The information below comes from Integrated analysis of gut microbiome and host immune responses in COVID-19 [2022]. While the data comes from COVID patients, there is a good chance that it may apply to other conditions. The items below are for the strongest p-values (most likely). My intent is to add all items with a value with a p Value of 0.05 or less as an experimental page, before creating the page I hope people will find similar studies that can be combined with this data. Please add as comments.
Gut_microbes
Blood_clinical_features
Correlation
Pvalue
Coprococcus_comes
CD8_counts
0.73173549
1.95E-06
Coprococcus_comes
CD3_counts
0.714258257
4.41E-06
Coprococcus_comes
Lymphocytes_counts
0.649787455
5.71E-05
Coprococcus_comes
CD45_counts
0.647009338
6.29E-05
Roseburia_intestinalis
CD8_counts
0.63669604
8.94E-05
Roseburia_intestinalis
CD3_counts
0.619860729
0.00015457
Roseburia_intestinalis
CD45_counts
0.611747625
0.000199038
Streptococcus_oralis
Eosinophil_counts
0.603056198
0.000258997
Gut_microbes
Organ_damage_related_factors
Correlation
Pvalue
Akkermansia_muciniphila
Creatine_kinase_isoenzymes
0.889522704
1.00E-11
Bacteroides_cellulosilyticus
Creatine_kinase_isoenzymes
0.88177523
2.63E-11
Streptococcus_oralis
Gamma_glutamyltransferase
0.788443186
8.39E-08
Akkermansia_muciniphila
Aspartate_Transaminase
0.784368908
1.08E-07
Bacteroides_cellulosilyticus
Aspartate_Transaminase
0.766075518
3.22E-07
Gut_microbes
Cytokines
Correlation
Pvalue
Ruminococcus_gnavus
IL17
0.513431966
0.002652723
Klebsiella_pneumoniae
IFNG
0.484801301
0.004922183
Klebsiella_pneumoniae
IL10
0.431582818
0.01364772
Lachnospira_eligens
IL2
0.393008018
0.026072229
Gut_microbes
Coagulation_factors
Correlation
Pvalue
Enterococcus_durans
APTT
0.358216937
0.044101648
Enterococcus_faecium
APTT
0.357152676
0.044780215
APTT is activated partial thromboplastin time, the time to form blood clots
I started Microbiome Prescription site using data uploads from ubiome, a firm that was founded by a crowdfunding campaign, went to venture capitalists, and went unethical due to pressure from venture capitalists and died. I received over 800 samples processed by ubiome.
Readers started to request the microbiome reports to be processed on the Microbiome Prescription site and I started adding them according to constraints of the reports available. BiomeSight.com, a UK firm, has been the most cooperative. We worked together to allow automatic transfers directly from their web site to the Microbiome Prescription site by using a API.
Xenogene | Metagenómica y Biología Molecular Reports were shared to me. I found that I could do an accurate extract from one of the reports they made available to users. The result was that they became the most comprehensive report as see by the statistics below
There were few uploads because of their higher costs. The report that I used is shown below.
They Changed Their Report
Recently I have had two people trying to upload their reports. The report was different than the above. I asked them to contact Xenogene to get the above; they were not successful. I examined their reports (they were several years apart), and found two different formats, as shown below:
While both give the same information, the structure of the page was different. The report do not give the hierarchy, for example, Eubacteriaceae was found in neither report. I looked for Blautia and could find species and strains — but no total, so you cannot apply Dr Jason Hawrelak criteria.
Summing up all of the species and strains under Blautia does not give a correct total, in some cases the Blautia total will be 2x higher. Why, because many strains and species has not received names and / or “fingerprints”.
Example of the synthesis to higher levels… Issues at the species level and below can be identified
Bottom line: I no longer recommend this lab
Despite these issues, I have updated the import to support both of the above PDF formats and synthesize all of the missing layers of the bacteria hierarchy. The new import should be on line by November 13th.
