Dear American Gut & uBiome: You Have Some Explaining To Do.

Uh oh.

This is concerning. Apparently, a science writer had her gut sequenced by the two popular gut sequencing services in the market right now — American Gut and uBiome. Not only that, but she had both done at the same time, from the same sample. Meaning, they should show the exact same results.

They didn’t. Not even close:

As you can see, the the ratio between the two predominant phyla, Firmicutes and Bacteroidetes, are almost flipped. FROM THE SAME SAMPLE.

Either one service’s sampling/sequencing method is faulty, or the general method that is being used by both services to do the gut sequencing is highly unreliable.

Either way, they’ve got some explaining to do. If this has legs, a lot of people will be — and should be — demanding their money back.

About these ads
Standard

57 thoughts on “Dear American Gut & uBiome: You Have Some Explaining To Do.

  1. Don Diego says:

    You need to compare table data side by side. Does uBiome provide table data? There are certain bacterial strains that are under “swing” classification. Actually more meaningful would be the comparison of rare taxa and those that are overrepresented compared to the population. That’s really the key, not so much the proportion at the phylum level.

    • Communicating phylum to genus level information is the primary feature of the products these services deliver. Just look at the presentation. If you want to argue that on a scientific level we should be looking at something else, you can, but it’s beside the point.

      • Don Diego says:

        You’re correct, Heisenbug, and the phylum is pretty much what most people will look at and they’ll leave it at that. But it’s confounding because it only tells half the story. And it’s unfortunate because it’s really at the sub-phylum level the presence of pathogenic bacteria seem to determine the disease state. You’ll never be able to tell who’s suffering from Crohn’s by looking at their phylum — there is very lilttle correlation; you have to go 2 levels down. In other words, ok, you have lots of Bacteroidetes, but how much of that is Prevotellaceae? You got Firmicutes aplenty. How much of that is Streptococcaceae? If these two vendors know what they’re doing, they should be identical within a margin of error. But more so from bottom up, not top down.

  2. Cooper Mor says:

    I’m glad this is finally being brought up.

    We’ve all read and seen a lot of research that suggest sick people can have significant dysbosis or altered microbiomes vs. healthy people. But, when I see sick people share their uBiome or American Gut samples – their bands are *usually* quite similar to the respective company’s standard status quo report. I have seen one from a person with IBS that had a massive (60-70%) proteobacteria band.

    Even disregarding Tina’s personal sample, if you look at their average results there’s inexplicable difference between the two.

  3. libfree says:

    I ask out of complete ignorance but is it possible for different sections of the same sample have different bacterial levels? Either way, I personally think its way too early to be able to garner much from these services. Your better making changes to your diet and seeing how those might effect you personally.

    • If that were the case, then these services would be utterly useless. I would hope that isn’t the case. I do agree with your second point — diagnostics are really just an adjunct to self-experimentation.

      • libfree says:

        I’d guess sample contamination or a mix up of some sort. I don’t think that there is a lot of information to take away at this junction. If we really want to know, we’d conduct this same experiment among several of us and see if we have the same problem. I’m a realist and realize that there will be an error rate, we just hope its a small one.

      • tatertot says:

        OK Everybody…money where your mouth is! I just ordered a $99 AmGut and an $89 UBiome kit. Anybody else?

      • Don Diego says:

        I actually ordered uBiome a week before (both saliva and fecal). If I had known Tina’s discrepancy, I probably would have just ordered AGP.

      • Allan Folz says:

        Tim, at 40g of RS a day, what is that — +4, +5 sigma? — your results are idiosyncratic enough if they drop your swab on the mailroom floor and try faking some numbers it’s going to be noticeable.

      • You know, it’s funny that this should come up. I just came back from a camping trip to places quite remote. Conditions were primitive and the morning routine involved a quiet place and a shovel. There were two very distinct components of one day’s movement, one dark and one light in color. Obviously related to the inputs of the prior day but nevertheless, both sections were quite different in color, consistency, and visual texture. (That’s as far as my analysis went – visual assessment only.)

        To account for such differences, perhaps mixing before sampling is in order.

  4. Thanks for posting this. uBiome does provide information at all taxa levels, not in a table form, but in an interactive online form. They allow you to see how your sample compares to other uBiome customers. American Gut provides a data table. None of numbers agree at the phylum, genus or family levels.
    It’s as if I were two different people. Which makes me wonder, “did one of these places mix up my sample with someone else?”
    If either of these is correct, I’m a bit of freak microbiome-wise, although I do seem to fall within the “normal” range as reported by uBiome.
    Here is the story I wrote for Science News on the initial results from uBiome. https://www.sciencenews.org/article/me-and-my-microbiome

    The only fair way to settle this may be to get a re-do.

