"Biodiversity is dead, long live biodiversity" 10 month microbiome data from BRStv.

[cpschult]

Aquabionetics, also told my reef at the time had less biodiversity than most tanks. And my tank was over 45 years old at the time.

This actually makes sense. Even with the instability you bring in by adding mud/from nature, most tank niches are well established. If you made a major change I’d expect those niches to open.

 

[areefer01]

Short summary of the overall exercise. (meet the tanks - Ep 3)
They started 12 tanks from different initial ingredients:
1) Dry sand and Rock - control
2) Dry sand and Rock + Coral frags/colonies
3) Dry rock, live sand in a bag
4) Synthetic live rock cured in seawater
5) Dry Rock and sand + live rock rubble cured in the dark, in dark sump
6) Live Rock and Sand directly from established tank
7) Dry rockand sand + 2 cups of established tank sand
8) Paper-wet indonesian live rock and dry sand
9) Gulf live rock shipped in water and dry sand
10) Dry rock and sand + Aquaforest reef mud
11) Dry rock and sand + 100% water from established system
12) Dry rock and sand + biobrick from established system

No test that uses both live rock and sand or is there a typo in the summary? TBS for example offers both.

They tested each tank via aquabiomics at week 2, week 4, 10, and 15.

Eli can probably answer this but is testing at week 2, 4, 10, and 15 enough for change to be noticed? More importantly how long does it take for testing DNA in marine aquaria (controlled) to show.

Idea 1: Biodiversity is dead, Balance is the new "biodiversity"
Aquabiomics gives two overall statistical measures/scores:
Biodiversity is a statistical measure of the bacterial families that make up more than 1% of the measured genetic material.
Balance is a measure of how much the present bacterial families look similar to those in established reef systems.

I've used the product but for general reference to this discussion what is the benchmark or standard used for reference? Also does it matter how a hobbyist score compares to X?

Note: I am of the opinion that the data model for Eli's work is still being done. Given time and access I'd wager some interesting trends could be seen. Who knows.


Edit: removed part of the post that I wasn't replying to - sorry.

 

[ISpeakForTheSeas]

Eli can probably answer this but is testing at week 2, 4, 10, and 15 enough for change to be noticed? More importantly how long does it take for testing DNA in marine aquaria (controlled) to show.

I don't know how long DNA testing takes to show, but - going based off of biofilm communities - I would guess that yes, testing at those weeks should show definitive changes in the bacteria communities in the tank:

Here's the specific article I was referencing:
https://www.researchgate.net/publication/356692776_Coral_reef_biofilm_bacterial_diversity_and_successional_trajectories_are_structured_by_reef_benthic_organisms_and_shift_under_chronic_nutrient_enrichment And a related one:

Prokaryotic community successions and interactions in marine biofilms: the key role of Flavobacteriia

Marine biofilm community structure on a model surface showed a rapid colonization followed by temporal dynamics associated to inter-specific cooperation.

academic.oup.com

 

[Dan_P]

I've meant to start this discussion thread for a while... so here we go.

Reef hobbyists have a lot of ideas about biodiversity and what it means in a reef system. A lot of these ideas (mine included) are reasoning from personal experience of myself and others, and how we think it ought to work. Mostly this is because the concepts are fuzzy and we just don't have much data - so reasoning from collective experience is the best we've got in these data-sparse areas.
But that's changing - there are a lot of concepts in this discussion that we now actually have pretty good data to ground ourselves in and think more concretely about - surrounding the concepts of biodiversity, maturity, balance etc. BRS did a series of 11 videos (full playlist) in 2022 "Biome Cycling" involving 12 side by side aquaria run in parallel for 10 months, with detailed observations and data. This was done at a scale and expense that the hobbyist can't duplicate and I think the results are of good quality, and the data is worth digesting even if I don't buy all his reasoning and conclusions. Given how much we think and talk about these ideas like the "ugly phase", biodiversity, maturity, balance etc - this data deserves way more discussion than it has received so far. It's a ton of data, observations, and reasoning to sift through. Way more than one post,
So I'll try to occasionally pull one big idea at a time that I think is well demonstrated by the data, and is worth talking about.

