We have a small number of samples annotated with this condition. From this our candidate bacteria are shown below. Almost all of the significant ones are excess counts.
This is from a special analysis identifying bacteria that are probiotics that have association to histamines and mast cell issues using the 394 annotated samples contributed.
Some species are to be avoided. Keeping to the specified species is strongly recommended,
Tax_Name
Tax_rank
Symptom Frequency
No Symptom Frequency
Suggests
Bifidobacterium adolescentis
species
33.5
48.8
Take
Bifidobacterium pseudocatenulatum
species
8.9
7.7
Avoid
Lactococcus
genus
42.9
55.6
Take
Bifidobacterium gallicum
species
18.8
27.7
Take
Bifidobacterium kashiwanohense PV20-2
strain
10.7
20.3
Take
Bifidobacterium indicum
species
27.4
39.8
Take
Bifidobacterium
genus
83.8
94.4
Take
Bifidobacterium catenulatum subsp. kashiwanohense
subspecies
13.7
21.9
Take
Bifidobacteriaceae
family
85.3
95.4
Take
Bifidobacteriales
order
85.5
95.4
Take
Lactobacillus rogosae
species
22.8
26.6
Take
Bifidobacterium bombi
species
13.7
21.3
Take
Lactobacillaceae
family
91.1
96.7
Take
Lactobacillus iners
species
10.2
14.1
Take
Bifidobacterium subtile
species
22.6
26.8
Take
Lactobacillus
genus
83.2
89
Take
Bifidobacterium thermacidophilum
species
12.2
14.7
Take
Lactobacillales
order
97
100
Take
Lactococcus lactis
species
14
18.3
Take
In terms of bacteria patterns found, the following were found statistically significant
Bacteria are like people, they interact and are influenced. The problem is how to detect the interactions that are clinically significant and the direction of interaction without grabbing stereotypes (i.e. all Italians belong to the Mafia, Irish are lazy, Egyptians are all Islamic Terrorists, etc).
From a collection of samples, we pull all samples that report both bacteria. We take these numbers and drop them into a tool like Excel. Chart the data and try to do a linear regression. This is often pro-forma in research papers because that is rote learning.
A Uniform Way
This is almost the same, except we do not use the actual numbers, but the percentile rankings. This produce stronger regressions values Using the percentile transform the data to a uniform distribution. R2 increased by 10 fold but really a long away from significance.
You can almost see signs of a trend in the middle of lots of noise.
A Non-parametric Way
We use classic Chi2. The process is simple
For bacteria A we determine the percentage with a value of 100 or higher, say 5%
For bacteria B we determine the percentage with a value of 1000 or higher, say 5%
We filter the samples to those with bacteria A being higher than 100
If there are no interactions than we expect 5% of bacteria B to be 1000 or more.
If we find that 30% of bacteria B is more than 1000, then it appears that high Levels of A results in higher levels of B
From the above we can compute a statistics,Chi2, and thus the statistical significance. In this case, very very significant.
This means that we isolate the impact of high values and low values which the earlier methods did not do, We do not know how the middle value interact but for clinical issues, it is abnormally high and abnormally low values that are of interest.
Implementation
The first question is to pick the high and low threshold values. People can pick arbitrary values and try them. I have my own preference a patent pending algorithm to produce ranges.
The number of calculations to be done were done in the following datasets with 5%ile and above, and 95%ile and above. The bigger the sample, the better sensitivity and more interactions likely to be discovered.
All: 5,191,562 possible pairs on 5189 samples –> 1,270,000+ Interactions found
Biomesight: 1,717,410 possible pairs on 2534 samples –> 275,000+ Interactions found
Ombre: 1,743,720 possible pairs on 1540 samples –> 220,000+ Interactions found
uBiome: 132,860 possible pairs on 791 samples –> 4,700+ Interactions found
For each pair of taxa we have 4 scenarios (Low versus Low, Low vs High, High vs High, High vs Low) Or about 32 million queries retrieving data sets and performing calculations. The bigger the sample size, the more items that are likely to be identified. For thresholds, we use a patent pending algorithm that appears to yield good results (shown above). The alternative would be to enumerate percentages and find the ones that work best (so 100 x 100 x 32 million = 320,000,000,000 queries).
Illustration of the code is below.
Select Sum(Case when c1.Percentile < 19.274700171330668
/* 577309 Low Percentile Threshold*/
then 1 else 0 End) Obs, /*Low Count that is filtered sample */
Count(1) Cnt, /*Filtered sample Count*/
cast(Count(1) * 19.274700171330668/100 as float) [Expected Value]
from UserCounts c1 Join Usercounts c2
on C1.sampleId=c2.sampleId
And C1.taxon=204516
And C2.Taxon=577309
Join Users U on C1.SampleId=SequenceId
Where c2.Percentile < 12.890741292051205 /* 577309 Low Percentile Threshold*/
Group by C1.Taxon,C2.Taxon
dependent
independent
Label
Obs
Direction
Expected
Chi2
%
204516
577309
L,L
158
>
79
98.5
200%
204516
577309
L,H
204
<
238
6
86%
204516
577309
H,L
138
<
202
39.5
68%
204516
577309
H,H
724
>
609
43.1
119%
Phocaeicola massiliensis
Paraprevotella
Example of results with counts
There is a question of using Chi2 or using the percentage increased or decreased.
