This is what I've dreamed of for so long! Testing for microbes in our tanks!

[Paul Sands]

Are users that send in samples also answering some sort of survey about their tanks (new, old, dinos, stn, etc)? It seems that it would be necessary in order to begin to correlate problems with their causes and eventually determine solutions.

 

[lexinverts]

Are users that send in samples also answering some sort of survey about their tanks (new, old, dinos, stn, etc)? It seems that it would be necessary in order to begin to correlate problems with their causes and eventually determine solutions.

Yes, you provide data on your tank via survey when you register your sample on the website.

 

[Dan_P]

I have no idea what levels they can detect and what exactly that translates to. I can’t see how one could quantify the number of benthic bacteria from bits of dna that might be in the water.

Good point. Thanks

 

[rmurken]

seems similar to an ICP test.
They send you a sample kit - follow directions to sample your water & send it back - they send you results of microbe species and relative concentrations in your tank

They probably don’t vaporize the bacteria into plasma but other than that, totally like an ICP test!

 

[rmurken]

One thing that will be interesting is to see the relative abundance of vibrio and mycobacterial DNA in the samples; two groups that can potentially infect the aquarist.

I have had necrotizing fasciitis (“flesh eating bacteria”). In my case was Strep B, not vibrio, AND I DID NOT CATCH IT FROM MY FISH TANK. But I am at least reasonably cautious about washing my hands if I have any cuts or scratches precisely because of the vibrio issue.

 

[Graffiti Spot]

Can we get a sample of the questions we have to answer about our tanks when we send in our water samples? I am curious about this as well as where this testing can take the hobby.

To the owner of the company, I am very interested in the differences your seeing between long term heavy carbon source users tanks and the tanks that don’t use carbon sources.
Also interested in knowing if anyone dosing a certain bacteria product has the same bacteria in their tank long after dosing. Assuming that bacteria was not there from the start.

 

[taricha]

As the owner of the company I'd be happy to answer any questions.

Thanks. Great answers. This is a service that the hobby is not yet ready for (and I mean that as a great compliment) because it's quite outside what had always been knowable. Give us a few months to catch up. We'll have better questions then

The one system that I was having trouble with dino, was, by far, the system that had the lowest microbial diversity.

Nice tease! Would love to see that if it can be shared.

Now THAT is interesting, and supported by anecdotes about sterile tanks/low nutrients leading to dinoflagellate outbreaks.

To dig in on this, the picture that gets painted in the published lit is that recognizable communities: Dino blooms, Cyano mats, Macroalgae support and to some extent are supported by bacteria that are associated with those communities. The tightness and necessity of the association varies. A ball of chaeto can still grow (slower) if you wipe out/disrupt its bacterial community. Cyanobacteria is almost impossible (even for professionals) to grow without its bacterial associates. Dinos will be somewhere in between, and probably more like cyano - a tighter association.

...For all we know, said tank with dinos could have a full microbial recovery if fed more— and just pouring in the absent microbes would do nothing, as it is not the “cause” of the issue. Just correlational.

Good Q. The analogy would be - it's much easier to increase your copepod population by adding phytoplankton than by just pouring in copepods. The same may hold for bacteria

One aspect of this that would be very interesting is comparing the microbiome from a "normal" (whatever that means!) reef tank to the microbiome from a carbon-dosed tank.

To the owner of the company, I am very interested in the differences your seeing between long term heavy carbon source users tanks and the tanks that don’t use carbon sources.

I said it'll be hard for us to ask good questions yet, but this is great one to start.
Maybe we define a system as Carbon Limited: gets fish food - no carbon dosing - and keeps some (even tiny) amount of NO3 and PO4 in the water.
vs.
Carbon Replete: in addition to fish food, is dosed organic carbon daily, and may test at/near zero on NO3 or PO4.
The info from Aquabiomics in OPs link suggests some families specialize in more complex foods, and others that use simpler foods would expand in response to easily available carbon in water. Would love to see if that holds true and is measureable.

One thing we've learned already - there was a thought that maybe carbon dosing just created a mass monoculture of bacteria. That seems blown out of the water by what's been posted already.

 

[AquaBiomics]

Lots of great questions, forgive me for those I miss!

Yeah that thought crossed my mind also. I figured that he tested something like 1ul samples from the water with maybe ten different syringes spread into the sample to pull it all out at the same time. Then he checks to see how many of those 1ul tests have what kinds of DNA. A total guess on my part but unless multiple tests are run on very small simultaneous samples I cannot see how you can get any kind of count.

I fear that I've spent so long analyzing modern DNA sequence data (which comes in millions of sequences at a time) that I forget only a genomics geek is familiar with how these are analyzed.

For context, I'll say first that the way I am analyzing the microbial community is standard practice. For example in the Earth Microbiome Project, the Human Microbiome Project, and TARA Oceans. I didn't invent this; I am borrowing methods developed by great minds in the genomics and microbiology communities and applying them to the reef tank. Nothing is proprietary or secret here. I can share my specific laboratory protocols if someone wants them -- there is no room for secrecy in science, and no need for it in a service like this.