Short summary of the overall exercise. (meet the tanks - Ep 3)
They started 12 tanks from different initial ingredients:
1) Dry sand and Rock - control
2) Dry sand and Rock + Coral frags/colonies
3) Dry rock, live sand in a bag
4) Synthetic live rock cured in seawater
5) Dry Rock and sand + live rock rubble cured in the dark, in dark sump
6) Live Rock and Sand directly from established tank
7) Dry rockand sand + 2 cups of established tank sand
8) Paper-wet indonesian live rock and dry sand
9) Gulf live rock shipped in water and dry sand
10) Dry rock and sand + Aquaforest reef mud
11) Dry rock and sand + 100% water from established system
12) Dry rock and sand + biobrick from established system

All tanks were given two clownfish and were kept dark for 4 weeks, then moderate "LPS lighting" from weeks 5-10, then high "SPS lighting" from weeks 11-15.
They tested each tank via aquabiomics at week 2, week 4, 10, and 15.
(What aquabiomics is and what was measured - Ep 4)


Idea 1: Biodiversity is dead, Balance is the new "biodiversity"
Aquabiomics gives two overall statistical measures/scores:
Biodiversity is a statistical measure of the bacterial families that make up more than 1% of the measured genetic material.
Balance is a measure of how much the present bacterial families look similar to those in established reef systems.

Here's the backwards thing that kills many ideas about biodiversity (like mine). In essentially every system that was started with any live material, The Biodiversity fell while the Balance rose. The biodiversity is initially higher especially in tanks started with a lot of live material like the live rock, but gradually falls. The Balance score - on the other hand, starts extremely low, but in every decent-looking tank climbs over time to more closely resemble typical reef tanks.
So early on, the high biodiversity likely represents a disturbed system, with many food sources of dead, disturbed and out of place organisms, and quick bacterial growth in response - but those early bacterial families look nothing like the eventual expected families that will make up a well-established tank.
Put another way, it is the death and loss of early bacterial diversity that helps shape the microbiome to look more like eventual reef systems. The Balance is a far better measure of this process of moving towards an established system than the biodiversity is, and perhaps the balance score isn't a terrible marker for biofilm maturity and system stability overall.
Ryan explains this idea of high biodiversity in tanks that look awful, and the balance being a better indicator here Ep 7 3:01-3:48

This data is in the videos Ep 5, 6, 7.









And here's Idea 2a and 2b in post 51

This took a bit of work. Thanks for this. You earn a “Hero of the Week” badge.

Aquabiomics still suffers from bacteria myopia. While bacteria are important microorganisms ( Duh, right), a reef tank is also occupied my photosynthetic microorganisms. The interplay between bacteria and algae can be very important in nature and likely for the “look“ of an aquarium. I would not be surprised to find that algae play a big role in determining what the bacteria population looks like and probably vice versa. The Aquabiomics notion of “balance” is incomplete and likely a correlation not a cause.

 

[taricha]

The Aquabiomics notion of “balance” is incomplete and likely a correlation not a cause.

Causation? Hah!
I'd just settle for good correlations that tell us things are happening together.

 

[taricha]

I would not be surprised to find that algae play a big role in determining what the bacteria population looks like

This idea also appeals to me, it implies we could know things about the bacterial biome we can't see by looking at the state of play with the algae we can see.
We'll see if the data gets us there or somewhere in that direction.

 

[C4ctus99]

So their measurement of “balance” was only matching bacteria to bacteria in established tanks? How many established tanks? What kind/how many different strains do they consider to fall under their “balance” category?

Guess I just feel like “balance” is a little subjective here…?

 

[ISpeakForTheSeas]

So their measurement of “balance” was only matching bacteria to bacteria in established tanks? How many established tanks? What kind/how many different strains do they consider to fall under their “balance” category?

Guess I just feel like “balance” is a little subjective here…?

Don't know how many tanks (I've wondered that myself), but the quote below has a recent test that shows which families of bacteria (yes, families, not species/strains - which would be way more intensive but also more meaningful for really showing the "Balance" as described) fall into their balance category (the second page of the report in the quote):

I received my microbiome report on my 9mo tank today. It promises to give me information on improving the balance score but I'm unable to locate those suggestions. Can someone help me interpret these scores please.