    • Allan Folz says:

      I think a re-do is definitely in order. I also think sending two or three anonymous and identical samples to each service would also be in order.

      I can appreciate this stuff being on the bleeding edge, but the core foundation of science is reproducible results. If none of this is reproducible (and how much have the services done to test themselves in this regard), we’re just playing parlor games and reading tea leaves.

      Glad it’s getting some mileage beyond twitter. I was disappointed the way uBiome didn’t respond at all. Also, Am Gut, after a few honest attempts, dropped the topic. Yeah, its twitter, but don’t leave us hanging. Good luck Tina in trying to get any follow-up and closure to this. Definitely appears to be grist for another article.

      • Tatertot says:

        Tina – Want to make a bunch of basement dwelling gut geeks really happy? Post your AmGut Taxa report here. You can do it right in a comment, just cut and paste from the AmGut website. It looks ugly, but we know how to read it.

        From what I’ve seen, AmGut is spot on as far as the diet and health of the person and the taxa report. Would love to see yours. Also, if you can do the same with the UBiome report, that would be amazing!

        Thanks!
        Tim

    • Hi Tina. Thanks very much for raising this issue, and for providing some more context here. uBiome and American Gut really do need to address this — it raises some serious questions about the utility and veracity of these tests. I don’t know if they have a dialogue with one another, but they should — it’s probably the only way to get to the bottom of this. Until then, whatever we get from them is pretty suspect. I agree that a re-sampling would help. Though even one time is enough to signal a problem. And I agree with Allan that this should probably be done multiple times with multiple identical samples.

    • libfree says:

      The number of samples that get switched in a perfect organization > 0. Impossible to completely avoid error. We just hope that the percentage is really small.

  5. Does the data from either project include confidence intervals or error estimates? It could be that both estimates are “correct” within a margin of error.

    It would be important to know whether any differences are due to randomness (which would suggest that the data is not useful, at least not for individuals) or to some systematic effect such as differences in methodology (in which case the results may still be useful).

    • Michael, I think what you say is spot on. If it’s randomness, which is what I worry about, then it really calls into question the usefulness for individuals. If it’s a systematic effect due to methodology, then we should know what that is. And that’s what I have asked of them. After all, we don’t have two standards for cholesterol and blood glucose testing.

      With regard to margin of error, it sounds to me like the results are different enough that, if they did both fall into a margin of error, then the results — or at least how they are presented to people — are useless.

  6. Penny says:

    Maybe it was the amount of poo on the sample… we had to redo our first AG sample because it had “too much poo” on the stick – yet, weirdly, they had results for both the original and the redo… and they were totally different – the first one had 38% Ruminococcaceae and 16% Dialister and the second had 38% Bacteroides and 11% Ruminococcaceae – however you spell that:-) I like the mix it up first idea myself…

  7. Lori2 says:

    I know next to nothing about biology and even less about testing so this question may well display my ignorance, but I’ll ask it anyway. I was reading the transcript of an interview Chris Kresser had with a British doctor concerning fecal transplants. The doctor stated that one problem with the do-it-yourself at home method was that 90% of the bacteria are anaerobic and therefore die before they can be implanted. How would this fact impact stool analysis? Can dead bacteria be identified as easily as living bacteria?

  8. Danae says:

    Thank you for posting this information. I agree that the discrepancy raises important questions about the accuracy, within- and between-sample variability, and reproducibility of this type of testing. However, I disagree with your calling American Gut a “gut sequencing service,” and implying that there should be some type of money-back guarantee. I submitted a sample to AmGut last August, and my impression was that I was submitting a sample to a research project, and donating money to further that research. The fact that they send me back information on my personal results is an extremely cool, but incidental, aspect of the project. (I participated in a university research project several years ago, in which I donated several vials of blood and answered questions about a specific health condition that I have. In that case, my only feedback was being able to see the aggregate data in the final published paper; there was no personal data at all, and I think that’s not unusual for research.) Anyway, I do think that the questions are important, and I hope that we see some answers.