This tool (16S rDNA Sequencing) is central in modern microbial ecology. Basically it works by sequencing DNA from a mixed community of microbes and counting the number of sequences that come from each type, to estimate the proportion of each type in the community sampled.

Here we sample a large community of microbes (60 ml of water, containing 100s of thousands to millions of microbes) and extract their DNA. Then, I target a specific part of each microbe's DNA (called 16S) and prepare those parts for DNA sequencing. I sample about 10,000 of these from the pool of bacterial DNA and sequence them. Assuming large enough samples are drawn at each step, the frequency of each type of bacterial DNA in the sequence data is a reasonable estimate for its frequency in the total population.

There are caveats and schools of thoughts and a dozen variations on these methods that biologists love to debate. Which would require a lot of background and would be boring to everyone but genomics geeks. But what I described above is the overall strategy. Does that help clarify it at all?

 

[Thales]

I can share my specific laboratory protocols if someone wants them -- there is no room for secrecy in science, and no need for it in a service like this.

That would be very helpful. If it is easier to share in email let me know. We do this kind of work in the lab I am part of and it would be great to know what you are doing so we have a good way of comparing results with stuff in our lab and stuff at my house. And, like you say, there is no room for secrecy (the way the ICP testing outfits are not sharing that kind of info makes it very hard for me to have faith in their results).

Thanks!

Rich

 

[AquaBiomics]

Can we get a sample of the questions we have to answer about our tanks when we send in our water samples? I am curious about this as well as where this testing can take the hobby.

To the owner of the company, I am very interested in the differences your seeing between long term heavy carbon source users tanks and the tanks that don’t use carbon sources.
Also interested in knowing if anyone dosing a certain bacteria product has the same bacteria in their tank long after dosing. Assuming that bacteria was not there from the start.

There is a link on the website ("Register Sample") but to prevent spam submissions it's only visible to logged in users. But I'm happy to share it here.

I would greatly appreciate any feedback on other details that we should be including. There is some tradeoff between making it so long no one will want to complete it, vs so short it's not useful. But I generally favor gathering more data about each sample rather than fewer

 

[AquaBiomics]

That would be very helpful. If it is easier to share in email let me know. We do this kind of work in the lab I am part of and it would be great to know what you are doing so we have a good way of comparing results with stuff in our lab and stuff at my house. And, like you say, there is no room for secrecy (the way the ICP testing outfits are not sharing that kind of info makes it very hard for me to have faith in their results).

Thanks!

Rich

Hi Rich,

I'll put the details together and message you. Any delay is just because I don't have it written up in a clean format.

Overall here is the gist of it. (Here is the punchline in advance -- this is the academics version of a short version!)

We are sampling 60 ml of aquarium water on a 0.2 micron MCE membrane, then fixing with 70% ethanol. I extract DNA using a standard extraction kit for environmental samples, then amplify the V4 region (515-806) of the 16S gene. This region was chosen to maximize overlap with other marine and especially coral datasets. I prepare sequencing libraries using a mixture of GoTaq and Pfu, and sequence on Illumina MiSeq. I sequence about 10,000 paired end reads for each sample, 250 bp in length (i.e. sequencing almost the complete amplicon from both ends).

I analyze the data using a collection of custom scripts I've written, available on github, along with other open source software. After removing low-quality or uninformative reads, I process the high-quality reads with the package "dada2" in the R statistical software. This reconstructs the paired reads, and identifies and counts the different types (but still just as anonymous "sequence variants"). I then compare these using software called BLAST against the GreenGenes database, a collection of microbial DNA sequences curated by the experts. In this way, I can identify which type of microbe each sequence came from.

Putting these together, we arrive at the identity and abundance of each type in the microbial DNA sample.

I hope this is useful or interesting despite the lack of detail,

-Eli

 

[AquaBiomics]

One more reply and I'll stop spamming for a while.

After such a warm welcome, I feel compelled to clarify so no one feels deceived. I have actually been among you as a hobbyist for some time, under the account EMeyer. I lurked here for years and learned a ton about the hobby, and more recently have been buying and selling a few frags on the forum to feed my coral addiction... err I mean, my hobby. I made this account AquaBiomics in an effort to keep my hobbyist transactions (e.g. buying coral frags) separate from "official" company related stuff. But I'm not role playing different characters or something, no secrecy is intended

 

[Reefahholic]

Very interesting! Would love to see results of well known tanks vs average reefers.

 

[lexinverts]

Very interesting! Would love to see results of well known tanks vs average reefers.

I’ve heard that he has samples from some “reefing celebrities” but I can’t disclose who.

 

[Flippers4pups]

One more reply and I'll stop spamming for a while.

After such a warm welcome, I feel compelled to clarify so no one feels deceived. I have actually been among you as a hobbyist for some time, under the account EMeyer. I lurked here for years and learned a ton about the hobby, and more recently have been buying and selling a few frags on the forum to feed my coral addiction... err I mean, my hobby. I made this account AquaBiomics in an effort to keep my hobbyist transactions (e.g. buying coral frags) separate from "official" company related stuff. But I'm not role playing different characters or something, no secrecy is intended

Well put and up front! Me like!