 

[taricha]

No test that uses both live rock and sand or is there a typo in the summary?

this one used live rock and sand from one of their favorite tanks.

6) Live Rock and Sand directly from established tank

Eli can probably answer this but is testing at week 2, 4, 10, and 15 enough for change to be noticed? More importantly how long does it take for testing DNA in marine aquaria (controlled) to show.

The data shows significant changes over time during this time frame.

First chart is the Diversity scores (the raw number of types detected.)

Notice how the diversity drops across the board during the early dark phase, while the reverse happens in the early dark phase to the tanks' balance scores below.

These are the Balance scores - a.k.a similarity to core microbiome of Reef Tanks ( again raw scores, not percentiles)

I've used the product but for general reference to this discussion what is the benchmark or standard used for reference? Also does it matter how a hobbyist score compares to X?

So their measurement of “balance” was only matching bacteria to bacteria in established tanks? How many established tanks? What kind/how many different strains do they consider to fall under their “balance” category?

I don't know how the aquabiomics database may have changed since then, but here's some discussion that gives a little peek at what's going on when a comparison is made for a balance score.

https://aquabiomics.com/articles/core-aquarium-microbiome
https://aquabiomics.com/articles/how-aquarium-microbiomes-differ

It is not known if or how it matters how your tank diversity or balance scores compares to the core microbiome of reef tanks.

 

[C4ctus99]

this one used live rock and sand from one of their favorite tanks.



The data shows significant changes over time during this time frame.






I don't know how the aquabiomics database may have changed since then, but here's some discussion that gives a little peek at what's going on when a comparison is made for a balance score.

https://aquabiomics.com/articles/core-aquarium-microbiome
https://aquabiomics.com/articles/how-aquarium-microbiomes-differ

It is not known if or how it matters how your tank diversity or balance scores compares to the core microbiome of reef tanks.

So just 20 sample tanks in each article to pull the consensus…

 

[Dan_P]

Causation? Hah!
I'd just settle for good correlations that tell us things are happening together.

Maybe the road to causation is paved with correlations

 

[Dan_P]

This idea also appeals to me, it implies we could know things about the bacterial biome we can't see by looking at the state of play with the algae we can see.
We'll see if the data gets us there or somewhere in that direction.

I forgot to include fungi, and my favorite, protozoa in the aquarium microbiome orchestra. But as you say, a correlation would be interesting.

 

[taricha]

Idea 3: Almost none of the 12 tank starting schemes acheived average (>50 percentile) Diversity anytime during the test period weeks 2 to 15. Only one system even got close, and it took 15 weeks to get to that result. None of the others got over 34 percentile at any point, with most systems spending the whole 15 weeks around and below 10th percentile.

But in Balance score, more than half the systems achieved an average ( >50 percentile) balance score by the end of the 15 week period. A third (4/12) systems got there by the 2nd measurement at week 4.

This happened in systems both with strong initial material influences (Coral, Indo live rock) and some with no intentional seed material at all (dry rock/sand). These very different starting materials were able to achieve a balance that looked like a typical reef tank in this time frame.
If we say this criteria is a goal - to have our major bacterial components look like the rest of the reef hobby - then this goal is achievable in a fairly short time regardless of starting material. Quite an optimistic takeaway.

But if we say diversity matching a hobby tank is the goal - then this is a long slog - even with carefully chosen material, and no guarantees that the system gets there at all. Not so rosy a scenario.

One thing both views (diversity and balance) show is that there are quantifiable answers to a question that we often pose. Cycling takes 1-2 weeks, but we feel like tank "maturity" takes much longer - on the scale of months. But we don't have good answers to what we think happens between 2 weeks and a few months. One of those answers is bacterial succession is clearly a much slower process than a couple of weeks.
This is a snapshot of all 4 microbiome tests over 15 weeks on all 12 tanks... No tanks are totally settled across any time frame within this 15 week period. Some are in more flux than others.

