Jump to content

Recommended Posts

µBiome is offering half off their basic sequencing service. I was thinking about doing it but I googled around and found a lot of people wondering how mature the technology - or at least our/their ability to interpret the raw data - is. For example, this article: "Here’s the poop on getting your gut microbiome analyzed".

 

Has anyone gotten their gut microbiome analyzed and learned anything worthwhile, for ex., something "actionable"?

 

Zeta

Share this post


Link to post
Share on other sites

Zeta,

 

I have, and I just got a second one (still being processed), via their recent offer of a free kit (which I posted about but which was unfortunately quickly oversubscribed). See this recent post on my perspective on uBiome, and in particular the Reddit Q&A I engaged in with the uBiome Co-founder on the current usefulness (or lackthereof) of their testing.

 

So if I'm skeptical about the usefulness of gut microbiome analysis, why am I doing it (again)? 

  1. It was free  :)xyz
  2. A couple months ago I started taking a probiotic for the heck of it, and I want to see if its made any difference in my gut bacteria population distribution.
  3. uBiome is one big citizen science project, and I'm interested in contributing to any such effort I can, with the hope that insights will be forthcoming based on my data combined with that from others.

Interestingly, the co-founder was really being sincere in the Reddit AMA when she mentioned they were working on better analysis tools. Today I got a message that they'd made substantial improvements to the interpretation tools available on their website. Sure enough, they do now provide some interesting additional analysis. Here is how they described the changes in their blog post about it:

 

We’re proud to announce some exciting new additions to the uBiome website:
 
1. Bacterial functions. How well do your bacteria process caffeine compared to other people? What do your carbohydrate and fat metabolism capacities look like? We’ve added 106 bacterial functions for you to explore in your gut microbiome data, and we’re giving everyone this new feature for free.
 
Login at http://app.ubiome.comand let us know what you think!
 
2. Diversity score. Is your gut microbiome more or less diverse than all other samples in the uBiome dataset? The chart below shows my actual data. In the gut, bacterial diversity is generally positive, so I guess I’m in good shape, diversity-wise.

 

So here are the new results I got for my original sample when I now go to there website and log in.

 

First, the new "Bacterial Function" stuff - which is basically a comparison between the number of bacteria I have vs. other sample populations (e.g. other vegans) that are associated with various metabolic functions. My data is the blue bars, and in the first graph the gray bars represent everyone who has had their gut microbiome sequenced by uBiome:

 

post-7043-0-93723800-1446147117_thumb.jpg

 

This is the graph of my data vs. other vegans who've had their microbiome sequenced:

 

post-7043-0-80473300-1446147387_thumb.jpg

 

Here is how they describe these bacterial function comparisons:

 

Your Predicted Functions
 
You are currently looking at predictions of the molecular pathways present in your microbiome. The graph above shows predictions of selected molecular pathways in your microbiome normalized to the average across all uBiome users. To the right of this box is a bar graph comparing your similarly normalized data to the normalized data of groups you can select. The functional widgets each show you how subsets of your data compare with that of different groups.
 
These predictions are made by looking at the collection of all genes present in the genomes of the organisms that are found in your microbiome, as well as closely related organisms. A system of gene and pathway classifications (KEGG) is used to predict what pathways these genes fall into.
 
To learn more about how people generally make these types of predictions see: Langille et al. 2013. Predictive functional profiling of microbial communities using 16s rRna marker gene sequences. Nature Biotechnology, 31.9, 814–821. http://doi.org/10.1038/nbt.2676
 

Note - the big difference between "steroid synthesis" and "caffeine metabolism" between me and other vegans tested (bottom bar graph), but the lack of such a big distinction when comparing me with everyone (top graph). These two spike were missing in a comparison with vegetarians as well, so I'm not sure they mean anything. It could be that the vegan sample size was very small. But then again, at the time of this sample (a couple years ago), I wasn't drinking any coffee - which could explain why my caffeine metabolizing bacteria count was lower than other vegans. Now that I am drinking coffee again it will be interesting to see if I have more bacteria in my new sample that are associated with caffeine metabolism. I think comparing my new data with my old data might make for more interesting insights.

 

But overall, I still don't find this all that helpful. Perhaps I'll get additional insights from a comparison between my two samples. I'll report back when I get the results.

 

The second new feature, the diversity metric, is not particularly useful either, but it appeals to my competitive nature  B)xyz. Here is my result:

 

post-7043-0-94771100-1446147783_thumb.jpg

 

So as of that original sample (circa 2013), my gut microbiome was more diverse than 98% of all vegans tested (and 97% of all people tested), according to the Simpson's Diversity Metric. So I got that goin' for me, which is nice.  :)xyz

 

I'm (competitively) curious how other CR practitioners score on this diversity metric.