WARNING: the genus level and above may often be low because of the total synthesis.
I did extract their recommended ranges and added it as an option:
Citizen Science Action
I would suggest emailing them and asking them if they make a CSV file available of the results (including the bacteria NCBI taxon number), if so, can you get a sample. If you do not get a positive results, do a return email asking you to be informed if they change and indicating that you are going to use Ombre or Biomesight instead…… The risk of loosing customers can often change business practice.
I emailed about six months ago with questions. Since then I’ve attempted an evaluation of my microbiome’s needs through a thorough look at your site’s AI suggestions. I attempted to implement some of those, with my primary care’s approval. I didn’t have much luck and I was looking to have another go at it, with a fresh Ombre analysis. I’m formally requesting a review for an educational post. Before I jump back into trying the AI’s suggestions again, I thought it would be foolish if I didn’t seek out the assistance of the one who designed it. I understand there are others in line and you may decline the request, but I appreciate that you are offering this to people. Hope is such a necessary lifeline with CFS.
It feels a little odd to be giving such detailed information about myself without having a firm “go ahead” from you. From what I understand though, you want all the info before you’ll consider the request. I’ll attempt to make it brief. I have also uploaded my symptoms to the website. I consent to your use of my information.
Symptoms
Fatigue
Exaggerated loss of muscle strength with exertion
Brain fog
Trouble reading and comprehending
Post exertional malaise
Constipation
Panic Attacks
-Intolerance to any probiotics, *DAIRY*, caffeine, alcohol, refined sugars, and a growing list of fruits and vegetables.
-Whatever herbs I try there is almost always an initial benefit, but then things go back to the way they were. (Possible bacterial adaptation?)
Diagnosed
“Chronic Lyme”
ME/CFS
IBS-C
Panic Attacks
Depression
~ Backstory ~
Dec 2012 – I had been working multiple full time jobs while eating very poorly. Essentially fasting, and what I did eat was high carb, low nutrient foods. No fruits, vegetables, or other nutrient rich items. I felt something snap in me in an instant. I felt panicky and went and ate a large meal immediately. From that moment on I’ve suffered from CFS. The most prominent features being fatigue, exaggerated loss of muscle strength with exertion, brain fog, trouble reading and comprehending, post exertional malaise, and constipation. My symptoms were the most exaggerated at that time, although recently they have started to get back to that point in time. Examples: I would eat a carb rich food like pasta and 30 minutes later I would literally be on the floor in a quasi lucid state; Two and a half weeks without a BM. My primary care at the time put me on antidepressants and thyroid medications, which did nothing for me.
March 2016 – I started to see an integrative medicine MD who thought that I had reactivated Lyme. He reasoned that it was dormant in my system from when I had it as a 5-6 year old, and that the stress allowed it to manifest itself again. I was on an absurd amount of supplements and various antibiotics. I found initial improvement that I felt stopped my decline and helped with some symptoms, but didn’t solve the CFS. The one drug that I felt the best on was Tinidazole. I stopped seeing him in 2017.
July 2017 – I stopped working as my symptoms had continued to get worse over the years. I started taking care of my sister who was diagnosed with terminal brain cancer. She died in 2019.
July 2018 – Started to see a new primary care. He confirmed the CFS diagnosis but refuses to help in any way. He encourages me to seek solutions on my own however.
Dec 2018 – Started to see another Lyme specialist who told me about the herbalist Stephen Buhner. I bought his book and attempted a slew of his proscribed herbs over the course of a year or two, with little benefit. He also put me back on doxycycline for 6-12 months. It did help but just by taking the edge off of my symptoms.
Fall 2019 – I started to develop panic attacks. It was clear to me that stress made them worse but that it was primarily an issue stemming from a physical problem, rather than an emotional one. This was evinced by the fact that certain foods could manifest them. I felt that whatever my problems were, they were physical and were slowing progressing.