Example:
High Paraprevotella, we get more high count and less low count of Phocaeicola massiliensis. In other words Phocaeicola massiliensis numbers increase as a result (i.e. median likely moved up).
Low Example: Paraprevotella, we get less high count and more low count of Phocaeicola massiliensis. In other words Phocaeicola massiliensis numbers decrease as a result (i.e. median likely moved up).
Looking at doing linear regression, we do not see the relationship.
Chi2 Low
Chi2 High
Number of Interaction Found
6
50
1427931
50
150
301973
150
250
35584
250
350
7966
350
450
3172
450
550
1303
550
649
695
950
1050
610
650
750
505
750
850
483
851
950
376
1051
1150
349
1250
1350
223
1150
1250
218
1350
1450
213
1551
1650
158
1450
1549
153
1651
1750
141
1850
1949
138
1751
1847
126
1950
2050
98
2050
2150
83
2150
2249
68
2252
2348
64
2350
2449
56
2451
2548
33
2551
2650
31
2951
3043
28
2753
2845
27
2659
2741
27
3057
3150
26
3153
3242
23
3253
3348
20
2856
2949
19
3862
3948
18
3365
3450
16
4057
4141
14
3470
3548
12
3752
3842
12
3555
3625
9
3652
3749
8
3956
4046
8
4162
4248
6
4278
4329
3
4867
4905
3
5763
5782
2
4373
4423
2
5102
5139
2
4771
4771
1
5189
5189
1
4737
4737
1
If you use your own limits, this can be used to determine if the limits are better or not.
Next Project
Many taxa shifts have nothing in the literature affecting the taxa for use in a clinical context. Identifying taxa with a strong interaction that we can affect should allow us to indirectly influence the target taxa. Yes, gets complex but with modern computer power, very possible to do.
I would say some small subjective improvements since last time, but no major changes. Reminder: I have a friendly MD in terms of antibiotics.
Metronidazole was on top in the last samples, I did it back then.
Comment: Metronidazole is no longer at the top but dropped down to 16% of the highest value. It appears to have done its magic in reducing the bacteria pointing to it as a tool..
Base Analysis
When people have multiple samples, I like to do side-by-side comparisons, especially when someone has been doing some of the suggestions suggested. The suggestions are computed and may not always work. Expert Systems and AI are not perfect; they typically do better than a person with only a few years of experience that has training in the discipline (better consistency, remember more facts, etc). How are we doing objectively?
Scores
We see two positive shifts in the latest sample: Increase of Anti inflammatory Bacteria Score and decrease of Histamine Producers.
Sample
2021-11-18
2022-05-20
2023-06-22
2023-09-04
Anti inflammatory Bacteria Score
54%ile
43%ile
63%ile
87%ile 🥰
Butyrate Bacteria Score
57%ile
56%ile
57%ile
54%ile
Histamine Producers
78%ile
82%ile
81%ile
67%ile 🥰
Percentages of Percentiles
We see a lot of bouncing around between samples. The middle two images matches the typical pattern seen with ME/CFS and Long COVID. Those shifts have faded over the last 3 months with a different pattern appearing indicating a different dialect of gut dysfunction.
Multi-Vector Comparison
The main numbers are below. The take away, less bacteria that are in the high percentile range (at 95%ile, 10 -> 28 -> 23 -> 8). The numbers bounce around with the middle two being similar and the other two also similar. There are no really clear shift in these measures.
Criteria
11/18/2021
5/20/2022
6/22/2023
9/4/2023
Lab Read Quality
8.1
5.5
4.7
7.2
Outside Range from JasonH
6
6
9
9
Outside Range from Medivere
16
16
15
15
Outside Range from Metagenomics
8
8
7
7
Outside Range from MyBioma
5
5
6
6
Outside Range from Nirvana/CosmosId
20
20
23
23
Outside Range from XenoGene
29
29
35
35
Outside Lab Range (+/- 1.96SD)
7
6
17
3
Outside Box-Plot-Whiskers
36
69
54
38
Outside Kaltoft-Moldrup
93
48
47
88
Bacteria Reported By Lab
652
508
542
558
Bacteria Over 99%ile
7
4
6
2
Bacteria Over 95%ile
10
28
23
8
Bacteria Over 90%ile
29
42
36
22
Bacteria Under 10%ile
208
41
50
175
Bacteria Under 5%ile
180
19
8
157
Shannon Diversity Index
1.853
1.826
1.272
1.556
Simpson Diversity Index
0.056
0.038
0.087
0.09
Rarely Seen 1%
2
2
7
1
Rarely Seen 5%
14
5
21
8
Pathogens
41
24
29
36
From Special Studies
The top match was the same on all of the samples, with an increase when there was actually COVID.
