The Journey Begins with your microbiome

Thanks for joining me!

This is a companion site to the analysis site at:

Most of the content was originally posted on with the pages on the left being a restructuring of selected posts from over a thousand posts on that site.

Also a PodCast:

Recommended Site For Testing

If you have ME/CFS or other financially disastrous condition, there is always a nasty cost factor for testing. My usual recommendation is for the cheapest, high quality provider that provides information for upload to my analysis site. Some sites provide a mountain more of information — but the benefit from that extra information is almost nothing (and it adds $$$$ and complexity).

  • is shutting down. This had been my personal usual site because using a variety of techniques, the cost was $25/sample. Don’t order from there.
  • (EU based) is an excellent buy using our discount code [MICRO]. They have also automated data transfer to our analysis site.
  • Thryve is what I am starting to use. Their reports may be processed here for independent suggestions. I would also recommend

Sample Comparison Tools

This set of tools is designed for two scenarios:

  • One person who has had many samples of time. Typically that person is looking for outliers to reduce or disappear
  • A family with samples from different members. Typically one person is challenged and since the family group has shared DNA and diet — the hope is that the bacteria grouping causing the challenge will be identified. Once identified, it may be actionable.
    • This scenario will have more tools added over the next weeks.

I have used them in these prior posts:

If you have two or more samples uploaded, you will see the top two items on the Available Samples page. These may be collapsed into one over the next few weeks.

Clicking the right button of these two will take you to a sample selection page.

Clicks TWO or more samples

The program will list all items below that matches all samples OR all samples except 1 (but at least two).

I selected a group of 5 samples from when I was having a ME/CFS flare.

The lack of dehydrogenases above would account for high lactate(brain fog) and agrees with research [Nicotinamide adenine dinucleotide (NADH) in patients with chronic fatigue syndrome]

IMHO, it correctly identified what was wrong with me.

KEGG Modules takes a bit of research to understand.

Doing some research, I found “. L-tryptophan is produced in the shikimate pathway from chorismate” Which lead to many ME/CFS articles on PubMed.

Again, it may take some research to understand what this is.

This informed me which bacteria were constantly high over 5 samples

The cause of ME/CFS flare was stress and a quick search on PubMed found several articles where this genus increased with stress, the latest was “Gut Microbiota Are Associated With Psychological Stress-Induced Defections in Intestinal and Blood–Brain Barriers“[2020]

Bottom Line

This set of tools does not give immediate answers; it gives you leads to investigate. For myself, the findings plus the use of PubMed studies weaved a story of what happened that agrees with the literature. This is very important because ME/CFS contains dozens of subsets. Often I have seen that what is helpful for one subset is harmful for another. I suspect this also applies to other conditions, such as ASD/Autism.

In this case, it identified one key family to reduce. Identified enzymes that I was short on. Lead to a possible supplements that I should consider because of the dysruption.

Thryve vs BiomeSight – the numbers

A reader asked which one to use. They can be compatible prices, especially this weekend with Black Friday specials A cost item that should also be factored in is shipping costs to and from. In the US, Thryve comes with a postage paid return package.

The Numbers

The upload page gives raw numbers. I am also going to dive a little deeper into the numbers

Elusive 1%

Adjusting for number of samples, they appear very similar.

Elusive 2%

These charts are those between 1% and 2% in occurance

Elusive 2-4%

This is the count between 2 and 4% Frequency. BiomeSight appears to have an edge.

Elusive 4-8%

As above, BiomeSight curves appears better. 100 is at the 84%ile with Thryve and at the 41% with Microbiome; in other words, Biomesight report more in this range per sample. This is important for the AI analysis, because we need a threshold count before we can detect patterns.

Elusive 8-16%

Thryves now pulls ahead. Biome Sight has 70 bacteria count at the median, while Thryve has 90

But wait! Does it report on what you are interested in?

In my last person analysis, there was two probiotic recommendations:

I would like to see those counts on my next sample…. so clicking on the above links, I see that stats:

Ouch, BiomeSight is the only one that reports either! Looking at the parent group, I see BiomeSight again reports better

Bottom Line

There is no clear better or worst — it depends on your needs.