    • Don Diego says:

      Danae, you agreed to provide your fecal sample which will be used by AGP for further research down the road. But you did so by providing AGP with compensation and you’re expecting to receive results which you’ll interpret on your own. It’s contractual and it goes two ways. It’s entirely reasonable to expect accurate results based on your contribution of confidential fecal sample and compensation.

      This is supposed to be a research project but it’s also a means of financing these organizations. And there are concerns what they might do down the road with the samples, which are your fecal DNA. They’ve said the information is confidential and will not be shared. But there’s no guarantee that they won’t use the information (on an anonymous basis) with biotech companies for drug development down the road. Or they could partner with health and life insurance companies to better identify risky patient populations. There are many for-profit ventures which could arise from their data accumulation; this is probably the initial stage where such services parade as a “research project” on a non-profit platform.

      In fact, the only way these guys are able to finance themselves is through this direct patient model: you need a vast sample to analyze the data. You can’t simply receive financing and do this, as you need data. That’s probably why this so called “research project” and the “$100 contribution in exchange for fecal analysis” became a plausible business model. Some of the cluelessness and the naivete you encounter is just mind-boggling.

  9. I’ve been chatting with a buddy in Bioengineering here at Stanford and they’ve developed some interesting new ways of measuring the microbiome — smarter, cheaper, faster, and all that.

    We were debating some of the parameters of getting your microbiome sequenced and this post hits on some of the quality issues that I think are important to people.

    I put together a survey to try to find out what people who are getting sequencing find important. https://docs.google.com/forms/d/1KpFfEBQVT0D9r-4ke6ZhLWQe9A3caPjVrAsFdWgChQ8/viewform

  10. Danae, I also think of AG as primarily a research project to which I am contributing. My problem is the larger issue of being able to reproduce results. The microbiome is an extremely hot area of research right now and people are having difficulties reconciling their study results with others. That really has to happen if there is ever going to be a clear picture of what is happening with the microbiome.
    My gut microbiome has been sequenced twice, but I still have no idea of what is really in there. I could get tested again, but the new sample won’t clear up which of the other two was “correct” as I’ve had antibiotics to treat bronchitis since my last test.
    I’d fully expect to see minor discrepancies, but it seems that the overall pattern should be the same.

    You asked for the data, and here are my top uBiome Family level taxa results
    Ruminococcaceae 26.5%
    Lachnospiraceae 22.6%
    Bacteroidaceae 20.3%
    Streptococcaceae 3.67%
    Clostridiales Family XI Incertae sedis 3.48%
    Porphyromonadaceae 2.46%
    Rikenellaceae 1.47%

    And uBiome Genus level
    Bacteroides 20.3%
    Faecalibacterium 12.2%
    Blautia 9.28%
    Ruminococcus 4.81%
    Streptococcus 3.67 %
    Roseburia 2.05%
    Sutterella <0.01%