 

[taricha]

And here's Eli Meyer giving a very interesting 2 minute answer to this question on Diversity and Balance score and what bacterial diversity means to him in terms of it being desirable and how his views have shifted after looking at a couple of years of hobby data.
Aquabiomics presentation to PNWMAS (1:19:05-1:21:20)

 

[Thales]

I did an aquabiomics test on two tanks on my system in 2020. Eli sent me this very nice email about the results (Not the entire email is copied below). Both the display (DT) and SHL tank are on the same sump, but separated by 30 feet, and the DT had been up for 17 years and the SHL for about a year. I have a test on the shelf, maybe it is time to send it again since I already paid for it. I think data we have from these tests is interesting, but that we need to be careful about it because of how the samples are currently likely not a wides swath of the hobby, rather people trying to sherlock out a problem, and that we have no real idea what the averages actually represent. I know this has all been said before, but it seems important.
Anyway, here is what Eli had to say to me about balance and diversity in the samples I sent him
"
I'm writing to follow up personally on this automated email, and share my perspective on your reports. Since you sampled two tanks, I'll discuss them together.



First diversity. The two tanks had strikingly similar diversity scores. If I sampled the same tank twice I'd expect to get this level of variation. Both tanks communities were more diverse than most tanks I've sampled (74th and 77th percentiles). Despite the debates on the forums, I view high diversity as a positive thing based mainly on comparisons with natural reef water samples (which are more diverse in pristine reefs than degraded reefs). From this pro-diversity perspective, both tanks' communities look as good as one could hope for.



Its important to consider diversity in the context of tank age -- all tanks start at zero, rise to some level during setup, and (we think, although the trend is not significant yet) decline in diversity over time. We find a few outliers here and there. I would say yours are outliers... unusually diverse for their age. I plan a detailed analysis of this (tank age vs diversity) in the near future and will write it up on R2R.



Next balance. The composition and relative abundance of both tanks' microbiomes differed quite a bit from the typical community. This score is R2 calculated from a correlation of each family's relative abundance in your tank vs the typical tank (average). Low scores indicate communities that are very different from the typical reef tnak microbiome. This doesnt necessarily mean anything is wrong - just different, at a microbial level. This score is included as a way of quantifying exactly how different it is.



You can view the community barplots to understand the reasons for these scores. Your home DT differed from the typical profile by having very low levels of Flavobacteriaceae and Pelagibacteraceae, which are major components of the typical community. It also had elevated levels of Alteromonadaceae, which are a normal part of the community that are often elevated when Flavobacteriaceae and Pelagibacteraceae are low. It also differed by having unusually high levels of Colwelliaceae and Thiotrichaceae, which are not normally major components of the reef tank microbiome.



Your SHL also differed from the typical community, but for different reasons. In this case, lower than typical levels of Pelagibacteraceae, and slightly lower levels of Alteromonadaceae and Flavobacteriaceae, This tank showed very high levels of Vibrionaceae. In this case, this is almost entirely Vibrio fortis, with a little contribution from several Vibrio sp., Photobacterium rosenbergii, and Photobacterium sp. V. fortis appears to be a pathogen of at the very least, seahorses, although I havent yet moved it onto the pathogen list in the report. With over 1000 sequences from this single bacterium in your sample, this may be worth noting..



We also see Colwelliaceae again in your SHL. In both cases this is Thalassomonas sp. Some members of this genus are thought to be coral pathogens, but since the types in your tank are not identified to the species level I don't think there is any reason for concern. I find exactly these same types in some of my own home tanks, and find them at high levels when I swab healthy corals.



Nitrifying microbes - both of your tanks showed ammonia-oxidizing microbes at levels within the normal range. Interestingly, your SHL showed only Archaea (Cenarchaeaceae) while your home DT had both AOA and AOB. Both tanks showed high levels of NOB, which is a far from universal finding. In both tanks, these were all Nitrospiraceae, the most common group I find in aquariums. "

 

[Dan_P]

Idea 3: Almost none of the 12 tank starting schemes acheived average (>50 percentile) Diversity anytime during the test period weeks 2 to 15. Only one system even got close, and it took 15 weeks to get to that result. None of the others got over 34 percentile at any point, with most systems spending the whole 15 weeks around and below 10th percentile.