Criteria
11/18/2021
5/20/2022
6/22/2023
9/4/2023
COVID19 (Long Hauler)
28%ile
33%ile
41%ile
28%ile
Next one:
15%ile
26%ile
20%ile
13%ile
The “next one” dropping implies some reduction of dysbiosis
Health Analysis
Using Dr. Jason Hawrelak Recommendations, there are many items on the edge of being in range with some items of interest (I strike out those that are unlikely to be of great concern):
Faecalibacterium prausnitzii at 27% of the microbiome or 96%ile
Faecalibacterium prausnitzii at 27% of the microbiome or 96%ile
I would suggest a test for candida to be safe. The data suggests a risk. If confirmed, candida would contribute significantly to gut dysbiosis [The interplay between gut bacteria and the yeast Candida albicans[2021]). I did a “back-flip” check of the top prescription items, and all of them reduces Candida (studies cited below).
Addendum – Predicted Symptoms
This was just added to the site today as a further refactor based on New Special Studies on Symptoms data. These are from [My Profile Tab]
Criteria
11/18/2021
5/20/2022
6/22/2023
9/4/2023
Forecast Major Symptoms
Neurological: Cognitive/Sensory Overload 40 % match on 25 taxa
DePaul University Fatigue Questionnaire : Racing heart 38 % match on 13 taxa
DePaul University Fatigue Questionnaire : Difficulty falling asleep 37 % match on 27 taxa
DePaul University Fatigue Questionnaire : Difficulty finding the right word 35 % match on 20 taxa
Autonomic Manifestations: urinary frequency dysfunction 66 % match on 6 taxa Immune Manifestations: Bloating 37 % match on 45 taxa
Neurological-Audio: hypersensitivity to noise 35 % match on 28 taxa
None
Neurological-Sleep: Chaotic diurnal sleep rhythms (Erratic Sleep) 50 % match on 18 taxa
Neurological: Spatial instability and disorientation 37 % match on 16 taxa
This can be helpful for judging possible severity (and potential improvement of some symptoms), for example: Neurological: Cognitive/Sensory Overload. See [Special Studies] tab.
2021 – 40% matches
2022- 24% matches
6/22/23 – 16% matches
9/4/2023 – 4% matches
Going Forward
COVID has had quite an impact on this microbiome. I am going to just go with the “Just Give Me Suggestions” option with the addition of what matched his diagnosis:
To explain a bit more. First I click the button below
And then click I could click the consensus report to see what the top items are:
Which are shown below.
In this case, I want to add Irritable Bowel Syndrome suggestions (on the Changing Microbiome Tab)
Instead of the usual 4 packages of suggestions, we have 5
When we look at the consensus report we see the same items there, but the values have increased.
The intent is put a little bias on the numbers towards specific conditions of greatest concern.
PDF Suggestions
I tend to favor the PDF suggestions because it simplifies things for many readers. Also the PDF gives a good list of citations (never complete) used to make the citations to persuade MDs to see that the suggestions are based on studies — a lot of studies.
The PDF suggestions are below (using the consensus view is another option for those more technically orientated). I clip from the PDF to keep the blog simpler for the typical reader.
This is a little longer list than usual, so I went to the consensus report to get priority data. Top value was 618, so 309 is the 50% threshold.
Minor note: quercetin with resveratrol is an avoid, quercetin is a take. resveratrol by itself is a negative (-113). At times, you need to look at the technical details/consensus to clarify things; the data we are using is incomplete and sparse…. If clearly contradictory suggestions appear, then don’t do them (thing an abundance of caution).
Because he has an antibiotic friendly MD, the following are the TOP antibiotics with notes:
“Since gentamicin has minimal gastrointestinal absorption,…it has applications in several clinical scenarios, such as bacterial septicemia, meningitis, urinary tract infections, gastrointestinal tract (including peritonitis), and soft tissue infection,” NIH StatPearls
“The FDA-Approved indications include acute infective exacerbation of chronic bronchitis, otitis media in pediatrics only, travelers diarrhea for treatment and prophylaxis, urinary tract infections, shigellosis, pneumocystis jirovecii, pneumonia/pneumocystis carinii pneumonia (PJP/PCP), and toxoplasmosis, both as prophylaxis and treatment. ” NIH StatPearlls
CFS Antibiotics are also above the threshold. Since the prior sample had a strong Long COVID or ME/CFS Profile, I would be inclined to include one of those below in the antibiotic rotation. The microbiome cannot make a diagnosis of most things, with most ME/CFS microbiomes there is a particular pattern which you had in your last sample but which has disappeared from your current sample which looks more like your first sample. I read this as recovering from ME/CFS…. in likely a fragile state since relapse is very common with ME/CFS.