    And here's the American Gut table:
    Kingdom Phylum Class Order Family Genus Relative Abundance (%)
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides
    46.92
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae —
    9.02
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae —
    8.06
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Rikenellaceae —
    6.34
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcus
    4.78
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Porphyromonadaceae Parabacteroides
    2.51
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium
    2.37
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia
    2.21
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales Alcaligenaceae Sutterella
    1.72
    Bacteria Proteobacteria Alphaproteobacteria RF32 — —
    1.55
    Bacteria Firmicutes Clostridia Clostridiales — —
    1.51
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Lachnospira
    1.29
    Bacteria Verrucomicrobia Verrucomicrobiae Verrucomicrobiales Verrucomicrobiaceae Akkermansia
    1.23
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae [Ruminococcus]
    0.94
    Bacteria Firmicutes Bacilli Lactobacillales Streptococcaceae Streptococcus
    0.92
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Roseburia
    0.88
    Bacteria Bacteroidetes Bacteroidia Bacteroidales [Odoribacteraceae] Butyricimonas
    0.84
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Succiniclasticum
    0.55
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Dorea
    0.55
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Porphyromonadaceae Porphyromonas
    0.55
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Peptoniphilus
    0.47
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Lachnobacterium
    0.41
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Coprococcus
    0.39
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] WAL_1855D
    0.35
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae —
    0.31
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Prevotellaceae Prevotella
    0.31
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] 1-68
    0.27
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Oscillospira
    0.27
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Finegoldia
    0.25
    Bacteria Bacteroidetes Bacteroidia Bacteroidales [Odoribacteraceae] Odoribacter
    0.25
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] ph2
    0.20
    Bacteria Fusobacteria Fusobacteriia Fusobacteriales Fusobacteriaceae Fusobacterium
    0.14
    Bacteria Proteobacteria Epsilonproteobacteria Campylobacterales Campylobacteraceae Campylobacter
    0.12
    Bacteria Proteobacteria Deltaproteobacteria Desulfovibrionales Desulfovibrionaceae Bilophila
    0.12
    Bacteria Actinobacteria Actinobacteria Actinomycetales Corynebacteriaceae Corynebacterium
    0.12
    Bacteria Bacteroidetes Bacteroidia Bacteroidales S24-7 —
    0.10
    Bacteria Proteobacteria Gammaproteobacteria Pasteurellales Pasteurellaceae Haemophilus
    0.08
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Anaerococcus
    0.08
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae 02d06
    0.08
    Bacteria Proteobacteria Gammaproteobacteria Pasteurellales Pasteurellaceae Aggregatibacter
    0.06
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae [Eubacterium]
    0.06
    Bacteria Firmicutes Clostridia Clostridiales Peptostreptococcaceae —
    0.06
    Bacteria Firmicutes Clostridia Clostridiales Peptococcaceae —
    0.06
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Anaerostipes
    0.06
    Bacteria Firmicutes Clostridia Clostridiales Christensenellaceae —
    0.06
    Bacteria Firmicutes Bacilli Gemellales Gemellaceae —
    0.06
    Bacteria Actinobacteria Coriobacteriia Coriobacteriales Coriobacteriaceae —
    0.06
    Bacteria Actinobacteria Actinobacteria Actinomycetales Actinomycetaceae Varibaculum
    0.06
    Bacteria Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Proteus
    0.04
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Veillonella
    0.04
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium
    0.04
    Bacteria Firmicutes Bacilli Lactobacillales Enterococcaceae Enterococcus
    0.04
    Bacteria Actinobacteria Actinobacteria Actinomycetales Actinomycetaceae Actinomyces
    0.04
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae Collimonas
    0.02
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae —
    0.02
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Dialister
    0.02
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Anaerotruncus
    0.02
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae SMB53
    0.02
    Bacteria Firmicutes Bacilli Lactobacillales Streptococcaceae Lactococcus
    0.02
    Bacteria Bacteroidetes Bacteroidia Bacteroidales [Paraprevotellaceae] [Prevotella]
    0.02
    Bacteria Actinobacteria Coriobacteriia Coriobacteriales Coriobacteriaceae Adlercreutzia
    0.02
    Bacteria Actinobacteria Actinobacteria Actinomycetales Actinomycetaceae Mobiluncus
    0.02

    • Tatertot says:

      Tina! You made my day.

      Here’s what I think is going on. AmGut has a bigger database of sequences to check against when determining what they are finding in the sample.

      The evidence is in the numbers, somewhere, there should be a list that adds up close to 100%. The two lists from UBiome add up to 80% for the Family level data and only 52% for the genus level. From what I can gather, 16s rRNA works by identifying microbes of a unique species or genus and the family and higher taxa then needs to be determined.

      With the AmGut report, the genera listed add up pretty close to 100% without using a calculator, 95%+ for sure. So even they can be off a bit.

      Since UBiome couldn’t determine the genus of almost half of your sample, I’d say there report can only be considered 50% complete.

      16s rRNA is supposedly used by both companies, but the report is only as good as their equipment and database. My guess is that UBiome has cheap equipment and an incomplete database.

      My rationale for trusting AmGut over Ubiome is that only AmGut shows a list that adds up to nearly 100% of genera, meaning they have accurately identified nearly everything in your sample.

      From: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2045242/
      One of the most attractive potential uses of 16S rRNA gene sequence informatics is to provide genus and species identification for isolates that do not fit any recognized biochemical profiles, for strains generating only a “low likelihood” or “acceptable” identification according to commercial systems, or for taxa that are rarely associated with human infectious diseases. The cumulative results from a limited number of studies to date suggest that 16S rRNA gene sequencing provides genus identification in most cases (>90%) but less so with regard to species (65 to 83%), with from 1 to 14% of the isolates remaining unidentified after testing (5, 11, 17). Difficulties encountered in obtaining a genus and species identification include the recognition of novel taxa, too few sequences deposited in nucleotide databases, species sharing similar and/or identical 16S rRNA sequences, or nomenclature problems arising from multiple genomovars assigned to single species or complexes.

