Back to work after ME/CFS for 10 years and Long COVID

This is a follow up post on ME/CFS x COVID :- Long COVID instead from June, 2022. The person on seeing the results stated “New Sample Looks Worse”. I am curious because so far all subsequent analysis showed objective improvements (and subjective improvements too!) for people with ME/CFS. I am prepared to be humbled.


  • Lifestyle-wise, I’ve definitely been experiencing a big increase in stress due to working a full-time job for the first time in almost a decade. The job is remote, I couldn’t do it otherwise, but it still requires some late nights and lots of childcare complications since my wife also works full-time and then some.
  • Diet-Wise: Not much has changed in terms of my diet. I remain 95% Gluten Free with the occasional slip-up or cheat. What’s interesting is a lot has flip-flopped in this sample, so maybe I was overdoing it on the last round of food suggestions, especially in terms of adding fiber to my smoothies, mostly resistant starch.
  • Supplementation-wise, I had been taking the recommended probiotics from my last sample at a pretty high and aggressive dose to very mixed reactions. I probably wasn’t rotating them often enough.
  • Paxlovid Experience: As I mentioned in a previous email, I had a pretty bad case of Covid around Christmas time. And to my surprise Paxlovid not only helped my acute Covid symptoms, but it overall made me feel much better than baseline. I was able to confirm this experiment in mid-January when another member of my family got COVID, but couldn’t tolerate their Paxlovid. So as an experiment, I took the remaining three-day course, and again almost immediately my brain fog, executive function, and most neurological symptoms lifted

Also, and this is where I think we might be able to confirm some of Dr. AI’s suggestions instead of you being humbled by them, I was getting desperate when starting the new job in January and coming off of my case of Covid. So I started throwing pretty much any “energy” and “anti-viral” supplement I had on hand or had short-term success in the past with, to try and get well enough to do a good job at my work. A lot of those herbs ended up on the avoid list on this sample, especially Baicalin, Oregano Oil, and Resveratrol (which is usually mostly Japanese Knotweed).

Other things on the avoid list now I had frequently been taking before this sample: salt (salty electrolyte packets added to water), fish oil, pulses (lots of beans, especially navy beans since those had been a strong Take for a few samples going), and Culturelle (lactobacillus rhamnosus gg) for diarrhea, because I’ve been experiencing more frequent and urgent loose stools since Covid.Reader


First, I did a recalculated to make sure all samples used the same reference sets. The reference data is recomputed weekly so we have a living analysis system. The data has been processed thru both Ombre Labs (OL) and BiomeSight (BS), so the details are below. It is the same three FASTQ files processed by two different software packages (for backstory see: Different Microbiome Results from Different Providers on Same SampleSame Raw Data via Ombre Labs(Thryve) and BiomesightThe taxonomy nightmare before Christmas…)

Lab Read Quality5.38.615.35.38.615.3
Bacteria Reported By Lab531567493671746687
Bacteria Over 99%ile3519816113
Bacteria Over 95%ile754637492422
Bacteria Over 90%ile1337157864750
Bacteria Under 10%ile5519328051161329
Bacteria Under 5%ile261732452350289
Bacteria Under 1%ile21451973583
Rarely Seen 1%7561578
Rarely Seen 5%243032615870
Outside Range from JasonH444555
Outside Range from Medivere121212151515
Outside Range from Metagenomics888777
Outside Range from MyBioma333888
Outside Range from Nirvana/CosmosId232323252525
Outside Range from XenoGene282828434343
Outside Lab Range (+/- 1.96SD)34219241417
Outside Box-Plot-Whiskers1358282976363
Outside Kaltoft-Moldrup150190223218290400
Condition Est. Over 99%ile707004
Condition Est. Over 95%ile211170318
Condition Est. Over 90%ile352342531
Enzymes Over 99%ile167171032420
Enzymes Over 95%ile338521021578540
Enzymes Over 90%ile481109134245183142
Enzymes Under 10%ile110311638126221441
Enzymes Under 5%ile1925843248112314
Enzymes Under 1%ile21881621264
Compounds Over 99%ile20824479516936
Compounds Over 95%ile46194128356502158
Compounds Over 90%ile639304235584694329
Compounds Under 10%ile598629691610663726
Compounds Under 5%ile592617668604628710
Compounds Under 1%ile584591628600606680

My initial impression is positive using Enzymes. The number of under production of Enzymes has been reduced significantly – between 50% reduction to 99% reduction. I am not that concerned with over production, typically the body discard surplus of chemicals (with a very few exceptions). I tend to be more concern over enzyme starvation. Special studies found that Compounds tend to have much weaker relationships than Enzymes. While included, I usually do not over-read the compound significance.