Summer 2021 – I started looking into gut health as a cause of CFS. I started the Wahl’s diet and found some improvement through that. It afforded me enough strength to go back to work for six months. I never entered fully onto the diet as it was prohibitively expensive, but I did keep some of the foods that helped.
2021 – I started to see a gastrointestinal dr. who was absolutely no help. He diagnosed me with IBS-C but didn’t have any answers or solutions. He did proscribe Amoxicillin, which I declined to take as I was not sure if it would help or hurt my gut bacteria.
April 2022 – I found your website, uploaded an Ombre test, and attempted some of the suggested herbs. I found initial benefits from cinnamon but any long-term attempt at any of the herbs is really a trial (and I’m not one to back down from a fight or a stranger to discomfort). I continued on some of them until I thought there might actually be a chance of dying from it. I just couldn’t tell if it was herxing or hurting.
Spring 2022 – Developed food sensitivities, primarily to dairy. Dairy gives me intense psychological issues. The best I can describe them is that they are like racing thoughts accompanied by the feeling that my head will explode and there is no way to escape. Food that once helped, like carrots and berries, now make my intestines feel like overinflated balloons: a lot of pain.
August-October 2022 – I started to care for my mother who was diagnosed with terminal stomach cancer. She died and the stress of it has amplified my already mounting symptoms to a fevered pitch. It almost feels like when things started back in 2012.
This story is unfortunately very typical for many people. My wish is that Microbiome Prescription will be able to help. I do not have “the cure”; what is generated are suggestions (many — so pick and choose what works for you), items modelled to have better than random impact on the microbiome.
First Look at the sample
We have two Ombre samples on the account:
May 2,2022
Oct 18,2022
With two samples from the same lab, my first step is typically to compare them. I omitted the KEGG data which was not illustrative of changes.
Criteria
Old Sample
New Sample
Lab Read Quality
4.2
5.6
Bacteria Reported By Lab
561
677
Bacteria Over 99%ile
2
7
Bacteria Over 95%ile
18
21
Bacteria Over 90%ile
38
43
Bacteria Under 10%ile
92
121
Bacteria Under 5%ile
37
62
Bacteria Under 1%ile
3
11
Lab: Thryve
Rarely Seen 1%
5
8
Rarely Seen 5%
23
54
Pathogens
40
30
Outside Range from JasonH
3
3
Outside Range from Medivere
14
14
Outside Range from Metagenomics
9
9
Outside Range from MyBioma
12
12
Outside Range from Nirvana/CosmosId
21
21
Outside Range from XenoGene
8
8
Outside Lab Range (+/- 1.96SD)
9
40
Outside Box-Plot-Whiskers
67
128
Outside Kaltoft-Moldrup
136
244
Condition Est. Over 99%ile
0
0
Condition Est. Over 95%ile
0
0
Condition Est. Over 90%ile
0
1
My read is that between the samples, the person has gotten worse. Why?
Outside Lab Range (+/- 1.96SD), Outside Box-Plot-Whiskers, Outside Kaltoft-Moldrup all have very significant increases,
This is also reflected in Bacteria Over ??%ile and Under
Not having any strong matches to (PubMed Studies) Conditions is unusual. It suggests that the compounding of issues results in the microbiome not falling into any established “box”.
We see that also with the distributions, a massive surge of under-represented bacteria (0-9)
While he attempted suggestions after the first sample, we see a mountain of microbiome changing events also occurred (especially stress which, for me, has been very significant cause of my own historic dysbiosis). Whether the suggestions helped or hurt cannot be determined.
Approach
Building a consensus fromLab Range, Box-Plot-Whiskers and Kaltoft-Moldrup seems the best approach.
Early Sample
From the consensus we see a list which agrees with what is often reported as helping ME/CFS from the earlier sample.
