The high lights are:

• Sample quality is better than the prior samples, we see more bacteria reported
  • Ombre is reporting more (as we saw in a prior post)
• Improvement in the critical bacteria identified by Dr. Jason Hawrelak, we are down to just 4 bacteria outside of range for both (from prior 7 and 9)
• Similar with Medivere, MyBioma,Nirvana/CosmosId, and XenoGene ranges
• On the opposite side, we have more outside of Box-Plot-Whiskers, Lab Range (+/- 1.96SD) and Kaltoft-Moldrup – this is not unexpected because we have a lot more bacteria. The criteria improved are the criteria with a fix set of bacteria that are usually there in most samples.
• The number of extreme enzymes levels (using Ombre) have been significantly reduced. Compounds have no clear shift

Measure	8/22/2022 BIOMESIGHT	8/22/2022 OMBRE	6/15/2022	5/16/2022	3/15/2022	11/21/2021
Lab Read Quality	5	5	4.8	4.8	3.6	4.2
Bacteria Reported By Lab	657	765	515	669	593	473
Bacteria Over 99%ile	0	9	0	1	0	1
Bacteria Over 95%ile	15	45	4	22	21	9
Bacteria Over 90%ile	38	82	22	40	48	21
Bacteria Under 10%ile	168	35	43	112	23	87
Bacteria Under 5%ile	84	16	18	60	5	46
Bacteria Under 1%ile	24	3	0	9	0	2
Lab Read Quality	BiomeSight	Ombre	Ombre	Ombre	Ombre	Ombre
Pathogens	47	43	37	42	38	33
Outside Range from JasonH	4	4	7	7	9	9
Outside Range from Medivere	15	15	17	17	20	20
Outside Range from Metagenomics	8	8	8	8	7	7
Outside Range from MyBioma	4	4	8	8	10	10
Outside Range from Nirvana/CosmosId	20	20	23	23	22	22
Outside Range from XenoGene	4	4	8	8	10	10
Outside Lab Range (+/- 1.96SD)	16	14	2	7	6	2
Outside Box-Plot-Whiskers	135	118	42	71	72	25
Outside Kaltoft-Moldrup	155	209	113	148	143	91
Condition Est. Over 99%ile	0	0	0	0	0	0
Condition Est. Over 95%ile	0	0	0	0	0	0
Condition Est. Over 90%ile	0	0	0	0	1	0
Enzymes Over 99%ile	2	8	91	48	53	5
Enzymes Over 95%ile	48	78	599	305	182	147
Enzymes Over 90%ile	99	154	719	629	438	458
Enzymes Under 10%ile	175	75	70	71	38	169
Enzymes Under 5%ile	101	33	19	28	4	78
Enzymes Under 1%ile	20	3	2	0	0	0
Compounds Over 99%ile	31	36	6	93	143	19
Compounds Over 95%ile	131	265	186	277	190	89
Compounds Over 90%ile	371	380	303	435	220	118
Compounds Under 10%ile	212	282	321	528	333	133
Compounds Under 5%ile	54	186	195	378	310	100
Compounds Under 1%ile	23	62	0	118	267	42

The data suggests further improvement with the latest sample — in terms of numbers, dramatic!

Going Forward

We are going to use the Special Studies to generate the next set of suggestions. Remember that the numbers below are sensitive to the number of bacteria detected (i.e. lab quality). The latest sample had the highest quality which means that percentages may increase due to that alone.

	Percentage Matches	8/22/2022	7/21/2022	4/20/2022	2/22/2022	11/22/2021
1	Alcohol intolerance or Medication sensitivities	52	34	36	36	39
1	Allergic Rhinitis (Hay Fever)	33	31	35	35	36
0	Allergies And Food Sensitivity	30	32	32	32	34
0	Autism	38	28	31	31	44
1	Bloating	24	22	24	24	32
2	Brain Fog	28	32	39	39	26
2	Chronic Fatigue Syndrome (CFS/ME)	33	30	28	28	30
0	Cold Extremities	31	36	29	29	31
0	Constipation	17	18	15	15	29
2?	COVID19 (Long Hauler)	39	36	31	31	39
0	Depression	40	30	33	33	35
0	Easily irritated	28	31	24	24	27
2	General Fatigue	26	26	28	28	28
1	High Anxiety	28	26	25	25	26
1	Histamine or Mast Cell issues	28	24	27	27	32
0	Inflammatory bowel disease	43	48	28	38	43
0	irritable bowel syndrome	31	30	26	26	36
1	Intolerance of Extremes of Heat and Cold	32	30	24	24	36
0	ME/CFS with IBS	31	24	24	24	27
2	ME/CFS without IBS	30	25	23	23	28
0	Poor Gut Motility	25	37	31	31	35
2	Post-exertional malaise	22	24	23	23	26
0	SIBO	38	36	27	27	34
0	Tinnitus (ringing in ear)	32	34	24	24	28
2	Unrefreshed sleep	37	27	26	26	29

What surprised me, and delighted me, was the relative consistency of the numbers for the same person over time for many symptoms. This time, because of the special studies, we are going to focus on generating suggestions specific to the issues of greatest concerns (2) above.

• Brain Fog
• Chronic Fatigue Syndrome (CFS/ME)
• COVID19 (Long Hauler)
• General Fatigue
• ME/CFS without IBS
• Post-exertional malaise
• Unrefreshed sleep

Having high IBS or IBD values does not equate to ME/CFS with IBS — prior to my latter flares, I saw no gut issues at all.

The top Items are similar to each other:

• Pulses (Black Beans,Broad Beans,dry beans, dry broad beans, dry peas, chickpeas, cow peas, pigeon peas, lentils, Bambara beans, vetches, lupins), navy bean, pea (fiber, protein)
• symbioflor 2 e.coli probiotics,colinfant e.coli probiotics and d-ribose (which feeds e.coli)
• arabinoxylan oligosaccharides (prebiotic)
• lactulose
• green-lipped mussel
• levan
• fasting
• galacto-oligosaccharides (prebiotic)
• non-starch polysaccharides
• carob
• raffinose(sugar beet)
• wheat bran
• barley,oat
• sesame cake/meal
• Reduce choline (Beef, Chicken Eggs)

The top of the avoid list are many B-Vitamins. In prior posts, I speculated that the low vitamin B-levels seen in the blood of ME/CFS patients etc is because of greedy gut bacteria that prevents it being absorbed. I just did some literature browsing and this speculation appears to agree with some research.

