Suggested Searches

Ruth Globus and Jon Galazka Talk About Biology Data from Space

Season 1Jul 13, 2018

A conversation with Jon Galazka, project scientist for NASA’s GeneLab, and Ruth Globus, a rodent research project scientist, at NASA's Ames Research Center in Silicon Valley.

NASA in Silicon Valley Podcast Logo

NASA in Silicon Valley Podcast Logo

A conversation with Jon Galazka, project scientist for NASA’s GeneLab, and Ruth Globus, a rodent research project scientist, at NASA’s Ames Research Center in Silicon Valley.

Transcript

Host (Abby Tabor):Welcome to NASA in Silicon Valley episode 99! This week we’re talking with two scientists, Ruth Globus and Jon Galazka.

Ruth is the Rodent Research project scientist here. She advises researchers, engineers and others on how to carry out their experiments on the International Space Station. That lets them study how gravity and other aspects of the space environment influence biological systems.

Jon is the project scientist for NASA’s GeneLab. GeneLab is an open science initiative gathering all kinds of data from biology experiments like these done in space and making them available to all for further study. He helps make sure that the information GeneLab offers will be just what the scientific community needs.

Now let’s listen to our conversation with Ruth Globus and Jon Galazka.

Music

Host: Thank you Ruth and Jon for joining me here together. You guys are both project scientists advising two different projects at NASA that work together. So, we’re going to want to talk about what each of you does, and then how you work together. So Jon, you’ve been on the podcast before, so people should go check out that episode, but Ruth, you’re new to us, so could you tell us a little bit about how you got to be a biologist working at NASA?

RuthGlobus: Sure. So, I was recruited to come to NASA originally to try to answer the question of how bone responds to the last of weight-bearing, normal weight-bearing. This was work being done by Emily Holton, who had worked with Russians decades ago in the Cosmos missions, and was one of the first to establish that in fact there is demonstrable bone less in space. We know that’s the case with astronauts as well.

So, I came after getting my undergraduate degree. I worked for a few years, and we learned a lot of interesting things, establishing a ground-based model to simulate weightlessness. And then I went onto graduate school, and I was recruited back later to say, we want someone back here who has a cell biology interest and capability to answer a little, ask questions at a mechanistic level, and also address the basic questions.

Host:That’s nice, to be asked back.

RuthGlobus: Right. Well, as a project scientist. So, I talk about wearing two hats. One is, I’m a research scientist, I have my own research program, and I’ve continued to be interested in the problem of how gravity and other aspects of the space environment influence the health of cells and bone. As a project scientist, I was asked to help develop the capability to fly rodents in space on the International Space Station, to begin to answer questions we’ve had for a long time as to how does long-duration space flight influence the various parts of the body, and develop an understanding that will both inform biologists on earth how parts of the body, how the cells in the body react to changes in loading and gravity, but also to help us figure out how can we make things better for astronauts?

We’re looking, NASA’s looking to go to very long-duration what we call exploration-class missions. That is, go to Mars. And now we’re going into uncharted territory with respect to understanding physiological changes and biochemical changes that occur in the body. To what extent do they stabilize, and how bad may it get? That’s what long-duration rodent experimentation enables us to do.

So, as the project scientist, circling back to your original question, as the project scientist, I help advise the engineers and the specialists who take care of the operations of doing the experiments in space. I advise them on the science, and how to best ensure the welfare of the animals, but also to achieve the specific science objectives of other scientists. We have scientists from the commercial side through CASIS as well as scientists who are supported through the space biology program at NASA, who ask a variety of different questions, and have their own—every experiment has its own demands placed on how you conduct it.

So, that’s my job as the project scientist, is to help make sure the work that gets done and the changes get made and the processes that get developed all end up achieving the desired science objectives while ensuring the welfare of the animals.

Host:Right, that’s really clear. So, CASIS you mentioned is the partner that helps?

RuthGlobus: No, CASIS is actually the lab manager for the International Space Station. So, their business is to identify and recruit and support commercial and non-NASA use of the space station. So, CASIS sponsors research by pharmaceutical companies. We’ve had seven missions to date. We’ve had principal investigators—that is, those are the lead scientists on a given experiment from companies like Eli Lilly and Novartis. We’ve also had a principal investigator from the Department of Defense. And we’ve also begun to support space biology program investigators. Those individuals typically come from academia. They also include some NASA scientists.