Outside of ranges were interesting — my preferred range Kaltoft-Moldrup, had less in the latest sample (with the numbers dropping with each sample). All of the 3rd party lab results were unchanged. To remind readers on the 3 suggested ranges assumptions:

  • Outside Lab Range (+/- 1.96SD) – forces on the bacteria analysis the belief that data is a bell curve — very false
  • Box-Plot-Whiskers – this method was created to deal better with data that is not a bell curve, but with the underlying assumptions of a skewed bell curve. – better but not ideal
  • Kaltoft-Moldrup – is the bacteria whisperer. It listens to the data and looks for atypical patterns. IMHO it produces the best identification of data of concern.

Conceptually the numbers under 10%ile and over 90%ile should be in the same ratio 1:1. What we see is shown below:

  • 2/2/2023 – 2.4 or 1.7 ratio
  • 11/1/2022 – 0.36 or 0.29
  • 4/11/2022 – 0.2 or 0.15

There was a major change, a flip in ratio with the latest sample. This was seen by the reader.

So is he better? IMHO — yes, the objective evidence is not as strong as I would like to see but:

  • None of the 3rd party ranges got worse (they remain unchanged)
  • His microbiome is producing a much richer amount of enzymes — which should cascade into a more balance system
  • The Kaltoft-Moldrup count continued to drop.

There are several events that adds noise to this analysis:

  • Going back to work (the fact that he continues to work must be viewed as solid evidence)
  • COVID and Paxlovid will alter the microbiome
    • I was unable to find any studies on the impact of Paxlovid on the microbiome 😞
  • Randomly tossing supplements into the mixture
  • There were huge differences in lab quality between samples (from 5.3 to 15.3)

Additional Analysis Points

Potential Medical Conditions Detected

The improvement seen in earlier post have persisted

  • 4/11/2022 : 26 items
  • 11/1/2022 : 5 items
  • 02/02/2023: 7 items

Bacteria Deemed Unhealthy

Nothing that is clear, randomness can explain the numbers well.

  • 4/11/2022 : 10 items
  • 11/1/2022 : 15 items
  • 02/02/2023: 12 items

Below we see the extreme %iles (0-9) improving over time. He went from a multitude of bacteria with token amounts to a more appropriate number with token amounts.

0 – 91040931859126
20 – 29342619552422
30 – 39342211383414
40 – 49141314192421
50 – 59161811212013
60 – 69241812282626
70 – 79172113233719
80 – 89211912312816
90 – 100261516432524

Going Forward

First thing is that my own experience in a significant ME/CFS flare was lots of swings in the microbiome results as I altered supplements and diet. The analogy that I often use is that the trip from the Port of ME/CFS to the Port of Health is not a straight flight like hoping on a plane (“as the crow flags”) but similar to travelling by a sailing ship that needs to make a lot of tacks (changes of boat directions) because of winds, shoals and reefs. A bad microbiome happens as a result of a long series of minor changes, we need to undo those.

Doing the usual trio of suggestions to build a consensus report.

In addition to that, we do some of the “has conditions and matches published literature” where the matching is over 95%ile

  • ME/CFS without IBS 12 of 18
  • Neuropathy (all types) 11 of 18
  • Hashimoto’s thyroiditis 8 of 17
  • Long COVID 68 of 167
  • COVID-19 39 of 96

This will produce a deep consensus report using 8 sets of suggestions. The downloads are attached



Looking at the consensus, we have 28 items that ALL of the suggestions agreed upon. These include:

A specific strain was recommended: Lactobacillus salivarius UCC118, but lactobacillus salivarius (probiotics) was a to-avoid — so I would skip that suggestion. The absence of any probiotics in this top list would inclined me to suggest skipping probiotics for the next round of doing and then testing.

The New Artificial Intelligence Diet (see More information) had #1 being … Cheerios!! — the reason was all of the fortifications added! Being gluten-free, then this is an easy to ignore.


I proceed to filter the diet suggestions by fruits first, and got the following:

When by Vegetables with this list:

Filtering by Roots, Tubers etc, had low values with only one nutrient contributing. Nuts etc had Sesame being the top item. Other items of note included: AvocadoBarleygrassCoca (as in source of cocaine – have fun asking for that at your health food store!).

On the AVOID list, we have:

What I find interesting is that vegetable/fruit juice-based diets, is pretty broad. Using the AI Diet, we can likely isolate which food and vegetables are the best choices and may well explain the “why” for the general vague diet term.

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 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.