Now, Wexler et al. show that some gut microbes may be able to pirate vitamin B12 from us as it passes through the digestive tract. Wexler et al. showed that a protein called BtuG on the surface of a type of gut bacteria called Bacteriodes grabs onto vitamin B12 with extraordinary strength. In fact, these bacterial proteins bind to vitamin B12 so strongly that they can even pry it away from our own vitamin B12 collecting protein.

Human gut Bacteroides capture vitamin B₁₂ via cell surface-exposed lipoproteins [2018]

This means that to avoid B-vitamins via food or supplements, despite having low blood level may be a rational approach. Getting Vitamin-B injections bypassing the gut bacteria may be a more effective approach to dealing with low vitamin-B. Choline is connected with B-Vitamins, see The B Vitamins and Choline: Overview and Methods [1998], hence the reduce choline suggestion above.

Probiotics

Escherichia coli was the top probiotic on the consensus. Using Ombre reading of the sample and the Kyoto Encyclopedia of Genes and Genomes, Escherichia coli is the top suggestion with items like Bacillus subtilis being 1/3 of contribution. The same results appear with the BiomeSight processing.

For flavonoids and polyphenols — it seems that supplementing with any should be avoided. The very few with a positive value has ‘equivalents’ being negative values. For amino acids: Conjugated Linoleic Acid, l-citrulline and proline (amino acid) are the top choices. Diet choices are low animal protein (which agrees with avoiding b-vitamins) and high fiber (which agrees with pulses above)

Bottom Line

Remember, these are not generic suggestion for anybody with ME/CFS. These are specific to this person’s microbiome. We see that the numbers for his microbiome has improved, but some items are still of concern. Our method of getting suggestions shifted over to using the Special Studies because he had a number of symptoms that were significant which were covered by the special studies, as well as his data being processed thru BiomeSight which we are using for special studies (apples to apples comparisons).

Again, the computations are based on the odds of items helping the microbiome to shift in the desired direction. Odds are just that, some will work, some will not — but ideally more will work than not. For the suggestions, remember ROTATION (Rotation — Essential for Changing the Microbiome), and pick items that work for you and your medical professional. The suggestions can be viewed as a series of dresses on a rack, you do not need to buy all of them to be fashionable. Use the ones that work for you.

Rotation as a key facet of fixing or keeping a microbiome healthy was first introduced to me in 1999 by Cecile Jadin, MD (Surgeon in South Africa) with the Occult Rickettsia Protocol from the Pasteur Institute of Tropical Disease (where her father worked). I have written about this prior, Rotation and Pulsing: Herbs, Probiotics, Antibiotics [2017], Rotate, Rotate, Rotate and Curcumin [2016], Continuous or Pulse Supplements? [2015]

Some literature on it:

• The effects of antibiotic cycling and mixing on acquisition of antibiotic resistant bacteria in the ICU: A post-hoc individual patient analysis of a prospective cluster-randomized crossover study [2022]
• Optimising efficacy of antibiotics against systemic infection by varying dosage quantities and times. [2020
• The effects of antibiotic cycling and mixing on antibiotic resistance in intensive care units: a cluster-randomised crossover trial. [2018]
• Fighting antibiotic resistance in the intensive care unit using antibiotics[2015].
• “with a 28-day on-off regimen with tobramycin or aztreonam” [2015]
• Antibiotic rotation strategies to reduce antimicrobial resistance in Gram-negative bacteria in European intensive care units: study protocol for a cluster-randomized crossover controlled trial [2014].
• Antibiotic rotation for febrile neutropenic patients with hematological malignancies: clinical significance of antibiotic heterogeneity[2013].
• The impact of an antibiotic cycling program on empirical therapy for gram-negative infections. [2006]

The Microbiome is not a Mechanical System

Think in terms of societies. Think about a gang that steals quality goods at a store. The antibiotic may be having a store detective. Issue solved – no more crime! Wrong, the gang will change to a different crime, they will adapt. It will become a continuous battle and escalation by both sides. Many people view the microbiome as a simple mechanical system unfortunately, bacteria continuously mutate (think of COVID mutations). If the antibiotic kills off 99% of the bacteria, the remaining 1% likely has a mutation that allowed it to survive. Give this 1% a few weeks and we are back to old levels with the antibiotics making no difference.

How does this apply to vitamins, amino acids, even food? The simplest explanation is that those items affect the growth of different bacteria. Different bacteria strains produces different compounds, including  bacteriocins (natural antibiotics). When you think about probiotics, remember that they often work due to their natural antibiotics (see above for literature).

To jump to the human body for an example, when you were 20 you likely drank a lots of pop, smoked.. ate a lot of fast food… and were “healthy” and fine. As you body ages (evolved), the same diet would worsen diabetes, weight, and a dozen other conditions that will evolve. The microbiome changes with time, in fact, it changes faster than your body. When you make one set of changes, like running 2 miles a day, the body may act up with structure damage as you age. Professional players in sports do not age well. Going on a specific type of diet may address one issue and trigger a different issue eventually.

What is the answer? All things in moderation. Instead of just one type of exercise, do a variety of appropriate exercises. Instead of a specific diet (which may deliver insufficient minerals or vitamins), rotate around different diets that addresses the issue you are trying to address. Every change you make, changes the microbiome– sometimes in unexpected ways.

In terms of suggestions from the microbiome prescription site, rotating suggestions is desired — usually there are many suggestions so it should not be hard.

What is the Ideal Rotation?

I do not know, I have inferred from the typical duration of antibiotic prescription that 7-10 days is a reasonable guess. I will give a simple example of one rotation that I do (maintaining), alternating the various forms of gluten (the specific types are listed after each).

• Oats – as porridge: avenin, C-hordeins, γ-hordeins, B-hordeins and D-hordeins
• Rye – as German 100% rye bread: γ-40k-secalins and high-molecular-weight secalins
• Wheat — as typical western baked goods: (ω5-gliadins, ω1,2-gliadins, α-gliadins, γ-gliadins and high- and low-molecular-weight glutenin
• Barley – as porridge: C-hordeins, γ-hordeins, B-hordeins and D-hordeins

My stools will change every cycle – IMHO, there is no perfect stool.