From Earlier Sample
Some illustrations from the literature of the suggestions
What I did above is called, cross-validation. This means checking if the suggestions generated by the model agrees with clinical experience. It does. This implies that items not seen in studies (like a grapefruit for breakfast) seems likely to have positive effects.
Cross validation is always a good step after getting suggestions. The suggestions using this trio of methods to select will mostly be good — but odd cases may produce poor results.
Example of how to do it
Latest Sample
Given the stress etc. I know that the microbiome will shift and may not be so easy to cross-validate. We see many of the same things, they have just rearrange themselves.
We have the simplified suggestions (shown above) with the to-take probiotics being:
bifidobacterium (animalis) lactis
lactobacillus gasseri
bacillus coagulans (with bacillus subtilis being a very strong avoid).
KEGG suggested Escherichia coli Probiotics — which is to be expected from ME/CFS. A low level of Escherichia coli has been reported in the 1999 Australian Conference papers.
My suggestion would be ONLY bifidobacterium (animalis) lactis (Custom Probiotics has it available as a single species without additives) and E.Coli (i.e. Symbioflor-2 )
The last version of suggestions is a food list derived from flavonoids etc in food. It is an experimental exploration (so a grain of caution is suggested).
The only thing that was positive was Barley which is on the avoid list. So nothing (safe) useful from this experimental method.
As a FYI, I checked for items that are adaptogens (helps with stress) and the following were on the to-take list.
The suggestions often have numbers beside them. The numbers are relative numbers for things in the same list. In simpler words:
One in Metric, – meters
One is Imperial/American, miles
One is Roman, league
One is nautical, knots
One is astronautical – parsec
They cannot be compared to each other. The goal of each list is find the best given the approach.
Postscript – 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.
I use modelling and various mathematical technique to estimate forecasts when there is no hard data available.
This person did his tests using OmbreLabs.com and then transfer the data to biomesight.com.
I did the special studies for a couple months taking symbio, florastor, GOS,Arabox., d ribose, pea fiber, etc.
Feel in general more energized, specially after the round of florastor which I had done just a four days then tested at the time so impact probably won’t fully show in this test. Where to go from here? Drop special studies focus or stay the course?
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 will start by adding new columns for the latest sample. The person measured with Ombre and then transferred data to Biomesight to get a second interpretation of the raw digital data.
Criteria
BS 6/6
BS 7/19
BS 11/22
OL 6/6
O 7/19
O 11/22
Lab Read Quality
2.1
5.4
4.2
2.1
5.4
4.2
Bacteria Reported By Lab
280
497
468
365
628
473
Bacteria Over 99%ile
2
7
0
5
6
1
Bacteria Over 95%ile
24
31
2
27
24
9
Bacteria Over 90%ile
49
58
12
49
51
21
Bacteria Under 10%ile
17
62
187
18
60
87
Bacteria Under 5%ile
5
30
90
10
28
46
Bacteria Under 1%ile
0
13
22
1
7
2
Rarely Seen 1%
0
4
0
0
9
0
Rarely Seen 5%
4
18
15
8
40
9
Pathogens
15
25
39
19
28
33
Outside Range from JasonH
4
4
9
7
2
9
Outside Range from Medivere
17
17
20
16
16
20
Outside Range from Metagenomics
7
7
7
7
7
7
Outside Range from MyBioma
9
9
10
14
14
10
Outside Range from Nirvana/CosmosId
22
22
22
23
23
22
Outside Range from XenoGene
6
6
10
11
11
10
Outside Lab Range (+/- 1.96SD)
6
13
2
10
14
2
Outside Box-Plot-Whiskers
70
84
29
64
61
25
Outside Kaltoft-Moldrup
70
113
70
112
182
91
Condition Est. Over 99%ile
1
1
0
0
0
0
Condition Est. Over 95%ile
2
4
0
0
0
0
Condition Est. Over 90%ile
5
6
0
2
2
0
Enzymes Over 99%ile
3
10
1
13
15
5
Enzymes Over 95%ile
46
32
16
69
82
147
Enzymes Over 90%ile
90
51
103
155
411
405
Enzymes Under 10%ile
102
219
354
55
138
169
Enzymes Under 5%ile
45
132
154
22
67
78
Enzymes Under 1%ile
6
47
29
5
2
0
Compounds Over 99%ile
9
7
29
104
126
19
Compounds Over 95%ile
56
76
233
385
397
89
Compounds Over 90%ile
292
313
347
533
548
118
Compounds Under 10%ile
72
125
264
183
248
133
Compounds Under 5%ile
39
64
163
109
127
100
Compounds Under 1%ile
5
21
41
16
17
42
Note: I just cut and pasted from “Multiple Samples” tab to Excel to make the above table.
