Microbiomes are rapidly evolving, and this computational biologist is out to debug the mystery of why.
We are what we eat but there’s also a host of microbes living in our guts that help us make the most of all that food. Computational Biologist, Dan Knights investigates the dynamic and rapidly evolving relationship between humans and the bugs living within.
Just like macro-organisms, bacteria have specific environmental needs — and in the case of gut bugs, that environment is the gastrointestinal tract. Central to the concept of the ‘microbiome’ is that a symbiotic relationship exists between this community of microbes and their human hosts. Yet when conditions change, so do microbiomes, and a lot is changing in our world right now.
Knights’ work focuses on understanding what makes a microbiome healthy, and what drives it towards illness. Read more about how he melds biology with bytes in the Q&A below.
Since microbiomes can contain so many different species at once, how do you collect meaningful information about them?
“The main way that we study microbiomes is to grind up their DNA and sequence it. If you try to grow the bugs, you get an enormous bias towards bugs that grow easily in the lab. For a long time, people thought one of the main gut bugs was E. coli. It turns out E. coli just grows really well in a petri dish. E. coli is in most people, but in a healthy individual, it only takes up about one tenth of a percent. We’ve found that instead, if you grind up the DNA and sequence it, you can capture all of the diversity that’s there.”
As a computational biologist, how does computer science affect your work?
“We spend about half of our time designing experiments, carrying out experiments, and analyzing the data. The other half of the time is spent developing new algorithms. The most exciting parts are when we’re developing new tools to support the experiments we’re running, so we really get synergy between the two disciplines of Biology and Computer Science.”
How do computer algorithms aid your research?
“The first way in which we use algorithms is to go from the raw DNA to understanding which bugs they come from, what they might be doing, or which types of genes they are. After we’ve done that raw interpretation, there’s another set of algorithms that we use to interpret the biological significance of the bugs we find. We want to know which are the good bugs and which are the bad ones, and we want to build a new test that will tell you how healthy your microbiome is, based on the species present within it.”
What does a healthy microbiome look like?
“A healthy gut microbiome has hundreds or thousands of different species living in it, and that’s actually more healthy than one that has 50 or 100 species. The diversity makes our data sets high dimensional, but there’s also high variation in bugs between healthy people, in a certain person over time, or between people with a given disease.
It’s not something like blood pressure, where you have a simple ratio. Instead, there are hundreds of numbers, so it presents a very interesting computational problem. What we’re doing primarily is trying to enable precision medicine with the microbiome. This means being able to tell as precisely as possible which species and which strains are in a person, and then to be able to classify them as being healthy or unhealthy. If a person’s microbiome is unhealthy, we say that it is in a state of dysbiosis.”
Does dysbiosis happen on a case-by-case basis, or is something broader occurring?
“Something broader is happening with modernization. We have cross-sectional data where we can see that people who are in a developing country tend to have significantly more diverse microbiota than, say, people living in the USA. If you can find very rural peoples living in developing countries, their guts are even more diverse. We don’t know what benefits those missing bugs are conferring, but we do have epidemic levels of obesity, diabetes, Crohn’s disease, colitis, asthma, and allergies — all of which are linked to a shift in your gut bugs, and all of which are also linked to increased exposure to antibiotics in kids.
It’s likely that we’re missing some of the bugs that we evolved to have. Which exact ones are crucial, we don’t know, but that’s one of the things that we’re studying in our primate microbiome project.”
What patterns have you found through these projects?
“We found that when we have samples from wild monkeys, two different species of monkey will have completely different gut microbiota. But as soon as they move into captivity, they lose most of their native bugs and acquire modern human bugs.
It seems as though there is an axis of dysbiosis. If you start with wild monkeys and continue along to captive monkeys, the next group you get to is non-westernized humans, and then all the way at the end, the farthest ones from the wild, are Americans. And that’s not where you want to be. Our studies of recent immigrants in Minnesota have confirmed this pattern.”
How could biotechnology help those suffering from dysbiosis?
“One way is to give someone new bugs — i.e. a probiotic — although what people typically think of as probiotics is a very limited set of bugs. You can find thousands of probiotics on Amazon, but if you look at the ingredients, it’s the same small set of strains over and over. We’re talking about a much broader set, so that would include bugs like Faecalibacterium prausnitzii. That’s a good bug. Everybody who’s healthy has it. Every time we study a disease, it’s depleted in people who have the disease or are at risk, and yet you can’t buy it.
It’s not a probiotic yet, but that’s an example of a bug we could potentially use as a therapeutic.
Another way is to give someone prebiotics, which are food specifically for your bugs rather than the bugs themselves. You could supplement with a particular chemical that you know encourages certain types of bugs to grow.
Or, there could be targeted antibiotics that would protect certain members of the microbiome without being as comprehensive or broadly destructive.”
Why is this dysbiosis happening, and how can computational biologists contribute to solutions?
“Everyone’s trying to put the pieces together right now, because it’s all one giant system. You can’t really study human endocrinology without considering the bugs, and you can’t study how the bugs are affecting the human without endocrinology and immunology, so research teams are converging, and projects are getting bigger, more multidisciplinary, and more reliant on computation.
It took decades to build a functioning model of a complete cell, in a very simple organism. To do that for multicellular organisms and bacterial communities is still quite far beyond reach. The golden apple would be to have a full computational model of everything in the human body. Everything you have to do in between where we are now and that point of
the comprehensive model is fair game for computational biology.
Some people are modeling low level physical processes between interacting molecules. Others are a level up from that, measuring the activity of enzymes and how different expression levels change in response to environmental conditions. Then there are people studying communities of cells and cell-cell heterogeneity, and so on up the line. All of those problems require computational biology.”