My own experience is that it is better to overcure ME/CFS and when there are signs of recovery…. no backflips of joy or running marathons; keep doing slow walks that becomes a bit further each week for 6-12 months. Your microbiome is fragile and can quickly slip back.
I prefer to use the strategy of going for prescription items that are both suggested from the microbiome and been shown to help with one or more of the diagnosis conditions. This usually encounter low resistance from physicians — they are clueless for the microbiome, but very accepting of published studies. An antibiotic that is used as a prophylaxis usually encounter little resistance.
KEGG Suggestions
This is done by using information from the bacteria found with some fudge factors. I am in discussion with some Ph.D. candidates to build this concept directly from the FASTQ files and will hopefully have this as an added feature next year.
The KEGG probiotics is the usual pattern for ME/CFS and Long COVID with the top one being the usual, with the top reasonably available ones for other families shown below. I usually like to compare the values with those from consensus to minimum risk (i.e. two thumbs up, we do; mixed, we skipped)
The suggestions should be thought as influencers. The human population is often a good analogy or parable for the microbiome population. Each influencer shifts the population in the desired direction. Based on Cecile Jadin’s work and several studies, I am a firm believer in short duration (1-2 weeks) of each influencers. Just as with human influencers, people stop listening if the same person just keeps droning on and on. If a different person starts speaking, you get persuaded more. If a mob start to shout, yet a different human behavior will occur. In terms of the microbiome, “stop listening” means mutations that are resistant to the item will start to increase. Items line vitamins and minerals can be taken continuously; items that are likely to have bacteria resistance developed should be taken for a week and then another item replace it.
Herbs and spices: cinnamon, ginger, black cumin, thyme, rosemary, quercetin (suggests just before each antibiotic with a few days of overlap because it has potential synergistic activity with antibiotics [2020], [2016],[2018] )
Remember our goal is to destabilize a stable microbiome dysfunction.
Questions and Answers
While there has not been significant changes in many of the vectors between this sample and the prior sample from a few months earlier, there has been two significant objective changes:
Significant improvement of Anti inflammatory Bacteria Score (higher) and Histamine Producers (Lower).
The lost of the ME/CFS – Long COVID spike in the 0-9%ile
Q: Do you/should I use the colored list now instead of the consensus list?
Either are fine, the color list (from PDF) is what I tend to use in post because it is easier for new readers to understand (and automatically sent on new uploads). The consensus page is more complex but allows people to apply their own logic and priorities.
Q: “Quercetin (suggests just before each antibiotic with a few days of overlap because it has potential synergistic activity with antibiotics”
Q: I just did Mutaflor for 8 days and felt really tired all the time (but in the end I also got a flu/cold, so maybe that was the reason and not mutaflor). Nevertheless, if it was a herx reaction, I wonder if I should have taken it for longer until the reaction disappeared? (I stopped it 4 days ago.) Not sure if this question even makes sense.
My personal choice would be to keep taking it for at least a week (perhaps 2). Remember that the traditional pattern for a herx is feeling bad for X hours and then things get better. The duration of the feeling bad usually decrease from day to day. Catching a cold makes interpretation challenging.
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 can compute items to take, those computations do not provide information on rotations etc.
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.
The answers above describe my logic and thinking and is not intended to give advice to this person or any one. Always review with your knowledgeable medical professional.
Recently I did a much more rigorous implementation of the Kaltoft-Moldrup ranges as part of a periodic review and refactor. Today, I tested a hypothesis that those ranges may be good for detecting the taxa (bacteria) associated with a self-reporting symptom. The results exceeded expectations.
The methodology is simple to understand. Suppose we have 1000 samples. The KM lower limit is 5%ile and upper limit is 85%ile across the entire population.
For those with brain fog as a symptom we found that instead of the expected 5% of 1000 (i.e. 50), we only have just 5 samples in this range. That is a dramatic (and statistically significant shift).
On the other end, we do not have our expected 150 (100%-85% * 1000) but 300 sample in this range.
We can conclude that those with less than the 5%ile is associated with brain fog as well as those with more than 85%ile.
Most of the time we will find just one shift being significant. At times, both can be — i.e. values cascaded away from the range is both directions.
I have determined the taxa/bacteria for a stack of symptoms using P < 0.01. Because different labs identifies taxa information (see Nightmare ). This is why there is often no replication of (or contrary) results in many published studies. I created a table that is lab specific for each taxa.