Host:Okay, right. So, the lead scientist on any of the NASA rodent research missions could be from these other institutions.

RuthGlobus: It could be anyone, yeah.

Host:And they’ve probably never flown an experiment in space before, right? So, that’s where you’d come in to help them get everything designed and setup correctly?

RuthGlobus: Well, the whole project supports that. So personally, my function is to evaluate whether the plans that are developed to support the experiment do what they need to do. That’s a simple way of.

Now, early in the project, we devise what’s described as a validation mission. So, we brought together science experts, scientists who’ve flown experiments in the shuttle era and animal veterinarians, to develop the first mission in which we could test out the plan essentially. And that mission had, half of that mission supported a Novartis principal investigator lead scientist, and half of it tested all of our procedures and processes. We carefully monitored the animals and all that.

Now, to get the most benefit from this mission and from subsequent missions, we collected samples from the animals, and we preserved them in such a way that subsequently various types of analyses and assays can be performed, and we can learn even more about them. So, we take the samples, we stored them at very cold temperatures to maintain their quality, and then we sent them out to different scientists around the country who have expertise and interest in analyzing them.

And one group which received some of these samples was Gene Lab. And so, I’ll leave it to Jon to explain how Gene Lab can further magnify the benefit gained from this kind of approach.

Host: Right, exactly, that’s where Jon comes in. Sorry, we’ve been ignoring you. So please, jump in here. So, just as Ruth is the project scientist for rodent research, you’re the project scientist for the Gene Lab project, right?

JonGalazka: That’s right.

Host: So, can you tell us, what is it you do, and what is Gene Lab?

JonGalazka: At the highest level, my job is to again inform the engineers and all the other team members of what the requirements of the scientific community are going to be. So, my primary job is to ensure the Gene Lab product, which we’ll discuss what that is, suits the needs of the scientific community. So, I have scientific training, I have scientific contacts, and so I can give that perspective to the rest of the team.

GeneLab is a relatively new initiative here at NASA, and at its core, what we’re trying to do is drive the production of more data sets, and also the reuse of data sets from biological experiments performed in space. And Ruth has already brought up the rodent research project, and it’s a great example of this. If we again go back to this validation study that Ruth mentioned, mice were flown, PIs and teams did primary analyses on these mice, and then tissues were deposited to a database, or a tissue storage facility.

GeneLab then was able to effectively write a proposal to access those tissues and generate more data from them. So, we generated data on the transcriptome of those tissues. And so, the transcriptome is meant to be a comprehensive view of the RNA profile of a cell. And RNA is very important to how a cell operates.

Host: Okay, it’s kind of what the cells are doing, it tells you that.

JonGalazka: Right. So the genome, you can think of it as being the shard drive of the cell, and then the RNA is produced from the genome, and that then gets converted into proteins. Proteins are doing actual chemistry. And so, the RNA is an intermediary between those two states. And there’s very nice technologies now to actually measure that RNA profile in a very precise and robust way.

So, we were able to do that from a number of tissues from this rodent research validation study. GeneLab, one of the core values of GeneLab is that it’s an open science initiative. So, we generate that data, and we don’t even look at it. We pipe it right out to the public. So, anybody can go to GeneLab.nasa.gov and download these data sets and analyze them. And the idea is that we want to get as many eyes on these from as diverse a background as possible—biologists, data scientists, high schoolers, et cetera—because you don’t really know where the next big discovery is going to come from.

Host: That makes sense. And there’s a lot to be discovered in there, I suspect, all this data from over the years and different types of organisms, right?

JonGalazka: Right. So, GeneLab has about 200 data sets now. A lot of our data by volume is from these rodent research projects, because those were done recently, when we can produce very large volumes of data relatively cheaply. Some of the older data sets are smaller because technology was not as advanced. But we have microbe data sets, plant data sets. Depending on your favorite organism, you can usually find something that’s relevant.

Host: That’s fascinating. I think when we were chatting before, you guys talked about data being really diverse, from different scientists, from different eras. Is that something you’re able to help with, with GeneLab?

JonGalazka: That’s one of our goals. The rodent research project, I think one of the successes of that project going back and up to now is that they have set up a relatively consistent cadence of flights and a relatively consistent structure of flights. And so, it’s really a model for what can be done with consistent schedule and design. We can now start to compare different experiments. So, this initial rodent research validation study, but now there’s new studies coming out. And yes, the goal of GeneLab is to organize that in a way so that making a comparison between those experiments is easy, and you start to pick up patterns from all of that that you wouldn’t see in a single experiment.

