I really wish this wasn't a thing. I'm getting tired of all the "startups" in my field (microbio) who are pushing microbiome-based treatments. They're not currently regulated by the FDA and they are downright dangerous. The healthy gut is primarily made up of (~90%) Firmicutes, Bacteroidetes, and Clostridium of which the majority are uncultured, unsequenced and completely unstudied. Shifting the balance a little too much one way or another can potentially cause symptoms of Autism Spectrum Disorder. There are at least two startups in stealthmode that I know of that are pushing microbiome matching between "healthy" donors and "unhealthy" recipients -- very much like snake oil salesmen. One person's healthy is another's unhealthy but these groups and their VCs apparently don't give a shit (VC funding in science is a while other can of worms).
If you want to be a microbiome hacker, go contribute YOUR microbiome to publicly funded by you research. I suggest the American Gut Project[1][2] for people in the US and British Gut[3] for those in the UK and EU. For $99 you can learn about your microbiome and contribute to largest, most successful crowdfunded research in the world. This project is entirely supported by people like you and me, directly.
So I can't speak to the specifics of the startups you're referring to (I don't know who they are), but I will say I'm very surprised to have read this.
The responsible development of new therapies is almost always a good thing. I'm confused to hear you say they're not FDA regulated, since the FDA is very clear: if it's intended to treat a disease, it's regulated. Perhaps you could explain what I'm missing?
Also technical note: these microbes are not unknown/uncultured/unsequenced. See MetaHIT, the Human Microbiome Project and the American Gut Project you mentioned.
They're not marketed to cure a disease, they are "probiotics". They are using the yogurt rule
>Also technical note: these microbes are not unknown/uncultured/unsequenced. See MetaHIT, the Human Microbiome Project and the American Gut Project you mentioned.
Yes the are, we have metagenomics on them (v4 rDNA sequence reads) not full seq data. We can't ID which bug and then grow it. We just know some of its rDNA is in the mix. Being able to phylotype =/= known microbe.
Ahh, you mentioned Autism spectrum, so I assumed that couldn't be the case.
> we have metagenomics on them (v4 rDNA sequence reads) not full seq data
There is plenty of genomic data beyond 16S. For the HMP, full sequencing of selected strains as well as hundreds of WGS metagenomes were done. And that's almost old news now, I'm sure there is even more being done currently. The result is basically either a full genome, or a set of giant chunks of the genome of the abundant microbes there. We don’t have all the genomes closed (although some are in the process!) but closing a genome is a very high bar and probably not worth it when you can get a lot of data from a fairly good set of large contigs.
Can the assemblies be chimeric/wrong/whatever? sure, some of them will be, but at this point most assemblies are fairly good :)
How do you think ubiome.com compares with these efforts? I did an analysis with them and feel like I got some insights, but I also felt that the way the data Is currently presented could lead to a lot of false conclusions, especially for people without a microbiology or statistical background.
I'm also curious about your statement about how many strains are unsewurnced and unstudied. I've read this elsewhere, and it makes me wonder if the species % data presented by ubiome is presented with false precision.
µBiome is vapid consumerism in my opinion. They don't contribute data to the research community (as far as I know).
The vast, vast majority of all bacteria are not culturable in a lab. That being said, we use rRNA/rDNA sequencing[1] to build phylogenies based off of BLAST E-values[2], or how likely it is two sequences evolved separately from each other. We have cultured and sequenced enough of the big three (and many of the top 100 gut microbes) that we can be statistically sure that they're related. We can't say for sure two different but closely related sequences are two different, closely related species but we can say they're related OTUs[3]
I quite liked that it was possible to download my raw data file from ubiome. I got my first results back a couple months ago. Then I got sick and in a while I'll know precisely how a course of antibiotics devastated my microbiome (but killed what it needed to, so I'll probably live.)
I really wish this wasn't a thing. I'm getting tired of all the "startups" in my field (microbio) who are pushing microbiome-based treatments. They're not currently regulated by the FDA and they are downright dangerous. The healthy gut is primarily made up of (~90%) Firmicutes, Bacteroidetes, and Clostridium of which the majority are uncultured, unsequenced and completely unstudied. Shifting the balance a little too much one way or another can potentially cause symptoms of Autism Spectrum Disorder. There are at least two startups in stealthmode that I know of that are pushing microbiome matching between "healthy" donors and "unhealthy" recipients -- very much like snake oil salesmen. One person's healthy is another's unhealthy but these groups and their VCs apparently don't give a shit (VC funding in science is a while other can of worms).
If you want to be a microbiome hacker, go contribute YOUR microbiome to publicly funded by you research. I suggest the American Gut Project[1][2] for people in the US and British Gut[3] for those in the UK and EU. For $99 you can learn about your microbiome and contribute to largest, most successful crowdfunded research in the world. This project is entirely supported by people like you and me, directly.
[1] http://humanfoodproject.com/americangut/
[2] https://fundrazr.com/campaigns/4Tqx5
[3] https://fundrazr.com/campaigns/4sSf3/ab/5qJ7f