Overview Page
This shows the number of bacteria found significant for each symptom. “All” usually will have more because it has more samples to use and thus can detect lesser associations better.
The following are permissible extracts (200 words per article). Click [more] to read more…
How the microbiome changes our idea of what it means to be human
YOU may, quite reasonably, think you are an individual of the species Homo sapiens. Once you have finished reading what follows, you will hopefully have been convinced that there is far more to us than that. Trillions of other organisms live on (and, more notably, in) your body. As you will see in the reports that follow, their impact on you is such that you will probably never think about yourself in the same way again. Your microbes change who you are and what it means to be you. With knowledge of this facet of ourselves growing rapidly, exploring it has never been more relevant.
Until recently, scientists believed that there were three discrete parts of our nature that reflected solid aspects of an individual self: the immune system, the genome and the brain. “None of these pillars of the traditional definitions of the self – immunity, genome integrity, the central nervous system – are free of microbial impact,” says Thomas Bosch at Kiel University in Germany.
The microbes that colonise us, collectively known as the microbiome, challenge the concept of a discrete self. These include bacteria, viruses and fungi, although the bacteria are the best-studied. [more]
The best way to care for your microbiome to keep it healthy as you age
Your gut microbiome is a vital support system for mental and physical health, supplying the body with all-important nutrients and helping tune the immune system.
As we get older, the balance of microbes in our gut changes. There are declines in beneficial types, such as the anti-inflammatory Faecalibacterium, and an increase in species that lead to more inflammation, which is implicated in multiple age-related conditions, including heart disease, cancer and cognitive decline. Many studies, with participants ranging from an isolated rural population in India to a wealthy semi-urban community in Italy, show striking similarities in the microbiome signatures of old age. One key finding is that people who have no significant health concerns in older age have an abundance of distinct beneficial microbes that are lost when there is a shift to physiological decline.
It isn’t clear whether the microbiomes of healthy older people are driving their vitality or are a result of the way they live, but an astonishing study in mice by John Cryan at University College Cork, Ireland, and his colleagues found that transplanting gut microbiota from young animals to elderly ones reversed age-associated impairments in brain function. [more]
Can probiotics and supplements really improve your gut microbiome?
If you have a condition like irritable bowel syndrome (IBS), a finnicky gut or just want to keep your microbiome in top condition, you might be tempted by products and treatments that offer a microbial tune-up. But what really works? Here are the main tools to engineer a better gut.
Probiotics
Probiotics are microbes that may help to restore healthy gut microbiota. If they also improve your mood, they are called psychobiotics. You can typically get them from eating naturally fermented foods like yogurt that contain beneficial bacteria, such as Lactobacillus or Bifidobacterium.
But as an adult, these microbes are unlikely to colonise your intestines. To the extent that they are helpful, their benefit comes while they are passing through. Such probiotics stimulate immune cells in the gut to reduce inflammation, increase mucus production and deter pathogens by producing lactic acid. But as mere visitors, they need daily top-ups.
Probiotic supplements have been used (with mixed success) for more than a century to help with the gut conditions of Crohn’s disease, colitis and IBS. They have also been shown to help with weight loss in people who are overweight and are increasingly being used for other conditions… [more]
Secrets of a long and healthy life reside in your gut microbiome
WHY do we age? As youngsters, we seem invincible. We climb trees, frolic in the dirt and blithely share alarming quantities of mucus. At college, we can thrive on a diet of ramen and beer, party all night and still sit an exam the next day. But in our 30s, we start to wind down. It becomes harder to maintain muscle tone and avoid illness. Our joints start to ache and our memory begins to dim. And it is mostly downhill from there.
People have long attempted to stop or reverse this process. But fountains of youth and secrets of immortality remain firmly in the realms of fiction. Our bodies wear out, even if we no longer do the back-breaking physical labour our ancestors did. And the world seems determined to grind us down with a plethora of disease-causing microbes. To help fend off these pathogens, our bodies recruit other microbes, vast numbers of which reside in our intestines, where we feed them in exchange for their services. But, as we age, this gut microbiota becomes less effective at fighting diseases too. [more]
Where does your gut microbiome really come from – and does it matter?
Imagine a remote island, recently formed by volcanic activity, in the middle of the ocean. At first, it is lifeless, but a growing variety of plants take hold, providing food for pioneering animal species, until eventually there is a diverse and flourishing ecosystem.
This is a useful way to think about how our gut ecosystems develop. “Your microbiome goes on a journey,” says Alan Walker at the University of Aberdeen, UK. “When you’re born, some bugs get in and then, when you start eating solid foods, other bugs replace them. There’s a dynamic process where your microbiome changes until you get to mid-to-late childhood. Then, through adult life, you’ve got a reasonably stable microbial community.”
Does a C-section affect a baby’s microbiome?