      From: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1594676/

      First, in this era of managed care in which clinical microbiological laboratories are obliged to contain costs, the cost of 16S rRNA gene sequencing remains a major obstacle. The laboratory can save costs by purchasing directly from providers of generic primer synthesis, general molecular reagents, and sequencing services, since they operate at a competitive market. In contrast, the cost to gain access to commercial databases for 16S rRNA gene sequencing remains high and hampers its implementation in a routine clinical microbiology laboratory. To make analysis of sequencing data affordable, use of public databases such as GenBank is an alternative. This brings us to the second problem: the deposited sequences in the public databases are not peer reviewed. Since anyone can deposit and name sequence data, the quality of entries in GenBank is compromised by sequencing errors and ambiguities, incomplete sequences, and insufficient strain characterization (6, 9). On the other hand, the quality of commercial databases is also not optimal (4, 5). This is more related to the limited number of strains in the databases rather than incorrect sequence entries. Third, although sequence data are objective in nature, the interpretation of these data to establish a species designation is not. There are no universally accepted criteria for 16S rRNA gene sequence-based identification of bacterial isolates. The level of sequence homology required between the isolate under investigation and those in the databases to conclude on genus and species designation is a matter of debate (3). This introduces a level of subjectivity in sequence-based species identification and, as a consequence, potential inaccuracies in the routine laboratory setting.

      • I only gave you the top 10 results from uBiome. The report goes much, much deeper and does give you 100% reporting at every taxon level. It’s just in a format that is not easy to copy and paste and I didn’t feel like writing down every single one because the list is very, very, very long.
        I find it much more complete than American Gut’s list as it is set up to explore ever deeper, sometimes down to the species level. With their “tree” interactive you can see the taxonomic relationships between the bacteria. And you can explore your results with other groups much more fully. As consumer experiences go, uBiome is far smoother and more satisfying. It’s great fun. I just don’t know if it’s right.

    • Tatertot says:

      And in looking at the AmGut Genus report, I see no bifidobacteria and very little proteo or actino. Good news on that is, not many pathogens but bad news is, not many beneficial microbes either.

      The bacteria you have in the Firmicutes and Bacteroidetes phyla are probably neither bad nor good, just regular gut bugs that everyone has irregardless of health status and diet-dependant, these could shift back and forth with the amount of fiber, grain, meat, etc.. you eat. The pH of your gut is probably a bit higher than it should be which is why the lacto and bifido are absent.

      I would think a good bit of added fermentable fiber in your diet, aka prebiotics, and taking some probiotics that contain lacto and bifido would help set you up for better gut health even if you don’t really have ‘problems’ now. Your gut profile maybe a bit better than most, due to the lack of actinobacteria, but also less diverse, due to lack of actino and proteo. Not all actino are bad, e. coli. for instance, but you have none of that, either.

      One thing I notice right off is that the UBiome report doesn’t identify anything that is not Firmicute or Bacteroidete flavored except the proteobacteria Sutterella at .01% which everyone seems to have. The AmGut shows you have representation from all the phyla, just at very low amounts.

  11. dugan says:

    Hi.
    I recently got my AmGut results and have been canvassing the internet trying to figure out what it means. I’m very happy to have stumbled upon this site, and have enjoyed following these threads. While I’ve learned a lot, I selfishly seek input from anyone who cares to share their thoughts. In summary: 53% B, 45% F, .8% ProtoeB, .7% ActinoB. 13x Prevotella, 14x Anaerostipes, 12x Gemellales, 5x Clostridium.

    I’m 47 male who was a vegetarian for 12 years, then a pescatarian for 10 and a meat-eater for the last 5 (though don’t eat a lot of meat). I eat lots of fruit, beans and vegetables, modest kimchi but also have a sweet tooth (working on it) and eat plenty of pizza, pasta, beer. Of concern – my FBS is creeping up (100-105?) and these past few years my immune system isn’t functioning as well – frequent cold sores and get colds/flu more often than I use to.

    Some specific questions:
    1. I have an exceedingly high amount of Prevotella. Besides grain/sugar consumption, significance?
    2. I seem to have a decent amount from the desirable Clostridia family, but maybe not enough?
    3. How to build immunity?

    Comments welcome. Thank you!!

    Dugan
    p.s. embarking on a potato startch/kefir/kimchi experiment.