BiomeSight offers free processing of Thryve FASTQ files which is big Kudo to Rose at BiomeSight. Thryve offered free processing once upon a time, but it does not appear to be offered any more (or it is sufficiently hidden that I cannot find it).

The new kid on the block, nirvanabiome, which uses, is 3x as expensive and does not appear to report any more bacteria types (which is surprising given their claims, I expected counts close to Xenogenes shown at the top of this post). I do not have sufficient samples via CosmosID/Nirvanabiome to do more analysis.

nirvanabiome also appears to be targeting the Autism market. I have a separate blog on Autism and the Microbiome, so I am interested if they will produce actual beneficial results or if this is “the best of intentions, the worst of execution” scenario.

Australia Microba Gut Test Uploads

I have implemented an upload for Microba, an Australian firms that claims “With the most comprehensive microbiome test available”. Instructions on how to do a download and upload is in this

I have tried several times in the past to do it. One of the biggest problems is that they do not use NCBI reference numbers or names. In fact, many of the bacteria they name — you will not find a single study on PubMed with that name. In other words — valueless information.

I have a mapping of their interesting names to NCBI names on line (and it will grow as samples are added and new names are added). The mapping is located here. I have repeatedly email them to make a download with NCBI taxon numbers available without success.

Left side is their name, right side is NCBI name. CAG-### is very vague.
Some additional “delights”

Only Selected Layers are Reported

They report only on the Phylum, Family, Genus and Species levels. Excluded are Orders, Classes and Strains. After the mapping, we are typically left with less than 100 bacteria taxonomy versus many more from other providers. I do not know how they define “With the most comprehensive microbiome test available”. Most means better than ALL…

Number of different bacteria reported by Sample

In short:

  • the information available is far less.
  • This is made worse by the use of atypical names for bacteria. If you are high “Peh17” and go to PubMed to see what will lower it, or what conditions are associated with it — you hit a blank page. They may provide advice — but the basis of that advice cannot be independently checked.
  • The sum of all Species/Genus/Family is 100%. This implies that they have identified every bacteria — impossible. They have scaled the numbers of the bacteria that they detected to 100%. A person with actually 40% of one bacteria in their gut could see a report of 45%, 65%, 85% — depending on what other bacteria is there.
  • The report is to 0.01% that is 100 / 1,000,000, a coarser measurement than some other tests.

Bottom Line

For those of you who have already tested with Microba, you can upload and MicrobiomePrescription will do as much as it can with that information. If you decided to do a retest— I would not recommend using Microba for the reasons sited above. I have heard that the UK firm BiomeSight is making it easier for Australians to use their service. I have heard that duties and shipping costs makes  Thryve Inside more expensive than BiomeSight.

The upload page is at:

Expect it to be a few days before 100% of your sample is ready — any new odd-ball names has to be researched and entered into the mapping table. At upload, you will likely be 80+% processed immediately.

One Stool, Two Samples, One Lab — What the shit!

A reader sent me the message below and gave permission to use his sample. I had, about a year ago, wrote The taxonomy nightmare before Christmas… that looks at the differences between lab results using the sample sample (as represented by a FASTQ digital file). We now try one more variation.

Last september I did (again) test my microbiome with Thryve. Because I had some general doubts about the validity of stool samples, I ordered two tests and took two different samples of the same stool and send them in under two different names.
…the results confirmed my doubts as I got different bacteria levels of the ten strains Thryve shows in their overview. 

STRAINS% sample 1% sample 2

So I do not doubt the reliability of each sample, but see that the validity of the sample is the problem. The results of a sample seem to be more or less random and not representative of the microbiome in general.
…so I think that any advice given, based on the results of one sample is arbitrary. If we are to take the importance of the microbiome seriously, we will have to consider a new way of getting a representative sample to have a solid base for interventions concerning our health.

Sampling Statistics

The typical sample seems to contain a round a 100,000 bacteria and is usually reported out of a million (scaled up). “Bacteria in faeces have been extensively studied. It’s estimated there are nearly 100 billion bacteria per gram of wet stool. ” [src] The sample that you sent it was likely no more than one milligram.