The first individuals that colonise an island can have long-lasting influences on its ecosystem, an idea known as the founder effect. Until recently, the thinking went that if the founder bacteria in a baby’s gut were unusual – because the baby was born by Caesarean section, for instance – this might disrupt their bacterial ecosystems. This idea has led some parents to take radical steps to get their children’s microbiomes back on the right track. [more]
What is the role of the microbiome in diseases like chronic fatigue
ONE of the most compelling discoveries about the gut microbiome is its influence on the immune system. Between 70 and 80 per cent of immune cells are in the gut, where they are constantly communicating with microbes. This crosstalk helps fight disease, strengthen immune responses and regulate inflammation, our body’s first line of defence against infection. Controlling inflammation is critical, as too much damages cells and helps drive chronic illness.
It is no surprise, then, that a growing body of evidence implicates the gut microbiome in various chronic diseases, from arthritis to Alzheimer’s. It is still early days, and most of these findings only point to associations. But they raise the possibility that gut microbes may contribute to, or even cause, some of our most intractable conditions, an idea that has already inspired new treatments.
It is now well established that gut microbiomes in people with conditions like multiple sclerosis, type 1 and type 2 diabetes, Parkinson’s disease and even asthma differ significantly from those of people without an underlying illness. Two papers published earlier this year showed that people with chronic fatigue syndrome – also known as myalgic encephalomyelitis, or ME/CFS – have less of a gut bacterium called Faecalibacterium prausnitzii [more]
How your microbiome is shaped by your friends, family, lovers and pets
“I jokingly say that your dating app profile should include your microbiome profile,” says Brett Finlay at the University of British Columbia in Canada.
The strongest evidence comes from work published in January by Mireia Valles-Colomer at the University of Trento, Italy, and her colleagues, who conducted the largest study to date of how our gut microbiomes are shaped by the people around us. They analysed DNA in the faeces of more than 7000 people from households around the world, including rural parts of Africa and South America and cities in the US, Europe and China, to find out which bacterial strains were in their guts and what proportion they shared with others. [more]
—————– from older issues ———————-
Your gut microbiome is linked to your fitness and biological age
The diversity of microbes in the gut could affect a person’s fitness and their biological age. Better understanding this may one day lead to probiotics that alter the gut’s microbial make-up to promote health.
Zsolt Radak at the Hungarian University of Sports Science and his colleagues studied 80 amateur rowers, aged 38 to 84, who participated in the 2019 World Rowing Masters Regatta in Velence, Hungary.
The rowers, whose training regimens ranged from practising every day to once a week, each provided a stool sample to identify the bacteria in their guts. The researchers also took blood samples to gauge the participants’ biological ages – a measure based on DNA markers, rather than the number of years someone has been alive.
On top of different labs interpreting samples in different manner which I have covered in the posts below, there is an additional issue to compound determining if a level of bacteria is normal or not!
I described it in my 2017 post, and updating information here.
One of the common misconception is that there is a “normal” microbiome that can be used as a reference. Below is a chart from “Metagenomic sequencing of fecal DNA[2016]“. Diet and environment makes a major impact on the distribution and volume of the bacteria.
“In a study of gut bacteria of children in Burkina Faso (in Africa), Prevotella made up 53% of the gut bacteria, but were absent in age-matched European children.”[2010]
The chart below is for healthy individuals in 12 different countries. In some cases neighboring very similar countries (Sweden [SE] and Denmark [DK]) have very different compositions.
This great variation means that testing the microbiome can only be done as group of individuals living in the same area with similar eating habits…. If you are a vegetarian living in Australia, the reference ranges provided by your Australian lab are very questionable for you to use.
An individual result without reference from people with the same eating habits and possibly ethnic background is very fuzzy to interpret. Yes, highlights may be common — like low E.Coli, Lactobacillus and Bifidobacteria…. but they likely apply to no more than 80-90%, the other CFS patients may have different shifts.
Then we also find that DNA also impacts the microbiome,
Host genetic variation drives phenotype variation, and this study solidifies the notion that our microbial phenotype is also influenced by our genetic state. We have shown that the host genetic effect varies across taxa and includes members of different phyla. The host alleles underlying the heritability of gut microbes, once identified, should allow us to understand the nature of our association with these health-associated bacteria, and eventually to exploit them to promote health.Human genetics shape the gut microbiome , 2014
People have asked me, “Did you get your microbiome done, what was it?” My honest answer was “No, such testing was not available when I last had CFS. I simply assumed that my pattern would be an appropriate match to that reported from the 1998 Australian studies”
Age changes the microbiome
” DNA of the Clostridium leptum group and pathogenic Enterobactericeae increase in the gut microbiome with age and can be detected in the same individual’s coronary plaques along with pathogenic Streptococcus spp., associating with more severe coronary atherosclerosis. ” [2019]
The presence of the Bifidobacterium, Faecalibacterium, Bacteroides group, and Clostridium cluster XIVa decreased with age up to 66-80 years of age, with differences reaching statistical significance for the latter group. Interestingly, the levels of some of these microorganisms recovered in the very old age group (>80 years), with these older individuals presenting significantly higher counts of Akkermansia and Lactobacillus group than adults and the younger elderly.
