Tuesday, October 29, 2019

Where do species show up together and why it may matter.

As part of my graduate research I’ve been building some new tools to try and determine how healthy watersheds are using patterns of diversity across their constituent streams. The results, which were just published in Ecology and Evolution, suggest the following:
  1. A significant portion of the health of watersheds can be explained just from how likely and frequently different species invertebrates (e.g. snails, insects, etc.) will be found at the same location in a stream.
  2. These patterns of co-occurrence are useful measures of watershed health, even if you don’t know the particular identities of the species involved.
Without going too far into the math, this is the ecological equivalent of saying something about the social conditions of cities just from looking at the structure of their Facebook social graphs. The implication of all of this is that we may be able to start assessing the health of other types of ecological communities just from these patterns of co-occurrence between species, rather than know the particular identities of the species involved.

Sunday, May 5, 2019

Using alpha, beta, and zeta diversity in describing the health of stream‐based benthic macroinvertebrate communities

Hot off the servers, my colleagues and I have recently published a paper on the use of diversity patterns of communities invertebrates (snails, insects, etc.) in various streams in assessing the health of entire watersheds.

Here's the quick version of what we found:

  1. Streams which are downstream of areas where more of the land is dedicated to human activity tend to have fewer unique species.
  2. At a larger geographic scale, in our case a regional watershed, higher upstream land use was also associated with each local stream community becoming less similar to one another in terms of the species present.  This could be called an Anna Karenina principle: every community declines in its own unique way.
  3. Using zeta diversity, a more general framework for assessing changes in diversity across multiple communities, we were able to construct a well-correlated proxy for the California Stream Condition Index, a measure of stream community health.
Any questions on our research?  Please contact me at levisimons@gmail.com

Sunday, April 28, 2019

The shining coast: An ecological survey of light pollution in southern California

Humans, and some of their predecessors, have been pushing back against the night for hundreds of millenia.  Up until the late 19th century this was confined to some form of fire.  Once we figured out how to generate electricity though things really began to take off in terms of lighting up the night sky, and quick.


Originally the only group of people who voiced much of any complaints about this trend were astronomers.  As cities grew, and illumination technology improved in both brightness and efficiency, astronomers were forced to retreat farther and farther away from any human habitation in order to see much of anything besides the Moon.

It turns out the astronomers weren't the only ones picking up and moving away from the glow of human activity.  In recent decades ecologists have begun to piece together a picture of the role of light in dictating how species communicate, avoid being eaten, and find their way.  When humans started to flood large parts of the planet with artificial light we were accidentally running a planetary-scale experiment.  How is all of this light affecting ecosystems in and around human settlements?  Are species migrating away from humans to cope, are some attracted to all of the light, and are some even adapting?  These are questions I'm hoping to address with colleagues at the University of Southern California over the next few years as we map out light pollution along the coasts of southern California.

Why not just used satellite photos to do all of this?  While we're using satellite images of southern California at night are readily available the data they collect on light pollution only gets what gets transmitted straight up through the sky from the surface.  What you won't get from those famous photos of Earth at night is an idea of how the whole hemisphere of the sky will glow looking up from the ground, which is an issue far more important to all of us critters moving about on the surface.

However getting all of those photos of the sky gets to be a labor intensive process, even with an army of graduate students.  So what we're doing is imaging the night sky at approximately 500 locations along the coast from southern Orange to northern Ventura counties, combining those measurements with satellite-based measurements of brightness and cloud cover, and building a model to predict what the night sky should look like along the coast using just satellite-based measurements.  Of course this means we still need to collect those 500 or so images, but that is a much easier task than 10,000.

Since June of 2018 I, along with our team at USC and some awesome high school students at the LA Zoo Science Magnet, have been imaging the night sky along the coast.  We should be wrapping up our photography in February 2019, but if you're interested in coming along I will be messaging the CRASH Space mailing list asking for temporary members of my science entourage. In case you are interested on where we have taken data, and where we need to do, please check out the map below.  The gray-scale points are measurements of the level of light pollution we've actually taken, while the rainbow-scale points are the locations of all of the points we'll eventually image.