From Isolation and characterization of gluten protein types from wheat, rye, barley and oats for use as reference materials [2017].

The goal is almost like trying to herd a group of cats.

Most of the analysis that I have done recently has been either ME/CFS, Long COVID or Autism. If you look at the page entitled Medical Conditions with Microbiome Shifts from US National Library of Medicine you will see a lot of different conditions that may be influence by microbiome manipulation..

Today I got an email asking “My mother in law has Progressive supranuclear palsy (PSP), and Diabetes  and is in a very bad state, there is no help from the mainstream medicine” with samples of her.

My first action is to see if there are any published studies on PSP and the microbiome. There was just one study: “Unraveling gut microbiota in Parkinson’s disease and atypical parkinsonism‘ 2018.

Progressive supranuclear palsy (PSP) is a less well-known neurodegenerative brain condition which is sometimes misdiagnosed as Parkinson’s disease or Alzheimer’s disease (or other forms of dementia). Because of the similarity to some Parkinson’s symptoms during the early stages of the disease, PSP is included in a group of diseases called Parkinson’s Plus Syndrome or Atypical Parkinsonism. However, PSP progresses much faster, causes more severe symptoms, responds very poorly to Parkinson’s medication, and has a significantly reduced life expectancy.

Parkinson’s Europe

At this point out a recent news story, Woman who smelled her husband’s Parkinson’s helps scientists come up with diagnostic test, Sky News, Sep 7,2022. When some ones gut bacteria changes, their smell change. On a little morbid note, in WW2, they could tell dead soldiers apart from their smell (he’s a German, he’s an Italian, he’s an American) [Story]. So avoiding scented products may have health benefits. In Vietnam war, it was different — the Viet Cong favorite smell was Old Spice, it means that there were Americans close by.

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.

Analysis

We lack any special studies nor any Medical Conditions with Microbiome Shifts from US National Library of Medicine(PubMed) which matches this person except for Diabetes. We do not have enough samples for Special Studies for Diabetes. For the PubMed, this person is sitting at the 95%, so a definite include. There is another PubMed that seem appropriate and which is at the 99%ile, Brain Trauma

Looking at data in detail, we see a definitely interesting microbiome. We have a high quality sample

Criteria	Current Sample	Comment
Lab Read Quality	10.6	High Quality
Bacteria Reported By Lab	960	Very high Count
Bacteria Over 99%ile	23	7 is expected, so high
Bacteria Over 95%ile	85	48 is expected
Bacteria Over 90%ile	155	96 is expected
Bacteria Under 10%ile	296	96 is expected
Bacteria Under 5%ile	267	48 is expected
Bacteria Under 1%ile	242	7 is expected
Rarely Seen 1%	56
Rarely Seen 5%	182
Pathogens	59
Outside Range from JasonH	8
Outside Range from Medivere	16	Candidate
Outside Range from Metagenomics	8
Outside Range from MyBioma	7
Outside Range from Nirvana/CosmosId	19	Candidate
Outside Range from XenoGene	5
Outside Lab Range (+/- 1.96SD)	37	Candidate
Outside Box-Plot-Whiskers	194	Candidate
Outside Kaltoft-Moldrup	347	Candidate
Condition Est. Over 99%ile	3
Condition Est. Over 95%ile	20
Condition Est. Over 90%ile	29

Dr. Jason Hawrelak Recommendations are at 99.7% so no red flags from that. There is a huge number of PubMed matches, I will only use the best ones for what was reported for this person. The following sets of suggestions are going to be used for our Consensus Report

• PubMed: Brain Trauma
• PubMed: Type 2 Diabetes
• Outside Range from Medivere
• Outside Range from Nirvana/CosmosId
• Outside Lab Range (+/- 1.96SD)
• Outside Box-Plot-Whiskers
• Outside Kaltoft-Moldrup

The consensus download is below

Probiotics

From KEGG, we see many of the Equilibrium and PrescriptAssist bacteria listed with the following further down the list (in descending order) (Remember : Lacticaseibacillus is the new name for Lactobacillus)

• Escherichia coli
• Bacillus subtilis
• Clostridium butyricum
• Akkermansia muciniphila
• Lacticaseibacillus casei
• Enterococcus faecium
• Lactiplantibacillus plantarum
• Enterococcus faecalis
• Lacticaseibacillus rhamnosus

From the consensus we have (in decreasing order), a similar list.

• lactobacillus rhamnosus gg (probiotics)
• enterococcus faecium (probiotic)
• Akkermansia muciniphila (probiotic)
• mutaflor escherichia coli nissle 1917 (probiotics)
• bacillus coagulans (probiotics)
• NEITHER Equilibrium nor PrescriptAssist were positive

My gut feeling is that the following products are likely a reasonable choice.

• L.Plantarium – Custom Probiotics
• Escherichia coli – mutaflor escherichia coli nissle 1917 (probiotics), if not available, Symbioflor-2 as an alternative.
• Akkermansia Muciniphila — Pendulum
• Clostridium butyricum – Miyarisan
• Enterococcus faecalis + Bacillus subtilis + bacillus coagulans – ThreeLac

I would suggest starting with whatever arrives first, starting with a low dosage and increasing every second day. Remember to review with your medical professional.

Vitamins

A vitamin B complex and Vitamin C are recommended

Supplements

The top items are all available on Amazon and other stores:

• melatonin supplement
• l-taurine
• epicatechin
• Hesperidin (polyphenol)
• mastic gum (prebiotic)
• sucralose

Herbs And Spices

Diet Style

Bottom Line

This person wife also has issues, see And now for a different condition… Part 2. The suggestions are different but creating two different menus everyday would be challenging. See that post for a possible solution.

One of the items cited that I did not include above was dopamine (prescription). Looking at a treatment site for PSP in an attempt to do cross-validation, we see that dopamine is a factor, but things are more complex, see: Excessive dopamine neuron loss in progressive supranuclear palsy [2008].

When I started this project, I saw the potential of using the microbiome as a method to identify candidate treatments in the absence of successful clinical treatments. This is the first attempt of putting this into practice.