Underlying these macro-level microbial alterations were demonstrable increases in select bacterial genera such as Veillonella (+14,229%) and Streptococcus (+438%) concomitant with reductions in Alloprevotella (-79%) and Subdolingranulum (-50%). To our knowledge, this case study shows the most rapid and pronounced shifts in human gut microbiome composition after acute exercise in the human literature.
“We analyzed the combined microbiome data from five previous studies with samples across five continents. We clearly demonstrate that there are no consistent bacterial taxa associated with either Bacteroides– or Prevotella-dominated communities across the studies. By increasing the number and diversity of samples, we found gradients of both Bacteroides and Prevotella and a lack of the distinct clusters in the principal coordinate plots originally proposed in the “enterotypes” hypothesis. The apparent segregation of the samples seen in many ordination plots is due to the differences in the samples’ Prevotella and Bacteroides abundances and does not represent consistent microbial communities within the “enterotypes” and is not associated with other taxa across studies.” [2016]
” All Egyptian gut microbial communities belonged to the Prevotella enterotype, whereas all but one of the U.S. samples were of the Bacteroides enterotype.
The intestinal environment of Egyptians was characterized by higher levels of short-chain fatty acids, a higher prevalence of microbial polysaccharide degradation-encoding genes, and a higher proportion of several polysaccharide-degrading genera.
Egyptian gut microbiota also appeared to be under heavier bacteriophage pressure.
In contrast, the gut environment of U.S. children was rich in amino acids and lipid metabolism-associated compounds; contained more microbial genes encoding protein degradation, vitamin biosynthesis, and iron acquisition pathways; and was enriched in several protein- and starch-degrading genera.
Levels of 1-methylhistamine, a biomarker of allergic response, were elevated in U.S. guts, as were the abundances of members of Faecalibacterium and Akkermansia, two genera with recognized anti-inflammatory effects.
The revealed corroborating differences in fecal microbiota structure and functions and metabolite profiles between Egyptian and U.S. teenagers are consistent with the nutrient variation between Mediterranean and Western diets.” [2017]
“This suggests that similarities between the Inuit diet and the Western diet (low fiber, high fat) may lead to a convergence of community structures and diversity. However, certain species and strains of microbes have significantly different levels of abundance and diversity in the Inuit, possibly driven by differences in diet.” [2017]
There is no clear definitive benefit from doing an individual microbiome testing — there is no definitive reference ranges. This is an inconvenient truth about the microbiome testing – rarely talked about and typically ignore.
My training is in statistics and artificial intelligence where there is no concept of definitive, just probability and fuzzy data.
The path that I have walked down on my Microbiome Prescription site accepts this problem and use a wide variety of methods (familiar to some of those people who are very well practiced and experienced in probability, statistics and artificial intelligence) to maximize the odds that suggestions will improve a person’s health. Both the simplified logic of influencers and the naïve application of the “hottest new” artificial intelligence fad are ignored. Many people cannot get their minds wrapped around the nature of this problem, IMHO.
This latter issue persists even if you get lab test results to agree.
” This work supports that sex is a critical factor in colonic bacterial composition of an aged, genetically-heterogenous population. Moreover, this study establishes that the effectiveness of dietary interventions for health maintenance and disease prevention via direct or indirect manipulation of the gut microbiota is likely dependent on an individual’s sex, age, and genetic background. ” [2019]
This is an update of a post that I did back in 2018 [Original Post]. A reader had messaged me about the safety of soil based organisms probiotics. For hundreds of thousands of years, soil based organism was a part of our diet because of the absence of safe water, soil was often on the food consumed. I recall reading that the human gut bacteria has strong similarity to that seen around root vegetables. This is not surprising, pulling roots out of the ground (without washing!!!) was likely common for most of these thousands of years…
“But I read about someone getting sick from SBO!”
There are two types of sick to consider: bacteremia or endocarditis (BAD), and herxheimer reaction (usually a good sign — I was really sick when I first started Mutaflor, it ebbed and I was much better afterwards)
The most dangerous probiotic (using report counts) are Lactobacillus.
There are similar risk from eating “safe” lactobacillus probiotics, cheese, yogurt, etc. Even deaths have been reported: “Lactobacillus-Cause of Death ” [2010]
” Lactobacillus has been used as a probiotic bacteria to treat diarrhea and is also present in dairy foods. It is hence commonly used. Lactobacillus endocarditis, an exceedingly unusual disorder, is accompanied by high mortality and poor response to treatment. ” – OUCH!