[Original published here on October 19th 2018]

The oyster gut microbiome and you.


Science is about humanity's search for the answers to fundamental questions ranging from 'Why is the speed of light the same for all observers' to 'Can I eat that?'. Two years into my graduate school career I now feel secure in knowing that I helped to contribute my little part to our collective understanding of things eating other things, as well as why you should not think too hard when eating oysters. My part of all this is a just-published paper with the snappy title "High turnover of faecal microbiome from algal feedstock experimental manipulations in the Pacific oyster (Crassostrea gigas)".  If you don't feel like reading the whole thing here's the tldr;
  1. If you change the diet of an oyster its gut microbiome will change quickly, on the time scale of a week or so.
  2. Different oysters will respond to changes in diet in very similar ways, at least with what is in their gut microbiome.
  3. This is of relevance as it implies that if any probiotics are developed for improving oyster crops in aquaculture then they can be quickly an uniformly incorporated into multiple oyster gut microbiomes.
Actually, I would recommend cooking oysters.
[Original published here on May 11th 2018]

Sunday, February 26, 2017

Oyster pseudofeces and you.

Oysters poop.  This fact is one of the pillars upon which I expect to eventually rest my thesis, the gold star sticker of academic glory.  In sifting through oyster feces my plan is to get a better idea of the type of microorganisms which live in their gut, the individual oyster gut microbiome, which has a good chance of being associated with the health of farmed shellfish.

Once a week I head out the lab on the island, collect oyster feces, extract DNA, and run through a genetic sequencer to get a snapshot of these microbiomes.  Sound easy enough right?  I've got my MoBio PowerFecal DNA kit and everything.

In science one always starts with a simple enough question, such as 'I tend to see A and B at the same time or place' (correlation), or for the more ambitious there is also the variant of 'I tend to see B given A' (causation).  However, since nature abhors a vacuum, and a vacuum containing nothing is the height (depth?) of simplicity, then it can be reasoned that nature abhors simplicity.  These complications, like infernal epicycles, will quickly work their way into any supposedly simple scientific project.  In my case it turns out oyster poop is not always oyster poop.

This is one of those facts I could have probably gone to the grave without knowing.  It is one of those complications which comes up when you're at least three decades into life, having made peace with being unemployable, and begin reading 'Scatalogical Studies of the Bivalvia (Mollusca)'.  It turns out that shellfish, such as oysters, have the ability to detect which bits of crud they suck in from the sea are digestible and which are not.

How they do this without a brain is not entirely understood, but the filters they use to ingest food can also sense if what they've drawn in is a piece of sand or a type of algae they can't break down.  At this point the oyster covers all of the indigestible bits of marine flotsam in mucous and expels the mass as something called pseudofeces (Diagram below).
Image courtesy of A Snail's Odyssey.

The problem for the sad marine biologist is then trying to sort this pseudofeces out from the feces.  It turns out this can be using the following procedure:
1. Frown.  Make sure enough people in your lab see you doing this.
2. Ask someone on Twitter how to tell the difference between feces and psedofeces.
3. Get an answer.  Wow, Twitter, really?  I thought people only used it to harass each other over gender dynamics in video games.
4. Get a microscope and look at the oyster secretion under about 10x magnification.  If you can still see intact cells of algae then its pseudofeces.  Feces under a microscope pretty much looks like feces, so that's reassuring.
5. Freeze the real feces for all of the gene sequencing later on.

Well, now I've got this part of my project down.

Thanks goes out to Carina M. Gsottbauer for helping resolve the pseudofeces versus feces issue.

Wednesday, February 22, 2017

Everyone poops, even in science.

Cities are pretty great inventions.  They're places where we can conveniently find a new type of burrito every day for a month, avoid eye contact on the bus while staring intently at our phones, and have important conversation in corner cafes (I hope everyone here knows I'm working on my new screenplay!).  What then does it take for a city to exist?  A modern city certainly needs tons of robust materials such as concrete, along with pipes and wires to keep everything from natural gas to cell signals work.  However humans have been try their hand at the urban game for roughly six millenia, long before most of what we would consider key infrastructure.