To use the “if I was a Martian” model… It is like a spaceship abducting a boatload of people in the Mediterranean…. If the boat is a cruise ship full of fat diabetic elderly Americans you will get one result. If the boat are full of starving Nigerians children trying to become refugees in Europe, a very different result. That is a disturbing concept when you mind is fixed on a deterministic precise definitive result. It’s a sample folks! For most industrial processes, dozens (or hundreds) of samples are required to get quality assurance. For the nerds, some readings: [2015] [Wikipedia]

Example: Two employees working for the same company at the same job earning the same amount and living in the same community. You stop each of them and take a sample of how much money they have in their wallet.
Would you expect them to have the same amount? Would they have the same number of pennies? dimes? quarters? Credit Cards?

A Sampling Analogy

Reviewing these two samples

Fortunately, I have sample comparison tools already on,

Diversity By Taxonomy Rank

I would expect differences in samples to increase as you move down the rank. It is similar to asking at one level [European, African, Asian] on the abducted ship above. At the next level [Swede, Dane, Italian, etc] , the counts between sample will diverge as you do more detail classification.

This is an illustration on why I do fuzzy logic on predicting symptoms with good success according to readers. Using studies from PubMed have been reported to produce poor results according to readers.

When the two samples are used to predict symptoms, we have a strong convergence. While the actors may be different, their impact are similar.

Adjusting for Natural Variation

Using counts without context is a good way to get upset without justification. I use percentiles to provide context and have a comparison page (which I need to revise). At the phylum level we see general agreement between the samples. One rare phylum was lacking in one sample (not found in 30% of Thryve Samples but only 6% of BiomeSight – hint: download the FASTQ files and process them thru BiomeSight [for free!]).

Medical Condition Matches

Going over to Pub Med Medical condition matches, we see a striking similarity between the samples as shown below. So for detecting medical conditions — they are almost identical to each other (despite the differences in bacteria)

End Products Predictions

Again, we have strong agreement between the samples using 3 buckets.

  • Both below 12%ile (i.e. Low)
  • Both below 82%ile (i.e. High)
  • Both in normal range

This means for this type of diagnostic evaluation — they appear to be the same.

Bottom Line

There are several questions that need to be asked (and an answer to one):

  • To the folks at Thryve (and, why are the numbers so different?
  • For users of my analysis site:, for diagnostic purposes there are few differences! We have general agreement for:
    • End Product Production
    • Medical Studies Matches
    • Symptom Matches
    • Detecting high or low levels by percentile

The critical difference between the information lab providers and my site is interpretation sophistication.

So, to answer the reader’s question “The numbers are in major disagreement, but the diagnostic significance of the whole sample is in strong agreement”. Doing the lab analysis is worth it — just ignore the lab’s “value added” suggestions/information.


I have just completed a series of charts showing timelines (over 1000 new charts for most users). Timeline are important because they show how things are changing. Showing a time line can be complicated because numbers from different lab software are not comparable (see Taxonomy Nightmare post) even when the same analysis data file (FastQ) is used!

To address the issues of different numbers, different symbols are used for different lab software. It is strongly recommented that you obtain the FastQ files from the lab that did the sample and process them thru:

The results should be 3 or more sets of reading for the single sample.

There are no magic most important number for all people. Nor can a single test provide a solution. Your microbiome changes over time, and as you attempt to change things, there will be unexpected shifts. If you are dealing with health issues, one test every 2-3 months is strongly recommended.

The data is recomputed at least once a week (there are a lot of numbers to recompute to keep the data current!).

The timelines are divided into 5 collections, as shown below

Expected Benefit

My own experience has been that I was able to improve (and in some cases, eliminate) medical concerns by using regular microbiome tests and altering diet, supplements etc to manipulate the microbiome. The challenge is identifying what is off, and then how to correct it.

This suite of timeline charts are intended to make that easier (although you may have many charts to look thru).

Bacteria Time Line

This collects the distributions for all bacteria types seen in all of your samples and allow you to inspect changes by time (and lab software!).

Showing using Log(Bacteria Count) – Default
Showing actual values —
Shown as Percentage of highest value of any sample

The bacteria rank goes down to Strains, when that is reported by the lab software.