RuthGlobus: So, what he’s describing is like a meta-analysis, which biomedical scientists will analyze, will collect the papers, the published papers from all the different studies that have addressed a particular topic and then ask, what’s the preponderance of evidence for or against a particular treatment strategy, or something like that?

What this approach enables is a meta-analysis across multiple samples, but not just of the conclusions that are drawn, but of the actual detailed data, as we described the transcriptome, the whole picture, the whole enchilada. So that when you, in some of the differences we’ve seen, we’ve had 27 space flight experiments with the original hardware on the shuttle missions, and the Russians have also flown rodents, mice and rats, on Cosmos. And there’ve been multiples—the Japanese now have submissions. So, there are a lot of variables having to do with cages and duration and whether the animals are male or female. So, all of those variables, when you can start to look at the big picture, take the detailed data across multiple experiments that might have some subtle differences, you can start to look for commonalties—what are the common things, what are you likely to see, and are you seeing in both, in experiments that are from animals that are both males and females, so you know any scientific conclusions you draw from your analysis aren’t limited by that. To whatever extent you can extrapolate, you know you don’t have to extrapolate to the point of, constrain yourself to a single, to only being relevant to female astronauts, for example.

And so, most of our early missions coincidentally were female mice, not male, and most of the data we have from astronauts is from males, just because of the population. So, that’s what having this kind of database where you can actually do what people in the field call mining, you can mine it, and then you can do these cross-experiment comparisons, and you can try to find out what are the, not the details, but what are the important commonalities?

Host: The bigger picture?

RuthGlobus: Yeah.

Host: Is that what you need to establish before you can start figuring out, for astronauts on long missions, they should be eating this or that for better bone health?

RuthGlobus: I don’t think so. It would be nice if it were as simple as that. That’s where you’re going, was as you build your scientific, I don’t want to call it a database, your framework of understanding, how does gravity influence a multicellular organism—start there. And then when you start to learn the principles underlying that, then you can start to formulate testable hypotheses. And that’s what you do. Ultimately, it could culminate in yes, take this pill and you’ll feel better in the morning on Mars, I don’t know.

Host: Could be our future. So Jon, you’re not a rodent researcher yourself, so how do you work together with somebody like Ruth, or how do you begin to approach working with rodent data.

JonGalazka: You listen very carefully. The key is to understand your limitations, and understand what you bring and where you need help. And so, we do try to reach out and ask the right questions. We have a new initiative at Gene Lab, we call them analysis working groups.

Host: Meta-analysis, is that what she said?

JonGalazka: Well, for any analysis. Meta-analysis working groups, and what these are, are groups of scientists form within NASA and outside of NASA. And we have four of them now—one of them is an animal analysis working group. And so, these groups are meat to look at all the GeneLab data and to critique it to give us feedback. And that’s a really critical part of the GeneLab cycle, is we’re trying to produce a product that is useful, and we need feedback from scientists who are specialists in rodent research, et cetera.

A simple example is just the metadata that a scientist requires for interpretation of these data sets. That type of metadata, metadata basically describes the conditions of an experiment—what’s the strain, what’s the age of the mouse, et cetera. Those simple categories of metadata will be very different for a rodent research experiment than it would be for a microbial experiment, a plant experiment, et cetera. And so, we really rely on those discipline-specific scientist to give us feedback on what’s necessary to lead to strong conclusions.

Host: Right. How would you know otherwise? And then if your data isn’t strong, you don’t want to produce it that way. So, that’ll affect maybe the future design.

JonGalazka: Right. So, another thing to understand about GeneLab is, we don’t actually do any experimental design. We effectively are reliant on the experiments that PIs like Ruth and other rodent research PIs are designing. So, they’re taking the lead in designing a really solid experiment, a hypothesis-driven experiment, if you will. We come in after and look at what’s available from those experiments, and try and generate the best data sets possible.

So in that way, a lot of the heavy lifting has been done for us already by these PIs. They’ve designed a nice experiment, that’s usually been peer-reviewed and has passed that threshold of quality. So, there’s a nice check there already for us. And then we come in and we feel pretty confident in our ability to generate and understand omics data, and so then we come in at that point and try and generate really useful data sets for this community.