 

Zeta - you in particular might get more out of a uBiome analysis than most people, being very technically competent and self-sufficient at analysis, and motivated to do so as a result of your health challenges. Put another way, I can't make heads or tails of data about my relative enrichment of bacteria responsible for various metabolic pathways like the following, but you might be able to find something interesting and relevant to your condition if you got this kind of data for yourself:

 

post-7043-0-98325800-1446148188_thumb.jpg

 

People like you Zeta are trying to do their own analysis to gain insights into health by taking multiple samples, with mixed and speculative results.  Here is one example. As you point out  - there is some concern over the value of this kind of analysis due to concern over its stability / repeatability, and what 'ground truth' really means, when big differences are seen between different services analyzing samples from the same bowel movement (i.e. uBiome vs. American Gut).

 

We might be able to gain interesting insights by comparing our data (CR practitioners that is) to that of other groups, and to each other based on differences in our diets.

 

--Dean

Share this post


Link to post
Share on other sites

What the heck. Your recommendation means a lot to me, Dean (even with the qualifications). It's only 40-something dollars, so why not.

 

And you're right: we might be able to do some interesting comparisons.

 

Note, everyone: sale ends tomorrow (Oct. 31)!

 

Edit:

 

Ah, and I just got a little discount code sent to me: I save, you save. Order via this link:

 

http://ubiome.refr.cc/HNCSFXK

Edited by Zeta

Share this post


Link to post
Share on other sites

Interesting new article in the NY Times discussing two new papers in Science [1] and Nature [2] calling for a coordinated national effort to study the microbiome, particularly in the human gut, similar to the large, federally-funded Brain Inititiative to study the brain. More and more people are recognizing the importance of the gut microbiome for determining health and longevity.

 

Coincidently, Dr. Greger's video today was on the same topic, although it had a similarly (pretty content-free) message - more study of the gut microbiome, and particularly the impact on diet on it, is warranted. He did have references to several good overview articles on diet, gut bacteria and health, with [3] looking particularly interesting.

 

--Dean

 

------------

[1] Science 30 October 2015: 

Vol. 350 no. 6260 pp. 507-508 
DOI: 10.1126/science.aac8480
 
A unified initiative to harness Earth's microbiomes
 
A. P. Alivisatos*, M. J. Blaser, E. L. Brodie, M. Chun, J. L. Dangl, T. J. Donohue, P. C. Dorrestein, J. A. Gilbert, J. L. Green, J. K. Jansson, R. Knight, M. E. Maxon, M. J. McFall-Ngai, J. F. Miller†, K. S. Pollard, E. G. Ruby, S. A. Taha, Unified Microbiome Initiative Consortium
 
 
Despite their centrality to life on Earth, we know little about how microbes (1) interact with each other, their hosts, or their environment. Although DNA sequencing technologies have enabled a new view of the ubiquity and diversity of microorganisms, this has mainly yielded snapshots that shed limited light on microbial functions or community dynamics. Given that nearly every habitat and organism hosts a diverse constellation of microorganisms—its “microbiome”—such knowledge could transform our understanding of the world and launch innovations in agriculture, energy, health, the environment, and more (see the photo). We propose an interdisciplinary Unified Microbiome Initiative (UMI) to discover and advance tools to understand and harness the capabilities of Earth's microbial ecosystems. The impacts of oceans and soil microbes on atmospheric CO2 are critical for understanding climate change (2). By manipulating interactions at the root-soil-microbe interface, we may reduce agricultural pesticide, fertilizer, and water use enrich marginal land and rehabilitate degraded soils. Microbes can degrade plant cell walls (for biofuels), and synthesize myriad small molecules for new bioproducts, including antibiotics (3). Restoring normal human microbial ecosystems can save lives [e.g., fecal microbiome transplantation for Clostridium difficile infections (4)]. Rational management of microbial communities in and around us has implications for asthma, diabetes, obesity, infectious diseases, psychiatric illnesses, and other afflictions (5, 6). The human microbiome is a target and a source for new drugs (7) and an essential tool for precision medicine (8).
 
-------
[2] Nature 526, 631–634 (29 October 2015) doi:10.1038/526631a
 
Microbiology: Create a global microbiome effort
 
Nicole Dubilier, Margaret McFall-Ngai& Liping Zhao
 
 
Understanding how microbes affect health and the biosphere requires an international initiative, argue Nicole Dubilier, Margaret McFall-Ngai and Liping Zhao.

 

-----------

[3] J R Goldsmith, R B Sartor. The role of diet on intestinal microbiota metabolism: downstream impacts on host immune function and health, and therapeutic implications. J Gastroenterol. 2014 May;49(5):785-98.

Share this post


Link to post
Share on other sites

uBiome really has to work on their interface. I'm irritated.