End Product Timelines

End products being produced by the bacteria are estimated from available data. Since each lab reports different bacteria counts, there will be some variability. Again we have 3 style of display: Log, Value, % of highest value.

There are about 150 choices to explore. If a chart is blank, then there was no bacteria matches — this may not be an item of concern because we have partial knowledge only of what is produced by which bacteria.

Medical Conditions Timelines

This uses studies from PubMed which report ‘higher’ or ‘lower’ levels. We use quartiles (highest and lowest) to compute values. In general, readers have reported that these numbers are less accurate than symptoms (see section below). We have 3 style of display: Log, Value, % of highest value.

There are about 150 choices to explore

Symptoms Timeline

This is from this site’s Citizen Science Artificial Intelligence algorithms. Only symptoms that have at least 5 bacteria very statistically significant are shown. The number of items may increase or decrease with time. At present, it’s around 100.

For myself, when I was above the 90%ile, I was having regular night sweats.

Sample Profile Timeline

This displays a variety of general characteristics as shown below:

A brief summary

  • Bacteria Count: Number of bacteria identified. More diversity is usually good, but excessive diversity, especially of rare bacteria, usually indicate issues
    • Rarest 1%: Seen in less than 1% of uploaded samples
    • Rare 2%: Seen in less than 2% of uploaded samples
    • Rare 4%: Seen in less than 4% of uploaded samples
    • Unusual 8%: Seen in less than 6% of uploaded samples
    • Infrequent 16%: Seen in less than 16% of uploaded samples
  • Firmicutes-to-Bacteroidetes Ratio: Some people deem it significant for some conditions
  • Prevotella-to-Bacteroides Ratio: Some people deem it significant for some conditions
  • Overall Symptom Health: This is a measure summing all bacteria matches for all symptoms
  • Overall Medical Condition Health: This is a measure summing all bacteria matches for all medical conditions

Bottom Line

This suite of charts gives a lot of analysis information. Most of these charts allow you to drill down immediately

Which takes you to a page like this:

Enjoy, learn, share.

Fecal Matter Transplants and Phages

FMTs have been tried for Chronic Fatigue Syndrome/ME with mixed success. The why of failures has been an ongoing interest of mine. We may now have a significant factor that has been ignored in these attempts.

 Fecal microbiota transplantation (FMT) as a special organ transplant therapy, which can rebuild the intestinal flora, has raised the clinical concerns. It has been used in the refractory Clostridium difficile, inflammatory bowel disease, irritable bowel syndrome, chronic fatigue syndrome, and some non-intestinal diseases related to the metabolic disorders. But this method of treatment has not become a normal treatment, and many clinicians and patients can not accept it. 

[Research progress of fecal microbiota transplantation] 2015

This week’s Economist had an extended essay on Viruses and the like: The aliens among us/The Outsider within, this provides good background.

In addition to this, there was a podcast reporting success with FMT was associated with higher Phage Diversity in the donor. Phages are the police of the microbiome.

In this retrospective analysis, FMTs with increased bacteriophage α-diversity were more likely to successfully treat rCDI. In addition, the relative number of bacteriophage reads was lower in donations leading to a successful FMT. These results suggest that bacteriophage abundance may have some role in determining the relative success of FMT.

The success of fecal microbial transplantation in Clostridium difficile infection correlates with bacteriophage relative abundance in the donor: a retrospective cohort study (2019)

My earlier posts on FMT

Bottom Line

This implies that for a greater chance of success and less risk, than DYI fecal transfer, that a lab that tests for possible infections AND for phage state may yield the best results.

Comparing Taxons on same FastQ file

David Morrison requested this feature. The FastQ file is produced by the physical lab machine. This file is then pushed thru software to produce a list of taxonomies. Different 16s retail providers use different software and as a result – different reports. For back ground see this “Taxonomy Nightmare before Christmas” post.


This feature is available only for those that sign on securely (i.e. request a login link to be sent by email).

On the samples page you will see some new buttons

Items are disabled if you use legacy login
Buttons enabled with secure login

Define FastQ files

Simply Click [FastQ Catalog] and create records for each FastQ file you are going to work with.