Most of the analysis that I have done recently has been either ME/CFS, Long COVID or Autism. If you look at the page entitled Medical Conditions with Microbiome Shifts from US National Library of Medicine you will see a lot of different conditions that may be influence by microbiome manipulation..

In the same email I got an second challenge: “Father In Law – Diabetes, Heart conditions and High Blood Pressure” with samples of her.

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.

Analysis

In this case we have an even higher lab quality than the wife, but a lot less bacteria reported. This means that the microbiome is likely a lot less fragmented than the wife.

Criteria	Current Sample
Lab Read Quality	11.5
Bacteria Reported By Lab	628
Bacteria Over 99%ile	4
Bacteria Over 95%ile	27
Bacteria Over 90%ile	48
Bacteria Under 10%ile	330
Bacteria Under 5%ile	295
Bacteria Under 1%ile	237
Rarely Seen 1%	12
Rarely Seen 5%	41
Pathogens	50
Outside Range from JasonH	7
Outside Range from Medivere	21
Outside Range from Metagenomics	9
Outside Range from MyBioma	4
Outside Range from Nirvana/CosmosId	25
Outside Range from XenoGene	6
Outside Lab Range (+/- 1.96SD)	14
Outside Box-Plot-Whiskers	54
Outside Kaltoft-Moldrup	246
Condition Est. Over 99%ile	1
Condition Est. Over 95%ile	1
Condition Est. Over 90%ile	8

Dr. Jason Hawrelak Recommendations has him at the 75%ile — so off, but not really bad. Following the same pattern of analysis as the wife (since we have no matching special studies):

• PubMed: Hypertension (High Blood Pressure 
• PubMed: Type 2 Diabetes
• PubMed: Coronary artery disease
• Outside Range from Medivere
• Outside Range from Nirvana/CosmosId
• Outside Lab Range (+/- 1.96SD)
• Outside Kaltoft-Moldrup

The consensus download is below

Probiotics

From KEGG, we do NOT find Escherichia coli near the top of the list. We see the usual odd bacteria strains from Equilibrium and Prescript Assist then:

• Bacillus subtilis
• Bacillus licheniformis

With microbiome labs/ megasporebiotic being a good match.

From the consensus we have (in decreasing order), a similar list.

• lactobacillus reuteri (probiotics) see this list for products that combines this with one or more of the following (also recommended)
  • lactobacillus plantarum (probiotics)
  • lactobacillus rhamnosus gg (probiotics)
• pediococcus acidilactic (probiotic) – for example: Imagilin / NutriLots
• bacillus subtilis (probiotics) – see above
• The following are shared with his wife (but lower priority)
  • enterococcus faecium (probiotic)
  • Clostridium butyricum
    

I would suggest starting with whatever arrives first, starting with a low dosage and increasing every second day.

Prebiotics

Vitamins

Most of the vitamin B’s are to be avoided. Selenium, magnesium, vitamin a and folic acid,(supplement Vitamin B9) are the top items.

Supplements

The top items are all available on Amazon and other stores:

• polymannuronic acid
• glycine
• proline (amino acid)
• Taxifolin
• grape polyphenols

Herbs And Spices

Diet Style

Specific Foods

One item really jumps out — Burdock Root (Gobo in Japan)- which is available as a supplement if not available as a fresh vegetable. It is high in Inulin (but inulin is much lower, just 81 — so other components may be playing a significant role)

Bottom Line

The diet style is a major contrast with the wife — this creates the frustration of needing almost a double food preparation. To address this issue, I imported both consensus list into Excel, used a VlookUp function to display the values besides each modifier and then identified items that are positive for both and then order by the total of each.

This allows one menu to be used for both of them. Perhaps a little less effective, but likely a lot less frustrating (and thus better compliance). I attached it as an example.

This is a follow up to a prior post:

• Jul 8, 2022. A Complex Microbiome Situation

This person did his tests using OmbreLabs.com and then transfer the data to biomesight.com. This allows us to use special studies to select bacteria. I am also, as part of my own learning (as well as the readers), going to do some comparison between the OmbreLabs and BiomeSight reports on the same data (i.e. FASTQ files).

I had another sample analyzed at Ombre, and there were already changes in my flora, even in a short amount of time. And they correlate with me feeling a bit better. So thank you. I’m still trying to crunch the data and make sense of the new results, and other than your great Dr. AI, I am using this new feature by Ombre which I find very clear (old sample first, new sample after)

Why Follow Up Posts are important

The first item is simple, does the model and suggestion appear to work. Everything is theoretically computed, not based on clinical practice or clinical studies. The second item is that these posts encourages people to try suggestions, or to do “self-serve” with the site.

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 appears to 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.

Comparisons between Samples

See this other review of a series of ME/CFS microbiomes from another person that I recently did, ME/CFS Follow Up Microbiome Samples. If you are new to this series, you may wish to review A new specialized selection of suggestions based on statistical significance for symptoms.

First, I do not know the best way to compare samples — what I usually do is put all of the numbers side by side. Special attention needs to be paid to Lab Read Quality. A poorer read quality results in less bacteria being identified.

Lab Quality is a measure of the total number of bacteria counted. The processing of a sample may detect just 30,000 bacteria or 300,000 bacteria. This impacts the number of bacteria detected and also the accuracy of the measures.

Also, Special Studies Percentage Matches is helpful to interpret these numbers better.