What cities need, now and forever, is large-scale food production.  That was true for Angor Wat and Chichen Itza, and remains so for London and Mumbai.  You can't hunt and gather your way to Manhattan, and a modern city is just as reliant upon the collective stomachs of its inhabitants as any of its ancient counterparts.

The most recent projections of human population growth put global population at near 10 billion by mid-century.  While the total population of Earth is not expected to grow beyond this point there is the expectation of an increase in the demand for food beyond basic grains and towards foods, such as meat, which provide a much higher density of protein and fat.  While it appears that we will avoid any sort of runaway population growth then we should expect agricultural demand to grow throughout this century long after any peak in human population.

Where then to get this food?  Close to half of the planet's land is currently dedicated, in one way or another, to the production of food for human beings.  While more intensive cultivation techniques, whether through changes in soil management and genetic modification, are likely to be implemented over the coming century there is a growing effort to expand our collective efforts and cultivate the sea.

Now aquaculture is not new.  Humans have practiced various forms of it for centuries.  What is new is, through a combination of the decimation of wild stocks and the expansion of cultivation in marine systems, that we have hit parity between cultured and caught seafood.  Just as we couldn't hunt or gather our way to Manhattan on land we won't be able to on the sea either.  The growing demand for protein, and in particular seafood, necessitates an improvement in our practices for cultivating the sea.

One such crop is the oyster.  In the case of my work it is the Pacific oyster, the most cultivated species of the oyster family.  Oysters lend themselves well to aquaculture.  They are sedentary, naturally grow in dense agglomerations, and will feed themselves by continually filtering seawater.  While that last part comes in handy it does mean that oyster nutrition is entirely dependent on whatever microorganisms happen to be floating by in their local patch of ocean.  This does not mean the oyster is alone in its noble battle to survive off of whatever it strains from the sea, for the gut of the oyster turns out to be its own little niche for various microorganisms to thrive.  This community, known as the gut microbiome, lives in the intestinal track of every oyster and is most likely a significant factor in how well it can digest food.

I say most likely a significant factor because, while the gut microbiome of some species such as humans have been studied in great detail over the past few years, that of the Pacific oyster has only just begun to be studied.  My interest in all of this is to get a better look at what lives in oyster guts because the ability to digest food is directly tied to the ability to incorporate plankton into more delicious oyster meat for humans.  To start to tackle my goal of finding if there are certain bacteria, or combinations of bacteria, which can improve oyster growth rates I'm starting with three questions.  One, does the diet of an oyster shape the composition of the gut microbiome, that is the relative abundance of all the different bacterial types in there?  Two, will two closely related oysters also have gut microbiomes more similar than if they were unrelated?  Three, is the growth rate of an oyster predictably influenced by its gut microbiome?

This spring I'm starting with the first question.  If the gut microbiome has any significant connection to digestion then it can be reasoned that different diets will cause different gut microbiomes in the same oyster.  This means I've got the glamorous task feeding multiple oysters, in this case fifteen, a controlled diet in individual tanks and then collecting their poop on a weekly basis.  The oysters are all in their individual tanks in order to prevent any microbial cross-contamination in the study.

Wait, come back, this is important.

Why look at poop?  Well, just as everyone has a gut microbiome everyone also poops.  It also turns out that collecting feces is a pretty standard method, as far as these endeavors go, of getting a snapshot of the gut microbiome that day.  In collecting oyster poop I can then, with the help of a particular type of genetic sequencing, get a census of all of the types of microorganisms living in the intestinal track on a regular basis.  Every month then the diet changes for ten of the oysters while leaving five as a control.  If the diet really has an effect of the gut microbiome the end result should be a stable gut microbiome for the five control oysters and a changing one for the ten experimental ones.

It turns out that you may one day, by taking a good look at oyster poop, that you may get more oysters for you buck.

Oh, and in case you're wondering, the ones you do eat have been eating at restaurants have been depurated.  This is a fancy word to work into your next conversation meaning 'pooped empty'.


