Interview with Dr D. Marshall Porterfield on leadership in the GeneLab Open Science program at NASA

0:05 

Good afternoon and a very warm welcome to Dr D Marshall Porterfield. 

 
0:10 
Thank you very much for joining us for today's interview where we will be discussing your leadership in the GeneLab Open science programme at NASA, and the advancement of integrated omics for space medicine. 

 
0:23 
Let's get started with the first question, which is what initially inspired your interest in integrating omics with space life sciences? 

 
0:34 
So I've been working in the area for my entire career. 

 
0:39 
My graduate work was focused on understanding mitochondrial stress responses that we found in some of the earliest plant growth experiments. 

 
0:49 
And also too, I started working in biosensors and developing really analytical systems to approach that system and do mitochondrial physiology at the cellular level and developing biosensors and really thinking about biosensing in the context of the broader field of biology. 

 
1:10 
I was actually working on a drawing one day and trying to see the flow of biological information and how it connects through at the cellular level. 

 
1:18 
Looking at signalling and biomolecules and what we are doing in biosensing at the time using enzymes, and really thinking about the use of the molecular probes and sequencing is almost like sensing - as we're sensing qualitative versus quantitative information with these different systems, and how sensing is actually leading to these potentially breakthroughs in biology. 

 
1:43 
So, the ability to sense the information in the DNA molecules and sequence that at the genomic level is a huge, huge breakthrough. 

 
1:55 
So we're looking at, you know, around 2000 with the Human Genome Initiative. 

 
2:00 
And also, I was influenced by when I was at Woods Hole at that time, Monica Riley was working on some of the earliest bioinformatics, and I was seeing literally how biological information was really creating a new field and how that was emerging in terms of these larger bodies of biological information that we have access to. 

 
2:23 
And we had used this platform for actually developing new biosensing and new biosensing strategies. 

 
2:28 
And really that's what RNA-seq has become. 

 
2:31 
But at the time, I was really looking at how we integrate all that, the different omics techniques that were emerging at the time. 

 
2:39 
To me, I was seeing those as different sensors of the same thing. 

 
2:42 
Really we're studying the central dogma and have a sensor technology allows us to intercept the central dogma biology and read that information from it. 

 
2:55 
So we can take snapshots with multiomics and starting to build that up really as a way to really crack the code of biology and really understand it at such an integral level that it would catalyse the next generation of life sciences and medicine coming from that. 

 
3:15 
And, and really, if you know, those analytical techniques and the emergence of RNA-seq as a research technique - that really became the focus of really the first phase of GeneLab. 

 
3:28 
But really the entire vision was really to see integrated omics emerge so that you can actually, in one experiment, see, you know, everything from gene expression all the way out to metabolomics and proteomics and really be able to connect all that. 

 
3:43 
If you have access to that kind of information, you can really start to decode biological systems that are really, really high level. 

 
3:50 
In a sense that's really what we started doing with GeneLab. 

 
3:53 
That's what it's it. 

 
3:55 
You know, if you look at the body of research and how that's advanced forward to some extent, it's actually enabling that kind of decoding. 

 
4:03 
Fantastic. 

 
4:05 
And, in that sense, how would you define precision medicine in the context of space exploration? 

 
4:16 
In the context of space exploration, precision medicine really is based on that, you know, working with individuals based on their sequence genome. 

 
4:26 
So every single crew member, you'd have their genome sequenced and you'd be able to basically programme all medical procedures and based on their genomic responses, you would actually have probably classes of drugs and everything pre-set up based on, you know, their predisposition for different diseases and different responses to different treatments. 

 
4:49 
You could actually have mapped all that out and actually have a lot of that really defined, you know, really, that was really an interesting issue when GeneLab was emerging because within NASA, within the agency, the argument was, well, this is what NIH should be. 

 
5:05 
This is under NIHs bailiwick. 

 
5:07 
NASA shouldn't be doing this because that's national, that's health. 

 
5:12 
But, my argument was since the crew members are so integral to these missions, we should be taking the lead on precision medicine because we're going to need it before anybody else really at the cutting edge of human exploration. 

 
5:28 
We're going to need these advanced technologies probably before it really emerges for the broader society. 

 
5:33 
So NASA should be pushing these things forward to create that spin off. 

 
5:40 
Can you talk us through some of what you would see as the key highlights or breakthroughs that you've achieved through your work in terms of GeneLab? 

 
5:57 
If you look at how that's evolved and the community has started to interface with and use omics and transcript omics as a, you know, foundational tool, it becomes the base. 