    Kingdom Phylum Class Order Family Genus Relative Abundance (%)
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Prevotellaceae Prevotella
    39.17
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae —
    11.22
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides
    9.35
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae —
    8.82
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium
    4.63
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae —
    2.25
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia
    2.19
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae Clostridium
    2.13
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Lachnospira
    2.01
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Anaerostipes
    1.66
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Porphyromonadaceae Parabacteroides
    1.63
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Coprococcus
    1.42
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae [Ruminococcus]
    1.41
    Bacteria Bacteroidetes Bacteroidia Bacteroidales S24-7 —
    1.23
    Bacteria Firmicutes Bacilli Lactobacillales Streptococcaceae Streptococcus
    1.23
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Rikenellaceae —
    1.18
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae [Eubacterium]
    1.15
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Roseburia
    1.11
    Bacteria Firmicutes Clostridia Clostridiales — —
    0.78
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] WAL_1855D
    0.64
    Bacteria Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus
    0.53
    Bacteria Firmicutes Clostridia Clostridiales Lachnospiraceae Dorea
    0.46
    Bacteria Firmicutes Clostridia Clostridiales Ruminococcaceae Oscillospira
    0.41
    Bacteria Actinobacteria Actinobacteria Actinomycetales Corynebacteriaceae Corynebacterium
    0.37
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Peptoniphilus
    0.27
    Bacteria Proteobacteria Gammaproteobacteria Pasteurellales Pasteurellaceae Haemophilus
    0.25
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales Alcaligenaceae Sutterella
    0.25
    Bacteria Bacteroidetes Bacteroidia Bacteroidales [Paraprevotellaceae] Paraprevotella
    0.24
    Bacteria Actinobacteria Actinobacteria Bifidobacteriales Bifidobacteriaceae Bifidobacterium
    0.24
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae —
    0.16
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Veillonella
    0.15
    Bacteria Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus
    0.15
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Anaerococcus
    0.15
    Bacteria Bacteroidetes Bacteroidia Bacteroidales Porphyromonadaceae Porphyromonas
    0.12
    Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Sphingomonadaceae Sphingobium
    0.10
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] Finegoldia
    0.10
    Bacteria Cyanobacteria Chloroplast Streptophyta — —
    0.10
    Bacteria Bacteroidetes Bacteroidia Bacteroidales [Barnesiellaceae] —
    0.10
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Dialister
    0.09
    Bacteria Actinobacteria Coriobacteriia Coriobacteriales Coriobacteriaceae Collinsella
    0.09
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae Herbaspirillum
    0.07
    Bacteria Firmicutes Clostridia Clostridiales [Tissierellaceae] 1-68
    0.04
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae Holdemania
    0.03
    Bacteria Firmicutes Clostridia Clostridiales Clostridiaceae SMB53
    0.03
    Bacteria Firmicutes Bacilli Lactobacillales Enterococcaceae Enterococcus
    0.03
    Bacteria Firmicutes Bacilli Bacillales Planococcaceae —
    0.03
    Bacteria Bacteroidetes Sphingobacteriia Sphingobacteriales Sphingobacteriaceae Sphingobacterium
    0.03
    Bacteria Actinobacteria Actinobacteria Actinomycetales Actinomycetaceae Actinomyces
    0.03
    Bacteria Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Proteus
    0.01
    Bacteria Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Morganella
    0.01
    Bacteria Proteobacteria Betaproteobacteria Burkholderiales — —
    0.01
    Bacteria Proteobacteria Alphaproteobacteria Sphingomonadales Sphingomonadaceae Sphingopyxis
    0.01
    Bacteria Firmicutes Erysipelotrichi Erysipelotrichales Erysipelotrichaceae Allobaculum
    0.01
    Bacteria Firmicutes Clostridia Clostridiales [Mogibacteriaceae] —
    0.01
    Bacteria Firmicutes Clostridia Clostridiales Veillonellaceae Phascolarctobacterium
    0.01
    Bacteria Firmicutes Clostridia Clostridiales Peptostreptococcaceae —
    0.01
    Bacteria Firmicutes Bacilli Lactobacillales Carnobacteriaceae Granulicatella
    0.01
    Bacteria Firmicutes Bacilli Gemellales — —
    0.01
    Bacteria Actinobacteria Coriobacteriia Coriobacteriales Coriobacteriaceae Slackia
    0.01

    • Tatertot says:

      Dugan – thanks for pasting that all here. I love these reports. I’ll bet that the bar chart just shows red and orange sections with a feint rainbow sliver underneath, similar to Pollan’s. To me, this is good and bad. Good in that you haven’t been overtaken by monster pathogens, but bad in that you may be setting yourself up for a hostile takeover should your immunity weaken or you get in an accident or need antibiotics.