Criteria	BS 6/6	BS 7/19	OL 6/6	O 7/19
Lab Read Quality	2.1	5.4	2.1	5.4
Bacteria Reported By Lab	280	497	365	628
Bacteria Over 99%ile	2	7	5	6
Bacteria Over 95%ile	24	31	27	24
Bacteria Over 90%ile	49	58	49	51
Bacteria Under 10%ile	17	62	18	60
Bacteria Under 5%ile	5	30	10	28
Bacteria Under 1%ile	0	13	1	7
Rarely Seen 1%	0	4	0	9
Rarely Seen 5%	4	18	8	40
Pathogens	15	25	19	28
Outside Range from JasonH	4	4	7	2
Outside Range from Medivere	17	17	16	16
Outside Range from Metagenomics	7	7	7	7
Outside Range from MyBioma	9	9	14	14
Outside Range from Nirvana/CosmosId	22	22	23	23
Outside Range from XenoGene	6	6	11	11
Outside Lab Range (+/- 1.96SD)	6	13	10	14
Outside Box-Plot-Whiskers	70	84	64	61
Outside Kaltoft-Moldrup	70	113	112	182
Condition Est. Over 99%ile	1	1	0	0
Condition Est. Over 95%ile	2	4	0	0
Condition Est. Over 90%ile	5	6	2	2
Enzymes Over 99%ile	3	10	13	15
Enzymes Over 95%ile	46	32	69	82
Enzymes Over 90%ile	90	51	155	411
Enzymes Under 10%ile	102	219	55	138
Enzymes Under 5%ile	45	132	22	67
Enzymes Under 1%ile	6	47	5	2
Compounds Over 99%ile	9	7	104	126
Compounds Over 95%ile	56	76	385	397
Compounds Over 90%ile	292	313	533	548
Compounds Under 10%ile	72	125	183	248
Compounds Under 5%ile	39	64	109	127
Compounds Under 1%ile	5	21	16	17

What are the key things seen above (most of the numbers are similar):

• Sample Quality are the same (expected from using the same FASTQ file)
• Ombre reports more bacteria
• Outside Range from Jason Hawrelak show a major improvement with Ombre Labs and no change with BiomeSight
  • As a historic notes, Jason’s numbers were developed using uBiome labs (adding more fuzziness to everything).
  • I view this major improvement per OmbreLab, to indicate the person’s improvement.
• For Enzymes we see more high production rates and less low production rate with Ombre
  • Remember that enzymes are estimated based on the bacteria reported and is an estimate only.
  • The percentiles for both Ombre and BiomeSight are based on other samples from the same lab (they are NOT intermixed – I removed that earlier this year)
• For Compounds, we see the same thing!

KEGG Computed Enzymes

I was curious what the top items were. Most of the bacteria are the bacteria only available in Equilibrium and PrescriptAssist, excluding those and looking at the top few — we see similar suggestions (and note E.Coli is not always #1 for ME/CFS people on all tests, just a frequent pattern what dates back to 1998 in some conference papers from Australia).

• BiomeSight 6/6: Escherichia coli
• BiomeSight 7/19: Escherichia coli
• Ombre 6/6: Streptomyces venezuelae, Escherichia coli
• Ombre 7/19: Escherichia coli

Special Studies Numbers

Only BiomeSight was used in the Special Studies (because of higher sample population). The person’s rating for each of the symptoms (2 – worst, 0 -none) is also added.

Why did the number increased so much? Look below at Lab Sample quality! We cannot pick a percentage match as being critical — because that percentage depends very much on lab quality!

BS 6/6	BS 7/19	Person	Symptom
2.1	5.4		Lab Quality
13	25	2	Allergies And Food Sensitivity
13	20	2	Bloating
11	24	2	Brain Fog
9	34	2	Depression
13	23	2	Easily irritated
8	22	2	General Fatigue
11	23	2	High Anxiety
12	20	2	Histamine or Mast Cell issues
13	23	1.5	Chronic Fatigue Syndrome (CFS/ME)
11	20	1.5	irritable bowel syndrome
13	23	1.5	ME/CFS with IBS
12	30	1	Alcohol intolerance or Medication sensitivities
10	23	1	Intolerance of Extremes of Heat and Cold
9	17	1	Post-exertional malaise
21	30	1	Small intestinal bacterial overgrowth (SIBO)
12	28	1	Unrefreshed sleep
16	28	0	Allergic Rhinitis (Hay Fever)
23	28	0	Autism
12	22	0	Cold Extremities
15	20	0	Constipation
21	29	0	COVID19 (Long Hauler)
26	45	0	Inflammatory bowel disease
8	23	0	ME/CFS without IBS
11	21	0	Poor gut motility
8	20	0	Tinnitus (ringing in ear)

Intrepretation

As cited in the introduction, the person reported feeling better. We also see a major improvement against Jason Hawrelak Criteria for a healthy gut (using Ombre numbers). With both labs we see an increased of rarely seen bacteria — which is open to many interpretations; statistically both increases looks like a move towards a typical gut. 5% of 628 bacteria is 31, we see 40.

Going forward

I am building a consensus report from the items marked 2 above using the Special Studies. The list is similar to other people with ME/CFS. We see 2 E.Coli probiotics (symbioflor 2 e.coli probiotics, colinfant e.coli probiotics) at the top with d-ribose (a sugar used by E.Coli). This is then followed by the earth based probiotics( General Biotics Equilibrium, Prescript Assist (Original Formula), Prescript Assist (2018 Formula)).

These suggestions agrees with the top KEGG suggestions (despite being calculated in a totally different way — one set used Genomics and one set used Clinical Trials)

The rest of the to take probiotics mainly fall into 3 groups: Saccharomyces boulardii (probiotics) bacillus (probiotics), bifidobacterium with bacillus coagulans (probiotics) being the top of this set. As is typical, lactobacillus is usually a negative.

Going over to vitamins, the strongest take is Ferric citrate. We have almost all of the B-vitamins being strong avoid — this is contrary to the conventional treatment wisdom which says vitamin B helps ME/CFS. I discuss this in a prior post and speculate that the reason that Vitamin B is low in blood test ME/CFS is that part of the microbiome dysfunction are bacteria that are greedy for vitamin B, hence it does not get to the body. Conceptually this speculation is testable with a lab reactor using the microbiome from a ME/CFS person.

Bottom Line

“This is too complicated” is what I can hear some people saying. This analysis digs into the nature of the data which is really not needed for most people. It is likely of interest to those treating microbiome dysfunctions as it illustrates many of the challenges in interpreting.

For most people, the process stays the same:

• Upload the data
• Try several different ways of generating suggestions
• Look at the consensus

Why is consensus important? Simple, we have very incomplete data and also have limited accuracy with the microbiome tests. Going the consensus approach is similar to using a Monte Carlo Simulation, an appropriate approach to deal with complex processes with many parameters that are fuzzy.

This person has been using microbiome prescription to reduce the symptoms with success and with objective measurements of improved microbiome. His MD is willing to prescribe antibiotics and the top three items (from hundreds possible) are all used by ME/CFS specialist — indicating that the model is in agreement with clinical experience of ME/CFS specialist (a.k.a. Cross-Validation).