RuthGlobus: And the full circle on that is that scientists can access the Gene Lab database and look—give a really practical and plain explanation of it is, look to see what pathways, what things are different in, say for example liver or the muscle of a flight animal, an animal that’s been in space for 30 days, versus control animals that have been maintained in identical conditions on the ground. So, the scientist can go look at what we describe as omics data, and analyze it and then formulate hypothesis. Let’s say that they find a particular pathway that they didn’t expect to be activated, you couldn’t make that up in your brain—I expect that to be different. But in fact they go in and look at it, and it is different. We already, there’s already been some papers that have taken that—lots of papers in the field that have taken that approach.

Now with that, the full circle part of it is they can take that result and they can then ask hypothesis-driven questions based on their analysis of that data set derived from Gene Lab. Now they can design their next experiment, they can write their next grant proposal for their next flight experiment informed by what they’ve learned from that data set. That’s typically how science works, but that’s how this project, the projects work together, building new experiments, providing that data, and then PIs out there going and saying, this is really worth testing, and then convincing the funding agencies to support them, to actually do it.

Host: So, digging through the Gene Lab data can spark new ideas, new hypotheses. I know you talked about lots of different organisms’ data in GeneLab. Can you make comparisons across organisms? Does that make any sense, the effect of microgravity on a bacterium versus a mouse?

JonGalazka: It’s a difficult comparison. The simple physics, the physics at different scales are different. And so what the effect of a lack of a gravity field on a microbe is going to be very different than on a mouse. That said, we’re all related. So a lot of, if you look through the tree of life and you look at the tree of gene families and the tree of life, there are relationships. You can find that a gene in a mouse is related to a gene in a plant. And so, you can start to piece that together. I think those, the further out you get from a concentrated study, the more careful you need to be. It’s certainly a very intriguing idea, and it’s possible. We just need to be a bit careful when you really try and take a big picture look like that.

Host: Yeah, I was curious about that. Also, GeneLab has quite a lot of data in it. But, do you describe it as big data, which is a popular word these days? Is it in that category?

JonGalazka: I was thinking about this on the way over, after lunch. And my answer today is yes, I think it is. Big data, it’s maybe not what people think of in Silicon Valley. We don’t have training sets. I think the typical example is, you have two sets of pictures, some with cats and some without cats, and then you can train an algorithm on that to then be able to detect cats.

We don’t have anything like that. We don’t have—we’ve talked about trying to set up a consistent cadence of rodent research experiments in space, but we’re nowhere near the millions of examples you would need to start to do that type of mining. That said, our database I think is up to 20 terabytes, and it’s big. It’s not as big as some databases, but, there’s challenges associated with simply moving data. So, if somebody wants to analyze this data, downloading it is a barrier. So in that essence, it’s big data. We need to use specific algorithms to analyze it. The primary data is essentially nonsensical. There’s a lot of steps that need to be done to refine it into something the human mind can actually conceptualize. So, there’s a lot of themes of big data that do apply to ours, but it’s a little divergent as well.

Host: All right. Well, that’s what makes the challenge interesting for you, right, working on these. So, 20 terabytes of data, potentially, all available, accessible for people to dig through and invent new ideas and questions to answer, all leading towards what? What would you say would be the goal of GeneLab’s data, or ultimately in your work, Ruth?

JonGalazka: For me it is supporting exploration. I think that the space biology program is ultimately meant to support exploration. So, we use model organisms to understand how biology responds to space. The next step is to translate that to the human body. Now, you understand better how the human body is responding to space. That allows you to design better, safe missions, more effective missions to put humans out into space.

There’s also an argument that what we’re doing in space is going to have direct relevance here on earth. Ruth mentioned Eli Lilly already, and I think there’s going to be interest from pharmaceutical companies to mine this data. The unloading of bones in microgravity is similar to disuse here on earth, and so there’s a lot of I think interest there to understand how those things may connect and what types of discoveries may come out of it. I think our primary mission is to support NASA’s goals, but we really do hope that a lot of different discoveries will come out of this data set.

RuthGlobus: Yeah, I mean, that’s well-said. And I think that I would only expand a little bit on that in terms of rodent research to say that rodents are a very commonly used model for biomedical research, and we understand a lot about them. We have a lot of tools developed that we can use to relate to humans. So, it’s a mistake to say a result obtained in a mouse is definitely going to be obtained in a human, but it’s definitely, rodents have been the really standard way to approach biomedical problems. So, we’re able to gain insight not just into—not only, I won’t say just, not only into mechanisms and the extent of the, assessing the extent of health risk, but also to test ways to intervene. So, that could be using a drug, or changing the genetic makeup of an animal to understand the contribution of a specific gene to a biomedical process.