Thursday, June 9, 2016

The Pangea of Commerce Part 2: Is it alive?

I've recently begun putting together the pieces of what will be my thesis project as a graduate student in marine and environmental biology over at USC.  In this project I will be testing the idea that the networks formed by ocean shipping traffic will have a measurable impact on the biodiversity in the waters in and around the ports in question.  My hypothesis is that the similarity in the composition marine microbial communities, that is the type and relative abundance of species, between two ports will track the rise and fall of shipping traffic between them.  My prediction then is that a rise in shipping between two port, say Los Angeles and San Diego, will cause the microbial communities in both ports to look more similar than before.

Seems straightforward enough, at least given modern ship tracking and genomic sequencing, but there are a number of complications which are going to arise in getting meaning out of this data.  One such complication is that sequencing the genomes of all the microbes in a sample of water will give you a census of microbial populations, but it will not tell you how active any of the species may be, or even if they're alive at the time of sampling.

Over the years a number of methods have been developed to try and measure the metabolic activity of microorganisms.  One method involves feeding sugars, with altered isotopic ratios, to a sample of microbes and measuring how quickly the altered sugars are incorporated into organic material of the microbes.  This can involve shifting the amount of carbon-12 to carbon-13, or oxygen-18 to oxygen-16.  Once a group of microbes has been feeding on foods with different isotopic ratios a microbiologist can send a sample of their biological material through a mass spectrometer and see how much of the altered carbon, oxygen, etc has been incorporated into the microbes' body masses.  This gives a good measure of how quickly the microbes are eating in the first place.  This method has been useful in studying microbes which can be readily cultured in the lab, however many of the species I will be looking at in ocean water are not readily cultured.  Also, I will probably be studying dozens, if not hundreds, of total samples.  This method will probably not scale very well on account of these issues.

Is there a method for measuring the activity levels of a large number of microbial samples, even when a number of the species involved cannot be readily cultured in the lab?  It turns out that there is, sort of.

In part 1 I mentioned the use of sequencing ribosomal genes as as method for determining the relative populations of microbes in a sample of water.  These genes are coded in DNA, which is the molecule that stores the raw instructional information for cells.  In this case the genes which code for an organisms ribosomes are known as ribosomal DNA (rDNA).  The way this information is then executed in a cell is through the use of RNA.  RNA is used to make copies of particular genes, which are written in DNA as messenger RNA (mRNA) with the assistance of transfer RNA (tRNA) .  This mRNA then works with a series of structures in the cell known as ribosomes, which are composed of ribosomal RNA (rRNA), to synthesize the proteins which perform all of the functions of life.  This process results in large variations in the amount of rRNA versus the corresponding rDNA.

As protein synthesis is associated with the activity of a cell it was then assumed that measuring the amount of rRNA versus rDNA would give a straightforward picture of how active a population of cells were.  If this were the case I could add some very precise activity data on top of microbial census.  Unfortunately it turns out that there is no set relationship between the level of rRNA versus rDNA and cell activity.  The rRNA to rDNA ratio varies between species, and doesn't even scale linearly with metabolic activity and cell division rates.  It also turns out that cells can have a high rRNA to rDNA ratio while they are currently in a state of low metabolic activity as a way for a cell to prepare for times of high metabolism, even if they are currently in a dormant period.

What does this mean?  The rRNA to rDNA ratio does increase during times of high growth and metabolism as cells are busy synthesizing proteins.  However the rate is non-linear and varies from species to species.  In addition to that there is a temporal uncertainty as it cannot be determined from one measurement if the cells are currently active or preparing for being active.  Will it then be worth sequencing the rRNA component of all of these cells along with their rDNA component?  Yes, if only because I plan to look at samples from the same ports on a repeated basis.  My hope is that repeated measurements will help to reduce some of the uncertainty in time, especially in differentiating currently active from potentially active cells.  At the very least this data may turn out to be quite useful for some aspect of my project yet unconsidered.

How then to determine the frequency and duration of measuring these populations?  I will try and address this matter in part 3.

The process of mRNA, tRNA, and rRNA synthesizing proteins.