 
6:12 
It's really become the baseline analytical platform. 

 
6:16 
We have really, everything we do has to have RNA-seq because that's how we measure, that's how really how we test our hypothesis. 

 
6:24 
We have, we think this pathway or this gene is going to be perturbed and we could do RNA-seq and we could see those things. 

 
6:29 
In the old days, we'd have a, you know, we do PCR on that gene. 

 
6:35 
We'd have a hypothesis and a gene, you know, that's really how we transition from a single hypothesis and single gene to this really multi hypothesis or any hypothesis, every hypothesis. 

 
6:47 
And that was really what the transition was with, you know, the transition we're trying to see move forward with GeneLab is that really open science approach where you don't really have to. 

 
6:59 
We could fly an experiment, you know, comparing space to ground, but the specific hypothesis that you're interested in doesn't matter. 

 
7:07 
You really can run, you can fly, you can do these experiments for the community and really, you know, really have the primary researchers be hundreds of people that are accessing the primary data. 

 
7:19 
And in a kind of ecosystem of competition, scientific competition, be able to share and then reinvest in the translational research. 

 
7:30 
It's going to take those fundamental findings and then translate them in ground based research to new knowledge. 

 
7:36 
And that's really what we haven't been very effective at, we see phenomena in space, but we don't actually have the throughput to actually convert that to new knowledge and which will give us new techniques and countermeasures for human exploration. 

 
7:50 
Yes, 

 
7:51 
So I think that touches on my next question, which is what would be the most significant technical or institutional barriers you have faced with launching GeneLab? 

 
8:03 
Well, when we first started on socialising this, it was actually back in the 2012 time frame. 

 
8:12 
So the strategic plan was released in 2014. 

 
8:15 
But in that period of working up to that, we actually developed that by working with the research community and brought in different stakeholders. 

 
8:23 
Actually, you know, we were working in interacting with NIH at the time with some of their early omics based efforts. 

 
8:33 
But really it was NIST and they're concerns about standardisation at the time of omics techniques - that really I think helped influence the process. 

 
8:44 
But really one of the main issues was, you know, the PIs and their acts in controlling their data and changing some of the rules, which now with NASA that all PIs are required to submit their GeneLab data, their sequencing data, and that it has to be open shared a year after they had completed their experiments. 

 
9:09 
And what that's done is really created a really rich body of active research data that's coming from primary experiments on the ISS. 

 
9:20 
And then people are coming back in and being able to do meta-analysis on those types of experiments. 

 
9:26 
And that's really opened up. 

 
9:27 
And, you know, really then I think the next phase of research because we've been able to take human data, mouse data, nematode data, fruit fly data, and really stack across species. 

 
9:38 
And that was one of the things I envisioned early on with GeneLab, being able to use species as a philtre. 

 
9:46 
You know, what's the same and what's different between a human, a mouse, a nematode, a fruit fly, and even a plant in space. 

 
9:54 
And those have revealed some of the most really fundamental responses in space are mediated by mitochondrial systems. 

 
10:04 
So that emergence of the, you know, really the primacy of the mitochondria as an integrator of overall health and also probably a primary receiver and integrator of physiological stress. 

 
10:19 
And actually probably the site of most physiological stress and signalling related to cancer. 

 
10:25 
That's really emerged as being one of the primary focuses for future research. 

 
10:31 
Absolutely. 

 
10:32 
I think, as you touched on it in terms of focusing on the metadata, 

 
10:37 
can you speak a little bit more in terms of your uses of AI and machine learning, and how this has evolved in its role within GeneLab? 

 
10:51 
Yeah. 

 
10:52 
So that was one of the things I that we had envisioned from the very beginning is that you would never be able to approach these types of data sets and extract the kind of information that we would need to until really we evolved AI/ML. 

 
11:06 
And you know, really, I was kind of observing this evolving over time in terms of bioinformatics and the types of tools that people are using and the crossover between computer science and biology starting to become stronger and stronger in those particular areas. 

 
11:31 
And, leading on from that, you touched on this, in terms of how you've navigated the transition from traditional research frameworks to open science and having to look at FAIR principles at a federal agency like NASA. 

 
11:48 
Well, the adoption of the FAIR principles was something that came later as GeneLab started moving towards working groups as a way to enhance the open science approach and data sharing. 

 
12:03 
And so of course, you know, AI/ML are really the integrating tools that allow you to integrate, you know, across these different types of experiments. 

 
12:12 
Really that's been a revolution on its own - all the availability of these open source tools and the working group communities, how they're starting to integrate these AI tools and that's further created more enrichment in and use of the data in the community. 