Identifying sample with FastQ file

Return to the Samples page, and click [Update of Delete Samples]

You may now assign the FastQ file to the sample intrepretation


You need at least 2 sample to be linked to a FastQ file. Returning to the Samples page, click the [FastQ Set Comparison] button

Select the FastQ file that you are interested in, click Compare button.

You will now see the results reported. Some taxonomy will be very similar and others very different.

What’s next?

Once sufficient users link their files, we can start doing the various analysis that David was asking for.

Post-Covid19 Syndrome

Most of the Post-Covid19 Syndrome symptoms has a strong match to the symptoms seen with Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS). This condition has no conventional medical treatment known. Treatments attempt to mitigate symptoms.

There is evidence that some people develop a long-term fatigue syndrome from coronavirus infections, Dr. Anthony Fauci said Thursday.

“There may well be a post-viral syndrome associated with Covid-19,” Fauci told a news conference organized by the International AIDS Society. The group is holding a Covid-19 conference as an add-on to its every-other-year AIDS meeting.

Fauci said the symptoms resemble those seen in patients with myalgic encephalomyelitis, or ME, once known as chronic fatigue syndrome.

“If you look anecdotally, there is no question that there are a considerable number of individuals who have a post-viral syndrome that in many respects incapacitates them for weeks and weeks following so-called recovery,” Fauci said.CNN

I am very familiar with ME/CFS as anyone who knows my story can attest. And I believe that while the model of why these symptoms are there is simple, the treatment is complex, not cookbook and must be individualized for each person.

This image has an empty alt attribute; its file name is image-8.png
Persistent Symptoms in Patients After Acute COVID-19

The Model

COVID19 sends out chemical signals to the body to produce chemicals (metabolites) that its need OR which create a friendlier environment for it. The signals alters the microbiome (gut bacteria) to be a factory for its needs (Viralforming the gut). Once COVID is eliminated, the alterations should return to the prior state overtime— unfortunately a percentage that take a long time or never return. The best documented example is the Bergen’s Giardia Infection.

“Gastrointestinal symptoms are common in patients with COVID-19, and had an increased prevalence in the later stage of the recent epidemic in China. SARS-CoV-2 enters gastrointestinal epithelial cells, and the faeces of COVID-19 patients are potentially infectious.”Review article: gastrointestinal features in COVID-19 and the possibility of faecal transmission

The microbiome consists of many co-operations between bacteria. Often there are over 2000 bacteria involved in various dialogs. Identifying the bacteria that are at abnormal (too high or low) is the start. The next step is modifying the bacteria by drugs, diet, supplements. At this point, we need to point out that there may be 100 or more abnormal bacteria that needs to be adjusted. Naive adjustments may make more bacteria abnormal.

My Proposed Process

This is the process that I have done with ME/CFS and it is likely that it may also work for many people with post-covid syndrome:

  • Obtain a 16s analysis from an appropriate source
  • Look at the suggestions for correcting and make appropriate changes
  • Two+ weeks after making changes, repeat the 16s analysis and make the next set of changes.

This is an iterative process. Often the suggestions will cause one symptom to lessen or disappear while leaving others untouched. This is to be expected.

For more information, see

BiomeSight (UK) is now integrated with Microbiome Prescription

Rose Walbrugh and I are proud to announce one click sending of data from based in the UK to After you get your BiomeSight data processed, you can send the data across without needing to download and upload. You will be sent an email with an automatic login link (no more making up and remembering passwords!).

The transfer is available as an external application on a mobile device, as shown above
Or with more details in a web browser.

Once the [Send] button is clicked, you will get an email like shown below that allows you to login and explore your data deeper.

Clicking the link will log you in automatically, and you will see your sample and it’s identified as originating with BiomeSight.

BiomeSight specific distributions of bacteria will automatically become available once there is a large enough sample.


As part of this celebration, a discount code “MICRO” is offered on BiomeSight services. This results in £60 off, which brings the price down to £89 per kit ($110). Local USA fulfillment is now setup. Expedited 2 day delivery at £4.95.

Bottom Line

Microbiome Prescription is dedicated to working with labs to enrich user experience and knowledge. BiomeSight has stepped up to the plate for cooperation and win-win attitude.