So a brief comparison of results.
Overall they concluded that for the ME/CFS +IBS group, decreased Faecalibacterium and increased Alistipes was the best indicator while for ME/CFS -IBS, increased unclassified Bacteroides and decreased B. vulgatus was the best marker.
Interestingly, at my very first test, when possibly IBS might be considered a slight issue, I did fit the low Faecalibaterium high Alistipes category, though I'm not convinced this was not just chance. Both these genera have varied widely over time, soetimes high sometimes low and not in any relationship to each other.
Unfortunately it is impossible to consider the marker for the -IBS category. B vulgatus is a species reliably detected by uBiome but I have no way of knowing what is contained in Lipkin's unclassified Bacteroides category so have no way of comparing.
I mention this because that first test with low F and high A is the only concordance I can find but I don't think it really means much.
If I look at other patterns discerned by the different analytical techiques used I don't find much agreement.
TDA analysis finds at the family level decreased Lachospiraceae and Porphyromonadaceae and increased Clostridiaceae. In every test I have had, I have the exact opposite pattern.
At the genus level, this analysis found decreases in Dorea, Faecalibacterium, Coprococcus, Rosburia and Odoribacter and increases in Clostridium and Coprobacillus.
In my gut Dorea and Clostridium are consistently low and Coprococcus and Coprobacillus are virtually non-existent. Faecalibacterium, Roseburia and Odoribacter are very variable - sometimes high sometimes low - and as far as I can tell there is no reason for the variability. So the pattern of my test would depend on when the test was done but this doesn't seem to bear any relationship to the state of my disease. It seems to be more random.
Similarly with conclusions from the linear analyses. While a few of the genera found to be low are also low in my gut, others of the low category are very variable for me while those of the high category are not in me.
I have looked at the analysis of metabolic pathways. I can't find much accord here either.
I can find only a few categories reported by uBiome that seem to exactly coincide with those of Lipkin. These are vitamin B6 metabolism, atrazine degradation (both of which were increased) and biosynthesis of unsaturated fatty acids (decreased)
My unsaturated fatty acid synthesis is slightly decreased (around 0.9 average), B6 metabolism is average to slightly below average and atrazine degradation is a bit lower (around 0.7).
What are we to make of all this?
Well first I suspect that Lipkin's average is very different from that of uBiome. His control group consisted of 50 people. uBiome's database is very much larger - many, many thousands of people. Because of the enormous individual variability of microbiome composition, the size of the group could have an enormous influence on average composition.
This may be reflected in Lipkin's reporting of the break-up of the two main phyla which together constitute more than 90% of gut inhabitants. He find 63.4% Bacteroidetes, 29.7% Firmicutes in controls (patients are very similar). This was very surprising to me in light of the many gut studies I have looked at.
The relatively high fat, high simple sugar Western diet greatly favours Firmicutes. Bacteroidetes thrives on high fibre unprocessed plant foods and high levels are typically found in traditional, agrarian societies. In Western societies, the ratio is almost always the other way around and often Firmicutes greatly exceeds Bacteroidetes.
The uBiome average is about 60%Firmicutes, 31% Bacteroidetes.
Whether the discrepancy is just related to this particular group of controls or whether it has something to do with the different technology used is impossible to say.
Potentially big differences in averages could completely negate the kind of comparison I have tried to do.
In any case, I don't think it is particularly helpful to focus on the list of organisms. Yes it is necessary to accumulate this info to understand if there are differences in overall composition but that is as far as it goes.
The significance doesn't lie in the actual organisms present. There is no single ideal pattern, there are many possible combinations, all of which work because of the considerable redundancy in the microbiome. Many different organisms can fulfill the same function.
So all of the studies showing differences in composition in various disease states are just the first step. Working out the meaning of the difference will be far more important and will be much harder to do.