 

I click on Dashboard. Logical. I see a bunch of results. Logical.

 

I see "Carbohydrate metabolism". Puzzling, but it's a link, so I assume clicking on it will explain what that has to do with bacteria in my gut. It doesn't. A box comes up saying "Enrichment of Carbohydrate metabolism vs. All Samples". What the hell does "enrichment" mean here? Enrichment of bacteria that have something to do with carb. metabolism? Probably. But have to do with carb. metabolism in what way? Enhance it? Slow it down? Oh, I see, there's a subtitle to that box: "KEGG Pathways Level 3". What is that? No link, no explanation. Right-click google. "Kyoto Encyclopedia of Genes and Genomes". OK??

 

Caveat! This is only a first impression. I'll keep digging.

 

Meanwhile, Dean, I'll send you my raw data, just for fun. We could create our own comparisons easily.

 

Summary of my results, though: pretty avg., although I have very high ether lipid metabolism, and very low alpha-linolenic acid metabolism.

 

Oh, and I have low levels of proteobacteria -- more evidence that I really don't have IBD!

 

Zeta

Share this post


Link to post
Share on other sites

Zeta,

 

I got your results, but honestly, I don't know what to do with them. I'm happy to send you mine raw data if you'd like to do some spreadsheet wrangling. But it might be informative to compare ourselves using the graphics and data analysis ubiome provides first. I've now had 3 uBiome tests, in 2013 when I was eating fewer calories (and less nuts) and not exercising nearly as much, and then two quite recent uBiome tests about 2 weeks apart in late 2015 while eating my current diet, with many more calories (largely as nuts) than in 2013, and a lot more exercise.

 

First, my gut bacteria diversity relative to other people is reported to have changed pretty dramatically between the three tests:

  • 11/21/2013: 96th percentile gut bacterial diversity
  • 10/22/2015: 51st percentile gut bacterial diversity
  • 11/04/2015: 83rd percentile gut bacterial diversity

Note - second and third of these readings were only 2 weeks apart, eating the exact same diet and living the same lifestyle. So either one's gut population changes rather dramatically over a relatively short period of time for no apparent reason (could be, but undermines the value of this kind of testing if it does...), or more likely, the sampling or testing methodology is not very stable...

 

Next, a few screen captures.

 

First, here is the data from the three tests, showing changes in major metabolic functions of my gut bacteria over time. The numbers like "~1.02X" for "Lipid Metabolism" I interpret to mean the fraction of gut bacteria I harbor that are supposedly involved in metabolizing lipids as compared with the general population (i.e. I have 102% of the average person) :

 

YSSoEf5.png

 

Note relative to most people, and relatively to my first, 2013 results, I'm eating ton of fruit in these second two tests, yet my fructose metabolizing gut bacteria (purple bar at bottom) has remained virtually unchanged over time, and is almost exactly the same as other folks they've tested (~0.98x), according to uBiome. That seems suspicious...

 

Also observe how my alpha-Linolenic acid metabolism varies dramatically between the recent two tests, which were only 2 weeks apart. I went from being quite low (~0.3X) for ALA metabolism on 10/22/15 to being reasonably normal (~0.81X) two weeks later on 11/4/15. Again, that seems strange, questionable and unlikely. Given the amount of flax, chia and walnuts I consume, I'd expect there to be a lot of ALA-loving bacteria in my gut...

 

Here is a bar graph comparing my levels of gut bacteria in various functional categories to "All Samples". My data is represented by the blue bars (obviously). Zeta, perhaps we could use this to compare between the two of us:

 

HW5R7Cc.png

 

My two real outliers are caffeine metabolism and steroid biosynthesis, both very low relative to the general population. Caffeine metabolism being low seems strange, since I'm pretty sure I consume quite a bit more caffeine than the average person. Steroid biosynthesis might reasonably be expected to be low I guess, given I'm practicing a CR(-like) diet.

 

Here is the graph of how my highest level distribution of gut bacteria taxonomy has changed over the three tests:

 

hRC6q6R.png

 

The scale in the above is some funky log transformation so that rare bacteria families still show up on the graph and can be tracked over time. I appear to have 2x the number of firmicute bacteria as bacteroidetes, whatever that suggests...

 

Here are my taxonomy distribution results compared with the general population:

 

F7niOqu.png

 

I appear to be slightly enriched in firmicutes and below the average in each of the other four families listed. That general relationship has held throughout the three tests. Zeta, as you can see I too am low in proteobacteria relative to the general population (2.62% for me vs. 3.31% for everyone else). I definitely don't have IBS :-)

 

I'm not sure how to interpret these taxonomy results, but it might be interesting to compare our results Zeta, to see how we compare with each other, and with the general population. 

 

--Dean

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×