So, I think that’s the particular relevance of rodent research versus some of the other types of organisms, although I think you do gain great value from looking at flies and others in space. I think for the biomedical relevance, right now rodents is as close as we can get.

RuthGlobus: One of the questions that we can get asked as we were working with rodents is, how do you make the decisions you make, and how do you ensure welfare of the animals? So, we have what’s called an institutional animal care and use committee. Every institution that does research with animals has one. And we are, we and all scientists who work with animals are subject to regulations related to ensuring the experiments we do are consistent with high standards for animal welfare and health. And that ensures the quality of our science as well.

So we have, NASA has a dedicated flight—the abbreviation is IACUC — has a flight IACUC and an attending veterinarian who monitors every experiment, who examines video from the animals and monitors handling of the animals to ensure that they are, that they meet those health and welfare standards. And that helps us ensure high quality science outcome.

Host: That’s a good point to make. What was that you pointed out to me once again space flight flips the usual order of things, when we often have rodent research first before a human?

RuthGlobus: Oh, I was just making, it kind of amuses me that most of the research that we do on earth, the biomedical research, you really, people conduct experiments with animal models, with rodents and other animals before they go to humans. And that’s a safety thing, and that’s actually a requirement, and it’s highly regulated.

So, before humans are given a medication, it has been tested in a minimum of two species. That’s a current FDA standard. And, it’s well understood. And the animal models have been developed for disease states. And so, that allows these biomedical testing.

But we’ve gone to space. We certainly have sent a lot of animals, but what we know about the biomedical consequences of spaceflight are predominantly from studying humans, astronauts. We’ve done a lot of short-term studies, I mentioned them, in the shuttle era with rodents, but long-duration studies have only been done in humans. No rodents have been up there for a year, or even six months.

Host: That’s really interesting.

RuthGlobus: The longest rodents have been up there is three months. And we learn a lot with rodents, because their life span is shorter than humans. And they acquire a lot of the same age-related diseases that we get. They can get cancer, they get osteoporosis, they get muscle-wasting sarcopenia with age-related muscle-wasting. Even their fur turns gray. So, they have a lot of age-related diseases, but that all happens in a two-year period instead of an 80-year period, which is how long it takes us to age. So, we’re able to, shorter periods in the rodent’s life can be used to extrapolate to longer durations on exploration-class missions in terms of the potential for health risk to the astronaut.

Host: Right. So, that’ll help us understand a lot faster what could happen to human explorers.

JonGalazka: Yes, that’s our goal, yeah. So, I think we’re working toward that. One of the things we haven’t talked about a lot that’s the platform for this research right now is the International Space Station. And we can learn a lot about the exposure of astronauts to microgravity on the ISS, and one of the big steps for NASA now is moving outside of low earth orbit. The International Space Station operates in low earth orbit. And what that means is it’s still protected from cosmic radiation. Once we move outside of the protective magnetic field of earth into deep space, which would be required to move to Mars, there’s a lot of questions about how systems are going to respond, biological systems are going to respond to that exposure.

Host: It’s a whole other ballgame, right?

JonGalazka: Right. And so that’s again, when we think about future rodent experiments, how would a mouse respond to these radiation fields, and knowing that should allow us to provide for better protections and mission designs for astronauts.

Host: So, it really is all about exploration. Awesome. Well, that’s all the time we have for today. So, if you have any questions out there, we are @NASAAmes on social media, and we’re using the #NASASiliconValley. So, send us your questions and comments, and we’ll get them to Ruth and Jon, and get back to you with an answer.

Thanks for coming over, you guys, this has been really interesting. Thank you.

JonGalazka: Thank you.

RuthGlobus: Thank you.

Host: You’ve been listening to the NASA in Silicon Valley Podcast. If you have any questions, on Twitter, we’re @NASAAmes and we’re using #NASASiliconValley. Remember we are a NASA podcast, but we aren’t the only NASA podcast, so don’t forget to check out our friends at “Houston We Have a Podcast,” and there’s also “Gravity Assist” and “This Week at NASA.” If you’re a music fan, don’t forget to check out “Third Rock Radio.” The best way to capture all of the content is to subscribe to our omnibus RSS feed called “NASACasts” or visit the NASA app on iOS, Android or anywhere you find your apps.

End