 

[Silver14SS]

One more reply and I'll stop spamming for a while.

After such a warm welcome, I feel compelled to clarify so no one feels deceived. I have actually been among you as a hobbyist for some time, under the account EMeyer. I lurked here for years and learned a ton about the hobby, and more recently have been buying and selling a few frags on the forum to feed my coral addiction... err I mean, my hobby. I made this account AquaBiomics in an effort to keep my hobbyist transactions (e.g. buying coral frags) separate from "official" company related stuff. But I'm not role playing different characters or something, no secrecy is intended

Neat service, I commend you for being so transparent!

 

[EmdeReef]

Hi everyone,

As the owner of the company I'd be happy to answer any questions. There have been a lot asked already in the thread so forgive me if I miss some. I'd like to start with a few of the more important points.

1. I promise: no snake oil. I'm selling a testing service; whoever compared it to ICP is exactly right.
You will notice my site is conspicuously absent of promises that "this bacterium is bad" or "this bacterium is good". We simply don't know enough yet to make those claims.

2. By building a large database of samples from hobbyist and aquaculture tanks, we will learn a lot. I already have some data in hand and am gathering more.

3. There is some skepticism about what this means. Skepticism is a wise default position, but please do note what my service claims or does not claim. I claim to be able to identify the microbes in your tank, and provide information about what those microbes do. I will stand by those claims, since they are the basis for most modern microbial ecology.

I do NOT claim that I can sell you a magic bottle that will make everything happy. Or even that we understand everything about the microbes we find. In the reports, I try hard to not overstate the conclusions.

I've spent the last 20 years in Academia, so I'm comfortable with disagreement and discussion. Don't be shy about disagreeing - my service is simply DNA sequencing, I'm not selling bacteria. So I won't be offended if someone has an argument why bacterial type X is unimportant. (If there is someone who thinks it doesn't matter what kind of microbes are in your tank, I would disagree, but I'm not sure anyone in the reefing world holds that opinion anyway! And it would be an interesting discussion in any case)

4. Finally (for this post), some have raised the question of planktonic versus surface associated bacteria. Its an important distinction. But please recognize that water is constantly circulating past the surfaces in our tanks -- the biofilm microbes show up in the water column too. Lots of evidence in hand now to show this.

With that said, our first round did show that for a more sensitive analysis of biofilm microbes, we should include a direct sample of the biofilm. This contains important ammonia-oxidizing and nitrate-oxidizing microbes. So current sampling kits include a biofilm sample too.

I'm gonna stop there for now, but I'll come back and answer questions. There is a link on the main page of the company website that describes a basic overview of the microbes that live in reef tanks. Soon I'll be adding my analysis of differences between tanks, and my cycling experiment studying the succession of microbes in new aquariums started various ways.

We have a lot to learn - but the way we learn is by collecting data. Thats what my service is for.

-Eli

I think this is great and has lots of potential! I did my thesis research focusing on (non-viral) reef pathogens, tough space - wishing you best of luck and will be trying your product out of curiosity!

Unrelated question, have you thought about developing a test for some of the more common problem pathogens like C.irritans or A.occelatum?

I know the testing approach as well as sample taking would have to be quite different but it would be invaluable for reefers running tanks fallow if testing could indeed shorten the dreaded ~80 days. I know a researcher at SUNY who was looking into this a while ago.

 

[AquaBiomics]

I think this is great and has lots of potential! I did my thesis research focusing on (non-viral) reef pathogens, tough space - wishing you best of luck and will be trying your product out of curiosity!

Unrelated question, have you thought about developing a test for some of the more common problem pathogens like C.irritans or A.occelatum?

I know the testing approach as well as sample taking would have to be quite different but it would be invaluable for reefers running tanks fallow if testing could indeed shorten the dreaded ~80 days. I know a researcher at SUNY who was looking into this a while ago.

Absolutely. Its straightforward to modify this approach to target different organisms; there are good eukaryotic markers (18S) that can be used to identify any of these non-bacterial pathogens.

I'm running some tests right now and including some of these samples in the upcoming run. The only details that are uncertain are sampling issues and sensitivity; in principle if you can get one of these pathogens (or the environmental DNA from them) into a tube, we can detect them.

 

[EmdeReef]

Absolutely. Its straightforward to modify this approach to target different organisms; there are good eukaryotic markers (18S) that can be used to identify any of these non-bacterial pathogens.

I'm running some tests right now and including some of these samples in the upcoming run. The only details that are uncertain are sampling issues and sensitivity; in principle if you can get one of these pathogens (or the environmental DNA from them) into a tube, we can detect them.

That’s great to hear! I believe the commercial farming test takes 10 samples but that may not even be necessary, I’ll try to find a paper that discusses methods.

 

[EMeyer]

To make sure I'm not overpromising, I should be clear that the sampling part is not trivial. It will be important to test sensitivity very thoroughly. Multiple samples like you suggest seems reasonable. Proving absence is challenging.

So while the PCR is straightforward the tests for those pathogens remains in R&D.