      You have all the RS degraders and even some bifidobacteria. I’d go ahead and try your RS experiment and see how you feel doing it. Try to think about all the prebiotics, though, and don’t limit to just RS. Green bananas, dandelion greens, raw carrots, onions and garlic, if you are into supplements, there’s inulin, pectin, psyllium, and potato starch. Shoot for 2-4 TBS of a combo a day as well as eating cooked/cooled rice, potatoes, beans, and green bananas. I have a feeling you’ll respond well.

      Also consider some probiotics like Shant talks about around here…L. plantarum, fermented foods, etc…

      Here’s a couple interesting gut bugs you own. Nothing to worry about, just interesting:

      Sphingobium species are different from other sphingomonads in that they are commonly isolated from soil, however Sphingobium yanoikuyae was isolated from a clinical specimen. They can degrade a variety of chemicals in the environment such as aromatic and chloroaromatic compounds, phenols like nonylphenol and pentachlorophenol, herbicides such as (RS)-2-(4-chloro-2-methylphenoxy) propionic acid and hexachlorocyclohexane, and polycyclic aromatic hydrocarbons.

      Herbaspirillum: A genus of gram-negative bacteria in the family OXALOBACTERACEAE, comprised of vibrioid or sometimes helical cells. They are chemoorganotrophic nitrogen fixers and are found free-living in the soil or in association with the roots of members of corn, rice, wheat.

    • Hi dugan. The most interesting thing to me about your report is the same thing you find interesting: your Prevotella dominance.

      I don’t believe in trying to use these reports to assess health at all, but if someone has a specific health issue that they already know about or are concerned about, then it makes sense to use it as a “detective tool.” And it does seem you have one thing you are concerned about: your FBS.

      Therefore, if this were my report and I had some concerning FBS issues, I’d play with the Prevotella. How? Well, the theory right now is that Prevotella can be a result of high grain consumption. Like wheat. And it does sound like you consume a fair amount of wheat/grains. Perhaps to the point where most of your fiber comes from wheat. Leading to an imbalance. The issue is that Prevotella has been linked with inflammatory disease (perhaps something that may contribute to blood sugar dysregulation).

      So if this were me, I’d try dropping wheat/grains or at least scaling back, and then upping different types of vegetable-derived fibers (RS, inulin, pectin, etc.). See what happens with the FBS. Maybe get another gut test done. This isn’t about going low carb, by the way. Not at all. It’s a wheat/grain specific thing (corn would qualify as a grain).

      Let us know what happens if you try it out.

  12. How would it even be possible that the results are so different? Unless there was a major mistake made…What is terrifying is the fact that some people may have been given false results and have therefore been given false information about their health. Awful.

    • Tatertot says:

      I was thinking about this last night. The UBiome should be advertised as a “random sampling of gut bugs”. And AmGut could probably be called, “a fairly complete sampling of your gut bugs.”

      Neither of these should be used to diagnose of treat health problems. Just for fun. If a result shows a huge pathogen load, get a doctor involved.

      • David H says:

        Hi Tim, Ubiome finished processing my results today and I can access a “dashboard” that is being beta tested. My results are broken down into phylum, class, order, family genus and sometimes species. Spent a few hours compiling this information into a Word document, that I’m happy to share with you if this is still of interest. If so, what is the best way to contact you?

  13. Dear Mr. H, I (attempted) to post this under one of your Eczema posts, but I wasn’t signed in, so I’m not clear if it posted or not, please forgive me if it is a repeat and as it covers a number of ‘gut biome’ topics, may be off topic for this particular post. I plan on repeating my American Gut test in a few months. Was going to try uBiome, but maybe I will do both at once and see what happens! Here is a copy of the previous post:

    Love your blog. I am NEVER sick, but 9 months ago had to have an emergency appendectomy – so I have been really studying up on gut flora. As an interesting side note, I got the attack within a week of having a crown on my R bottom molar…hymn…
    FINALLY got back American Gut results back (took 6 months). Hazda like – makes sense as I eat virtually no grains and limited milk products, happy to have lots of butyrate producers, but trying resistant starch experiments to see if I can increase the bifidos. Previously, my primary carb source was non-starch vegs – so depending on how you count their carb calories, I was on a pretty low carb diet (under 100 grams total per day) and Jaminet suggests deducting ALL veg carbs)

    QUESTIONS: 1) So does butyrate from butter actually make it to the large intestine?