This is a follow up to these prior posts:

• Dec 30, 2021. Rosacea, Circulation and mild CFS.
• Apr 27, 2022 Follow up Microbiome Analysis from a prior post

Why Follow Up Posts are important

The first item is simple, does the model and suggestion appear to work. Everything is theoretically computed. The second item is that encourages people to try suggestions

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.

Comparisons between Samples

First, I do not know the best way to compare samples — what I usually do is put all of the numbers side by side. Special attention needs to be paid to Lab Read Quality. A poorer read quality results in less bacteria being identified.

Lab Quality is a measure of the total number of bacteria counted. The processing of a sample may detect just 30,000 bacteria or 300,000 bacteria. This impacts the number of bacteria detected and also the accuracy of the measures.

Criteria	8/31/2021	12/3/2021	3/25/2022	8/11/2022
Lab Read Quality	7.8	3.6	6.2	5.5
Bacteria Reported By Lab	461	379	479	383
Bacteria Over 99%ile	7	5	3	3
Bacteria Over 95%ile	20	24	11	13
Bacteria Over 90%ile	32	40	21	23
Bacteria Under 10%ile	283	123	237	189
Bacteria Under 5%ile	222	66	143	107
Bacteria Under 1%ile	161	9	44	23
Rarely Seen 1%	3	2	14	7
Rarely Seen 5%	9	7	33	14
Pathogens	37	30	44	31
Outside Range from JasonH	4	4	7	7
Outside Range from Medivere	15	15	15	15
Outside Range from Metagenomics	6	6	8	8
Outside Range from MyBioma	7	7	7	7
Outside Range from Nirvana/CosmosId	18	18	23	23
Outside Range from XenoGene	5	5	7	7
Outside Lab Range (+/- 1.96SD)	14	9	6	8
Outside Box-Plot-Whiskers	41	58	38	33
Outside Kaltoft-Moldrup	211	100	123	111
Condition Est. Over 99%ile	0	0	0	0
Condition Est. Over 95%ile	4	3	1	1
Condition Est. Over 90%ile	9	6	5	7
Enzymes Over 99%ile	17	19	30	10
Enzymes Over 95%ile	105	82	219	68
Enzymes Over 90%ile	139	126	296	183
Enzymes Under 10%ile	783	369	514	645
Enzymes Under 5%ile	542	186	264	423
Enzymes Under 1%ile	271	37	49	86
Compounds Over 99%ile	33	28	62	47
Compounds Over 95%ile	140	127	231	254
Compounds Over 90%ile	346	307	298	338
Compounds Under 10%ile	310	227	297	308
Compounds Under 5%ile	211	111	224	173
Compounds Under 1%ile	132	47	67	65

The next table is also very dependent of Lab Read Quality. The apparent improvement on 12/3/2021 is likely artificial because the counts are low due to low read quality.

	8/31/2021	12/3/2021	3/25/2022	8/11/2022
Percentile	Genus	Genus	Genus	Genus
0 – 9	73	24	51	51
10-19	15	18	32	24
20 – 29	12	13	18	12
30 – 39	4	10	9	14
40 – 49	6	8	9	3
50 – 59	4	8	7	2
60 – 69	4	4	9	3
70 – 79	7	10	7	10
80 – 89	7	4	8	5
90 – 99	14	18	8	8

	8/31/2021	12/3/2021	3/25/2022	8/11/2022
Percentile	Species	Species	Species	Species
0 – 9	87	29	57	58
10-19	24	21	29	24
20 – 29	14	15	21	16
30 – 39	10	16	14	14
40 – 49	2	6	14	3
50 – 59	12	9	17	10
60 – 69	9	10	10	7
70 – 79	8	9	14	7
80 – 89	7	15	5	4
90 – 99	11	13	10	9

So how to interpret this wall of numbers? People can cherry-pick the numbers to say improvement or no improvement. The difference of lab read quality is a big factor because they impact the count for most of the items above. The Outside Box-Plot-Whiskers numbers show continued improvement. In short, the changes shown were less than I was hoping to see.

There is one more method of comparison — using special studies. In this case we see the average matches. Doing a little math, the expected drop of percentage due to lab quality size between 8/31/2021 and 8/11/2022 is a 10% drop. Those that exceeded 20% are color with 😊 below. Nothing became 10% worse. Note that the 😊 also agrees with comparing to 3/25/2022 (the prior sample). Other items remained unchanged. Items reported by this person are 😧 – Strong issue, 😟 – a bit of an issue

Study	8/31/2021	12/3/2021	3/25/2022	8/11/2022	Average
Inflammatory bowel disease	49	32	47	47	43.8
Small intestinal bacterial overgrowth (SIBO) 😟	51	29	48	34😊	40.5
Allergic Rhinitis (Hay Fever) 😧	41	27	45	41	38.5
Autism	45	34	44	27😊	37.5
COVID19 (Long Hauler)	40	24	43	29😊	34.0
Irritable bowel syndrome 😟	40	22	44	30😊	34.0
Alcohol intolerance or Medication sensitivities	43	23	39	29😊	33.5
Histamine or Mast Cell issues 😧	44	17	43	24😊	32.0
Post-exertional malaise 😧	36	23	40	29😊	32.0
ME/CFS without IBS 😟	39	22	36	30😊	31.8
Poor gut motility	42	21	40	24😊	31.8
Brain Fog 😧	37	23	33	33	31.5
Depression	32	27	32	34	31.3
Allergies And Food Sensitivity 😧	38	23	36	25😊	30.5
Cold Extremities 😧	42	23	33	22😊	30.0
Intolerance of Extremes of Heat and Cold 😟	38	16	39	27😊	30.0
ME/CFS with IBS 😟	36	21	36	27	30.0
Bloating 😧	37	22	34	26😊	29.8
General Fatigue 😧	34	20	31	34	29.8
Unrefreshed sleep	31	22	36	28	29.3
Constipation	37	21	29	25😊	28.0
High Anxiety 😟	33	17	33	29	28.0
Easily irritated 😟	32	16	34	23😊	26.3
Tinnitus (ringing in ear) 😟	24	15	29	22	22.5
Chronic Fatigue Syndrome (CFS/ME) 😧	25	21	20	23	22.3
Average	37.8	22.4	37.0	28.9	31.5
Lab Quality	7.8	3.6	6.2	5.5

What is my conclusion? Most of the measures above deteriorates into noise with the exception of data from Special Studies, where we seen improvement in many measures, but not all. In one real statistical sense this makes sense: many are based on common sense and the ones showing clear improvement on statistical significance.