“In recent years, infections caused by Lactobacillus and Bifidobacterium made up 0.05% to 0.4% of cases of endocarditis and bacteremia. In most cases, the infections were caused by endogenous microflora of the host or bacterial strains colonizing the host’s oral cavity. According to a review of cases of infections caused by bacteria of the genus Lactobacillus from 2005 (collected by J.P. Cannot’a), 1.7% of infections have been linked directly with intensive dairy probiotic consumption by patients. ” [Lactic acid bacteria and health: are probiotics safe for human?]. [2014]
Some more citations…’ bacteremia is a bad bacteria infection, endocarditis is a bacteria infection of the heart. It has been only in the last few years
Probiotics are generally safe. No probiotic is 100% safe. To me, soil based bacteria are likely more beneficial then lactobacillus because they went along with our ancestor’s diet long before we started domestication of milk producing animals. There may be considerable basis to the hygiene hypothesis which would result from our modern pathological obsession with sterilization of food in the belief that all bacteria are bad.
The general belief is that issues arise with a weak immune system, after surgeries, and with a “leaky gut”.
People like (and expect) absolute certain answers. Statistician and Artificial Intelligence Engineers NEVER expect absolute answers… they expect fuzzy answers and just work with it.
There are two problems with determining levels from the microbiome:
Correct identification of the bacteria from the test
Determining if the bacteria produces the substance.
These studies are used by many labs to determine amount being produced.
Microbiome Prescription Use Genetics
Is the bacteria capable of producing the chemical? Surprise, surprise, surprise… this list disagrees with the inference studies above. We still have the challenge of labs reports misidentifying the bacteria. Which is more reliable? Well, with genetics, we do not know if the production process is turned on or not. We do know which ones are incapable of producing.
The Wish List
I have tossed this request over to a person that has the academic skills (and creativity) to explore. Take the FASTQ files and the data from KEGG: Kyoto Encyclopedia of Genes and Genomes and see if you can determine the amount of genetic material producing each of these products. We totally side-step the key point of failure — identifying the bacteria!
That is, create software that takes in FASTQ files and provide estimates for all of the applicable Enzymes present! Remove bacteria naming from the process.
Possible software includes: Piphilin, Tax4Fun, PICRUSt2, PICRUSt
Human Analogy
You want to be rich. So you look at the rich and see expensive cars, big homes, trophy spouses etc. So you deduce that you just need to have those and you will become rich!! After all, there is a strong statistical association!! The alternative is to look at wealth production (the genes) and you see a different picture: high yield stocks, inherited money, professional licenses, etc. It is the same with looking at what bacteria produces.
Today I looked at a sample and when I looked at raising butyrate, there were ZERO bacteria selected. In other words, there were no butyrate bacteria reported by the lab. This could be defect with the reference library used by the lab or a hundred other reasons. I was also hit by requests for suggestions from OATS test.
I did a series of detail posts on OATS in an Autism context which some may find informative:
I have created A Priori Reports for Pub Med reported conditions and it seems logical to create the same for compounds produced using KEGG: Kyoto Encyclopedia of Genes and Genomes data. Some examples are: Histamine (reduce), Butyrate (increase), and for SIBO(decrease): Hydrogen, Methane
This is experimental.
Logic Used
KEGG reports at the strain level using the full genome of each strain. From this information, we approximate the species genome (assuming the strains in the species are reasonably representative).
If the existing tests are 100% accurate at identifying all of these strains…. , we should have a smile. The reality is that tests do not report on many strains, and often disagree on species!! The problem is that they are not “safe” in their identification. For why, see these three posts:
To get around this defect, I used patterns from AI where dealing with imprecise data is normative. I assume that we could get reasonable suggestions by imagining that every species listed is present in equal amount and then generate suggestions from this synthetic microbiome.
Where is it available on the site?
Main Public Menu
The most commonly asks are under Information From Studies. They will produce a PDF download.
Using OATS Test Results
I have created a video of using both of these pages with explanation of what and how we are doing things.
and produce a PDF report that aggregates the suggestions over everything marked out of range.
You will see a summary of what you entered:
A list of items to take or to avoid
Logged In – Advance Display
The “kitchen sink” is available after logging in and setting display level to advance. KEGG Derived Data appears on the menu
Under this menu are two items: one for compounds (like Organic Acids) and one for enzymes (like the one that creates histamine).
On these pages you can get the list of bacteria used for each calculation.
Cross Validation
I usually do some spot checks for reasonableness of suggestions. I picked the highest value item to reduce histamines: Thyme, and was lucky to find a study specifically for this.
This is all an academic exercise for me. I do believe that this approach will likely produce superior results than random trial and error; or relying on influencers for approaches. As always, review with your medical professional before starting.
Recent Comments