What do I see above?
Sample Quality are the same (expected from using the same FASTQ file)
Rare and very high bacteria have a significant improvement
All of the statistical out-of-range measures(Std Dev, Box-Plot, K/M) reduced the count significantly.
Most of the expert suggested ranges increased.
Condition profiles dropped to zero.
We see the fragileness of some measures to the software being used to interpret the raw data.
Enzymes Over… one dropped from prior and the other increased from the prior
My general opinion is that the person has improved objectively. The algorithm explicit goal is to reduce all of the statistical out-of-range measures. Ideally, it will also “fix” the person but that is more complex, we lack sufficient knowledge to hand pick the bacteria. We can get associations to specific bacteria — association is NOT causality often (despite many politicians claiming such!).
Going Forward
KEGG Computed Probiotics
Both labs resulted in the same priority: Escherichia coli at the top, the soil based mixtures, then Bacillus subtilis (I personally prefer to get it “au natural”, i.e. in the traditional Japanese Soy based food, Natto).
I am going to skip them, mainly because the results are erratic until I get a better understanding of this. See Caution: Special Studies Suggestions.
There is a possibility of both being right. Right meaning shifting from the current dysfunctional equilibrium. This could be visualize as shown below. I am seeking understanding and building different approaches. Each approach could work for some (but not others). Too many factors for certainity.
Flavonoids: Only Barley was a positive (for Ombre) , it was not on Biomesight list
Food Contents: Again, only Ombre had positive suggestions: Brazil Nuts and Olive Oil. Biomesight data disagreed on the Brazil Nuts
Why so few? Why labs contradict? This comes down to two key challenges: Different interpretation of the bacteria from the digital data; a low volume of studies on the substances we are using to build food suggestions. Also, the suggestions are based on some of the contents of the food; there are other parts of the food that will have other effects. This is why direct food, herb and spice studies are best. Every food is a complex mixture of chemicals. Some may help, some may hurt. Care must be taken to avoid the simplistic logic that “Super Breakfast Food contains barley, thus it is good/healthy to eat!” Ignoring the 10 grams of sugar in this product.
Thus, these suggestions should be taken with a grain of salt. They are better than random choices, but far short of what we would ideally like.
Bottom Line
I ran some of the Special Studies suggestions and did a download of simplified consensus. Between approaches, we had agreement on taking:
Probiotics:
akkermansia muciniphila
bifidobacterium (animalis)lactis
lactobacillus salivarius
saccharomyces boulardii
Other
Vitamin K2
Calcium
Echinacea
Omega-3 fatty acids
Pomegranate
Rutin
Tea Tree oil
As well as agreement on avoiding
fructooligosaccharides (FOS)
jerusalem artichoke
Flaxseed
Vitamin B2 Riboflavin
“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. I am trying to get better understanding. It looks at some of alternative methods of getting suggestions. It is likely of interest to those treating microbiome dysfunctions as it illustrates many of the challenges in interpreting.
For most people, the best process stays the same:
Upload the data
Try several different ways of generating suggestions, building a consensus from
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 that produces better results.
Recent Comments