Going Forward

For most of my prior posts used the logical reasoning and clinical studies (which used different labs and software than the samples that I was looking at). With the special studies, we have upped our game (potentially) – the bacteria deemed significant were determined by the same lab and software of our sample, plus the study sizes was much larger than published clinical studies — hence better detection.

To build the consensus I will use the special studies, I filtered to reported issues and high percentage of matches, namely:

• Allergic Rhinitis (Hay Fever)
• Histamine or Mast Cell issues
• Post-exertional malaise
• Brain Fog
• Allergies And Food Sensitivity
• Cold Extremities
• General Fatigue

Remember that most of the special studies found that infrequent bacteria with a low value was what was statistically significant. This is turning the usual logic on it’s head. As I state, this is all experimental but based on studies and statistics.

The top suggestions are below

• navy bean
• Pulses ( dry peas, beans, lentils, chickpeas, etc)
• arabinoxylan oligosaccharides (prebiotic)
• non-starch polysaccharides
• red wine
• wheat bran
• proton-pump inhibitors (prescription)
• l-citrulline
• xylan (prebiotic)
• pea (fiber, protein)

Antibiotics

As expected, most antibiotics and prescription drugs are to be avoided. A few with positive impact includes:

• macrolide ((antibiotic)s) – used by Dr. Cecile Jadin in treating ME/CFS
• sulfonamide (antibiotic)s – suggested for treated ME/CFS [2019]

In terms of generic suggestions, rifaximin (antibiotic)s is by far the top antibiotics, cited here on Health Rising: Rifaxamin – citing use by Dr. Teitelbaum, Dr. Peterson, De De Meirleir and Dr. Myhill (all ME/CFS specialists).

In short, all of the top suggested antibiotics are applicable. My personal approach would be do all three of them in a pulse manner a la Jadin, 10 days on, 20 days off and then move to the next one.

Both above and generic suggestions have proton-pump inhibitors (prescription) being the top choice for other prescription drugs.

Probiotics

The top probiotics list have the usual dilemma: both e.coli probiotics and lactobacillus probiotics. It’s a dilemma because they tend to be hostile to each other. My typical rotation resolution would be 2 weeks of each and then move to the next:

• saccharomyces boulardii (probiotics)
• e.coli probiotics (take it with d-ribose, on the to take list and a food for e.coli)
• bifidobacterium
• lactobacillus
• Equilibrium and/or Prescript Assist

I should note that some are strong to be avoided (watch out for mixtures!!!)

KEGG Suggestions

The KEGG suggestions top items were the bacteria found in Equilibrium and Prescription Assist, except for the top choice, Escherichia coli. A probiotic suggested by Dr. Myhill, a ME/CFS specialist in the UK. The next common conventional items are

Bottom Line

The suggestions above were done solely from special studies. The key question is are they reasonable? I would say yes based on the antibiotics suggestions — all of them have been reported to help ME/CFS patients. We also have agreement between KEGG probiotics and these suggestions.

There is a potential conceptual symmetry between the two approaches (working off extremes and using special studies that are often dealing with rare low bacteria). Bacteria influences each other in very complex ways.

The full list of suggestions is available above.

I am hoping this will be a model for other Long COVID people to start the recovery process. This person used Biomesight. Those results allow the data from special studies to be used on his microbiome sample.

A word of warning, tests like GI-MAPS will not report on most of the bacteria found to be low in the Special Studies — you need much more detail reports!

Suggested Parallel Reading: CFS Patient after COVID using the Special Studies Results

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.

Backstory

COVID in February 2021. 37M at the time, athletic/fit. Crossfit x 3 a week, playing football weekly Only mild gastritis prior to COVID. No other health issues. Moderate severity Covid, lots of symptoms.

And then Long COVID and CFS/ME type of symptoms mostly fatigue, PEM and GI problems (pain, food intolerance, bloating..etc) I’d say it’s a moderate/mild case of CFS/ME. But after 18 months still not back to previous levels, can’t walk too long otherwise i crash. I’d say i am around 75%.

High Level Overview

Looking at Health Indication, we find no significant medical conditions flagged (consistent with prior life style). There is one bacteria of potential concern: Prevotella copri, accounting for a whooping 56% of the microbiome! It is interesting that this was also seen in another recent review, see CFS Patient after COVID using the Special Studies Results. In terms of Dr. Jason Hawrelak Recommendations – he’s at the 99.7%ile — extremely healthy!

Using Special Studies

Interpreting the updated table shown below can get a little complicated (i.e. not naively simple) see Special Studies Percentage Matches for details

We are going to use the 7 items below – items matching his reported issues. In an independent study that I did, I found that the pattern dims over time as the microbiome evolves. His person is 20 months post-COVID.

• COVID19 (Long Hauler)
• Small intestinal bacterial overgrowth (SIBO)
• ME/CFS with IBS
• Inflammatory bowel disease
• Post-exertional malaise
• General Fatigue
• Bloating

The Prevotella copri concerns me because it’s the mastodon in the room (bigger than an elephant, and a bit hairier!). This specific bacteria is NOT typical for long COVID, but I suspect many will find one or another tyrant to dominate in excess in the face of massive minority representation– hence check for high bacteria counts with high percentile. It was also high in the study cited above, CFS Patient after COVID using the Special Studies Results. I went thru the My Biome View to tag the ones that have a high percentile with with a large count. The purpose is to inhibit these, so they will not inhibit everything else.

The results were almost the opposite of the consensus below for B-Vitamins. It presents a dilemma, a choice that needs to be made. At the moment, I favor the working from the special studies approach (pending feedback from people who tried it). Conceptually, it is a more probable approach — incidentally, it is not the approach usually done (and those approaches, historically, have had very little success to date).

The Consensus

I did not want to toss in any more sets of suggestions. From the start we saw the dominate item in his microbiome was undergrowth of a multitude of bacteria and the domination of one — we have gotten what helped the weak and inhibits the strong.