    2) I had /have intense abdominal distention from increasing load of BRM potato starch combined with eating more fermented foods and taking probiotics (I don’t usually take them), so I have scaled back to dehydrated green plantain and green banana chips. Some bloggers write that if you have major gas to Inulin (I do), it is a confirmation of SIBO. I am wondering if this (SIBO) is my problem and wondering ways to confirm this short of the lactulose breath test? I did a 1 day experiment of eating mostly lamb and about 3/4 pound of pureed asparagus and I was fine, so I will order the FOS product and see if it works better for me than the inulin. The distention could also be low stomach acid (per the ‘baking soda’ home test). I have been experimenting with Betaine HCL (when 6 oz meat consumed) and some other digestive enzymes, as well as stomach acid stimulators such as ginger, ACV and bitters. Not very controlled experimentation I know – I need to settle down on that, but given my excitement and symptoms, it is hard to control everything at once! (FODMAPS vs. SCD type info), etc.
    3). My final question/comment is the “exact gut bacteria discovered for Crohn’s” article in the comments. The articles didn’t link to any exact bacteria, but seemed to report the already known dysbiosis of Crohns/IBD. Is it (from different studies): IBD patients had lower F. prausnitzii abundance than H and IBS (P<0.001). CD patients showed higher E. coli counts than H and UC patients.(H=controls)
    Also, I have read from other studies there is a shortage of butyrate producers in IBD's, which would lead to fewer TREG cells in the GALT. Would really appreciate clarification anyone has. Thank you so much for the time and effort you put into your blog and experimentation!
    p.s. sorry for the jargon/acronyms, but your readership seems well informed and I think will recognize most of them.

  14. dugan says:

    Thank you Heisenbug and Tatertot for your replies. Helpful info. I have definitely scaled back on bread/pasta/desserts and increased beans and fish so hope to see some changes in time. I generally have a favorable (low) CRP, but would that indicate low inflammation in gut, too? I’ve done the PS for a week w/ 4 Tbs in kefir. Ordered an inulin/FOS probiotic and bought a 3 gallon tub of kimchi, so my n=1 is gearing up. Some promising BS readings but also confounding variables so will wait a few weeks to report back. Thanks again for info!

  15. Penny says:

    So, this is a response from American Gut because each time I go into my son’s taxa report online, I find a different set of bugs… some are the same, but this time there appeared to be more of a variety…

    “As we gather more data, we occasionally re-analyze past
    data. When we re-analyze the data, we use a technique called
    rarefaction. Rarefaction normalizes the counts for each
    sample, such that an equal sampling effort can be assumed.
    It is inherently random, which results in small differences
    in your sample microbe abundance percentages. You can read
    more about rarefaction if you would like here: http://en.wikipedia.org/wiki/Rarefaction_(ecology).

    As such, I wouldn’t say that one is correct and one is
    incorrect. Right now it isn’t possible to say that our
    profiling of the microflora paints a 100% accurate picture
    of what is present. The accuracy depends on how advanced our
    molecular techniques and data analysis methodologies
    are.”

  16. I haven’t gotten my uBiome sample analysis back yet (and I submitted the samples at different times), but I think one factor may be in how the calculations are made.

    For example, my official American Gut report was fairly different than the analysis I did on my raw .fastq files in MG-RAST:

    http://cdwscience.blogspot.com/2013/11/my-american-gut-individual-report.html

    Basically, certain bacteria are known to grow in the tube for the shipping process, so the American Gut calculations try to correct for this computationally. At least for my sample, I think the computational correction was probably too strict (for reasons that I explain in the above blog post). In contrast, the uBiome samples are submitted with a chemical that is supposed to stop the bacteria from growing, without harming the bacteria DNA. So, my guess is that this will have at least some effect.

    Does uBiome provide access to the raw data? I am curious how similar the uBiome “official” report is to a report generated from MG-RAST (for example)

  17. Mr. Heisenbug,

    In case you haven’t seen it, here is our (uBiome’s) blog post from a couple weeks ago looking into this question. http://www.ubiomeblog.com/?p=92 Sorry I didn’t post this here earlier!

    This is a really interesting question, so thank you for raising it. Please don’t hesitate to let us know if you have any more questions or suggestions. Great work you’re doing here!

    Zac

  18. Pingback: Links 5/17/14 | Mike the Mad Biologist

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s