But in Balance score, more than half the systems achieved an average ( >50 percentile) balance score by the end of the 15 week period. A third (4/12) systems got there by the 2nd measurement at week 4.

This happened in systems both with strong initial material influences (Coral, Indo live rock) and some with no intentional seed material at all (dry rock/sand). These very different starting materials were able to achieve a balance that looked like a typical reef tank in this time frame.
If we say this criteria is a goal - to have our major bacterial components look like the rest of the reef hobby - then this goal is achievable in a fairly short time regardless of starting material. Quite an optimistic takeaway.

But if we say diversity matching a hobby tank is the goal - then this is a long slog - even with carefully chosen material, and no guarantees that the system gets there at all. Not so rosy a scenario.

One thing both views (diversity and balance) show is that there are quantifiable answers to a question that we often pose. Cycling takes a 1-2 weeks, but we feel like tank "maturity" takes much longer - on the scale of months. But we don't have good answers too what we think happens between 2 weeks and a few months. One of those answers is bacterial succession is clearly a much slower process than a couple of weeks.
This is a snapshot of all 4 microbiome tests over 15 weeks on all 12 tanks... No tanks are totally settled across any time frame within this 15 week period. Some are in more flux than others.

You are a data machine!

I have some alternative interpretations of this data. For now I have a question.

Eli still seems to be struggling to interpret the data. Just wondering how well he is analyzing his data. How well is he defining good tank vs bad rank? How rigorously has he defined the criteria? Is ”good” only defined as good coral growth? Has he consider that coral growth is the only thing driving the microbiome to his “balance” point?

 

[Dan_P]

And here's Eli Meyer giving a very interesting 2 minute answer to this question on Diversity and Balance score and what bacterial diversity means to him in terms of it being desirable and how his views have shifted after looking at a couple of years of hobby data.
Aquabiomics presentation to PNWMAS (1:19:05-1:21:20)

Yeah, diversity in terms of the bacteria population alone was never based on hard evidence. I guess it was the overwhelming amount of data that changed his mind. Hopefully, his live rock business won’t suffer.

I predict in a couple years “balance” will also lose its luster when enough data is available to show that “balance” is a result of good coral growth not good coral growth is caused by ”balance”. The data might already exist.

 

[taricha]

Just wondering how well he is analyzing his data. How well is he defining good tank vs bad tank? How rigorously has he defined the criteria?

Not at all, really.
He's talked about the fact that reef tanks are such unstudied systems, that he's taken an agnostic approach to what a "good" tank is. So he simply throws out from the statistical average any coral aquaculture facilities (they are not really reef tanks), experimental setups, and any systems reported as bad or crashed, dead livestock etc.

Other than that, it's just a statistical average of hundreds of hobby systems. So the stacked color bar that says "typical" really is meant to represent typical - not "good" tanks.

 

[Randy Holmes-Farley]

Thanks for the efforts and nice analysis of this experiment. I don't have the time to check every video, and you've laid out the issues and proposed interpretations nicely.

There is certainly a ton of interesting science to learn about bacterial changes in real reef aquaria. With so many different niches in a reef aquarium, it's an extremely challenging situation to make interpretations about what is happening where and why in the tank. That certainly doesn't mean it isn't worth doing, but at this point in time it seems to me like an academic tool to run pilot experiments (like the BRS experiment) and to evaluate specific hypothesis.

 

[taricha]

With so many different niches in a reef aquarium, it's an extremely challenging situation to make interpretations about what is happening where and why in the tank. That certainly doesn't mean it isn't worth doing, but at this point in time it seems to me like an academic tool to run pilot experiments (like the BRS experiment) and to evaluate specific hypothesis.

Right. Compared to physical sciences, biology/ecology is so complex it's difficult to reduce the number of variables to get anywhere close to manageable for a system like a reef tank.
And right now when we run these microbiome tests, and ask which we learn more about...
I think the tanks tell us more about how the aquabiomics tests work, rather than the aquabiomics test tells us about how our tanks work. (But there's certainly some of both)