I found the avoids to be an interesting combination, no red meat and no chicken (matching Reduced choline on the to take) .

We see that something like a B-Complex should be avoided. I discuss this issue more in the other blog post that I cited above.

Computed Probiotics from KEGG Enzymes

This produced a few items that are reasonably easy to get as single species probiotics. Remember, these are calculated by a totally different mechanism – using the genes of the bacteria in your microbiome and the genes in these bacteria. The top items were:

• Escherichia coli (which is near the top of our to take list)
• Bacillus subtilis (on our to take list) and other similar bacillus species
• Akkermansia muciniphila (which shows up once, with a positive value (74.5) on our consensus)
• Clostridium butyricum (a minor avoid on the consensus with 3 for and 4 against)
• Lactobacillus plantarum (a minor avoid on the consensus with 3 for and 4 against)

My basic take-away are using just 3 probiotics in weekly or fortnight rotations (one at a time)

• Escherichia coli (Mutaflor or Symbioflor-2)
• Bacillus subtilis many sources
• Akkermansia muciniphila (available only in the US unfortunately)

If one of the above cannot be obtained, I would suggest using Clostridium butyricum or Lactobacillus plantarum as the third element in the rotation.

Supplements suggested by KEGG

Although this is using an old algorithm that I have not updated, the list is below.

• alpha-galactosidase (Enzyme) – Percentile: 11
• Amylase (Enzyme) – Percentile: 8 – On Consensus: Take
• beta-alanine – Percentile: 2 – On Consensus: Take
• Glycine – Percentile: 4 – – On Consensus: Minor avoid
• iron – Percentile: 7 – On Consensus: Take
• L-Cysteine – Percentile: 3 – On Consensus: Major avoid
• L-glutamine – Percentile: 14 – On Consensus: Major avoid
• L-Histidine – Percentile: 12 – On Consensus: Take
• L-methionine – Percentile: 10 – On Consensus: Major avoid
• L-Serine – Percentile: 11 – On Consensus: Take
• L-Threonine – Percentile: 16
• magnesium – Percentile: 4 – On Consensus: Take
• NADH – Percentile: 4
• Selenocysteine – Percentile: 4
• zinc – Percentile: 16 – On Consensus: Take

Remember we are dealing with fuzzy data, my usual rule is do positive stuff where there is universal agreement, avoid stuff that are negative or where there are contradictions (I do like playing dice with my health).

Bottom Line

Because of the special studies and this person using the appropriate lab, this was actually a simple analysis to do. The traditional analysis showed “nothing wrong”, a familiar restrain from medical professionals to Long COVID patients. Our special studies and distribution by percentile showed things are wrong. Having 56% of the bacteria being Prevotella copri is saying something is very wrong.

I often try to use analogy of human populations to explain what I see. In this case, we have dozens of small tribes battling each other allowing a dominating force to seize most of the space. There are many historic examples, often under the name of “Divide and Conquer”.

In this example, the high number of low representation bacteria we saw in the overview matched the high number of low number of bacteria we observed in our special studies.

After two months of trying the suggestions, I hope this reader will do a new sample to see how well things shift from these suggestions.

Questions

Q: “Excuse me if I’m missing something but is there any reason why we are focusing on only Commensals,  Prevotella, why not on Probiotics at all?  I understand it’s way above the range, and it’d like to keep it low ideally, but what about the rest of Microbiome?”

A1: First “the rest of the microbiome” issue – the obvious response is a simple “If it is not broken, don’t fix it”. The above analysis used over 100 different bacteria. Our focus is on the bacteria where there is significant statistically evidence that they are connected to Long COVID. The numbers above are general health. As cited above, with Dr. Jason Hawrelak General Health Recommendations you are better than 99.7% of people. There is a huge variation in recommended ranges coming from labs and specialists — who are you going to rely upon? I am a statistician and I follow the numbers (and the z-scores), in other words, not working off opinion based largely on treating people who do not have Long COVID. I am NOT focusing on commensals, I am focusing on what was shown to be statistically significant.

A2: “why not on Probiotics at all” — Excuse me, I name four key probiotics: Escherichia coli (Mutaflor or Symbioflor-2 are retail products), Bacillus subtilis et al (microbiome labs/ megasporebiotic looks like a good commercial choice), Akkermansia muciniphila (Pendulum) and Clostridium butyricum (miyarisan). You will find alternative brands for some using the probiotic page.

If you mean bifidobacterium and lactobacillus probiotics — they are not indicated in general, in fact, they often appear in the to-avoid list. Example, Long COVID often has brain fog, see this study: Brain fogginess, gas and bloating: a link between SIBO, probiotics and metabolic acidosis [2018] which calls out those two as contributors.

Comment: To answer your question, I’ve had lots of symptoms in the beginning, but for now only mild fatigue and PEM plus gut issues, so I’d say definite ME/CFS with IBS, some Rhinitis, Alcohol intolerance and Long hauling.

I have recently changed the display below to show the percentage of matched instead of just the number of bacteria matches (the number will appear if you hover over the link as a tool tip). The numbers may be prone to misinterpretation, hence this technical page.

The candidate bacteria comes from special studies — it is important to note that often these bacteria are rarely seen, so having a 100% match is effectively impossible. We also have the dilemma of a single sample versus a collection of samples.

The rule that I am using is simple, a match must:

• Have the bacteria (if it is missing, it is not deemed a match)
• The bacteria count must be either:
  • below the study mean – 3 standard deviations of the mean if the study found it to be a low mean value against the reference population
  • above the study mean + 3 standard deviations of the mean if the study found it to be a high mean value against the reference population

Naively, assuming a normal distribution, the odds of a single match is around 1%, so with 200 items to check, we would expect 1% for a random person.

You should NOT view these as predictive, for example both ME/CFS with IBS and ME/CFS without IBS are on the list with the same value!!! Instead, your existing condition(s) should be used to select only the ones that apply to you. You could arbitrarily do all of the high ones — I do have a concern about that approach, you are creating noise that may make suggestions less effective.

One last item is the quality of the read (i.e. how many bacteria was actually detected in the sample). Since we are dealing with rare bacteria, bacteria (that are actually there) may not be detected and thus you have a lower percentage match. So do not view the percentage as absolute. but relative to others in the sample.