 
12:30 
So really, you know, the emergence of those tools. 

 
12:32 
I think when we were first were talking about the use of AI/ML in, in GeneLab in those original days, it was before really AI/ML was actually highlighted. 

 
12:42 
So maybe, you know, people's conception of what we were talking about then was really different than, you know, acceptance of what AI can do now. 

 
12:50 
So there's much more acceptance of that. 

 
12:52 
So that actually transforms our ability to, you know, integrate all this data at these different levels and then pull in the metadata too. 

 
13:02 
So that's one of the things that really was enriched by GeneLab. 

 
13:06 
It was, it was almost ironic when we first started developing GeneLab and started advancing the concept because one of the arguments against GeneLab was, well, what are you going to do about the metadata? 

 
13:16 
What about metadata? 

 
13:17 
I'm like, yeah, what about it? 

 
13:19 
Perfect, perfect. 

 
13:21 
Well, yeah, we're going to take care of that metadata problem and we're going to integrate it into GeneLab. 

 
13:25 
And the result has really been amazing. 

 
13:28 
The papers that are coming out, these meta-analysis papers that are coming out of GeneLab now really have transformed the field. 

 
13:35 
Fantastic. 

 
13:39 
So can you talk a little bit more about your partnerships or interdisciplinary collaborations that were involved in GeneLab’s development? 

 
13:51 
Well, one of the things that's really interesting about the emergence of GeneLab - that maybe is hidden in the history and that's related to the twin study that would be of interest - is that when we first started advancing that idea GeneLab, this was before the twins study had been formalised. 

 
14:16 
And we'd actually worked with the Russians on, you know, with our annual working group meeting. 

 
14:22 
And it was in one of these meetings where we were negotiating how we were going to allocate our future research, shared research on the ISS. 

 
14:30 
And the Russians started talking to us about doing a one-year mission. 

 
14:34 
Well, it turned out that was really motivated because they were trying to do a tourist flight for Sarah Brightman to go up to the ISS and sing a song. 

 
14:42 
And in order for her to do that, she would have to go up, sing her song and then get she didn't want to stay for six months. 

 
14:48 
So she would take one of the seats for the one of the astronauts that would have to come down. 

 
14:53 
So the Russians were, they were kind of coy saying we want to do a one year mission. 

 
14:57 
Later on it came out, we figured out that they were actually trying to facilitate this tourist flight. 

 
15:02 
So we agreed to the mission because it did create some opportunities scientifically to do the one year experiment. 

 
15:09 
And then later on when the astronaut office actually selected the crew member who would do the one year mission. 

 
15:15 
Scott Kelly was selected without consideration for his, you know, his brother. 

 
15:20 
And it was actually Scott Kelly who kind of brought that up. 

 
15:23 
Well, my brother is a twin. 

 
15:24 
Would that help? 

 
15:25 
And that catalysed some talk within HRP and then they realised that that would be perfect in the context of this emerging era of GeneLab. 

 
15:34 
And that's what how the twins experiment became that first genomics based experiment for NASA and really the foundation for really what NASA is moving forward now in terms of precision and personal medicine. 

 
15:48 
And then I suppose if we could look at some of the biggest limitations you've come across in terms of current multiomic data integration for Space Biosciences and how you've overcome those? 

 
16:04 
Well, the approach we're using right now is really kind of a blending of ecophysiology and biophysics approach. 

 
16:14 
So when we get these omic data sets back, we're getting information. 

 
16:18 
But what we really want to know is what's happening at the cellular molecular level, what's happening, specific cell type, can we get, you know, the emergence of single cell omics is really actually a major influence in what we're doing right now because it allows us to really extract much more information from the primary transcriptomic information. 

 
16:41 
So what we're doing is able to, you know, basically look at different varieties of organisms that have that are pushed into different ecosystems and they have different types of physical biophysical stress responses they have to deal with. 

 
16:54 
So you can kind of use that as a fingerprint, a reactome kind of fingerprint. 

 
17:00 
And so that's really been one of the major tools that we're starting to use is really this kind of physical AI type of approach where we're really trying to understand the physical environment that relates to these omics responses. 

 
17:14 
And that also involves biophysical modelling and creating really almost like a digital twin of the organism and the biophysical environment. 

 
17:24 
And now we're starting to really integrate the cellular and tissue data responses into those organismal models. 

 
17:33 
So we can really start to integrate the physics and biophysics and start to generate these models that would actually start us to be able to decode these, the broader genomic information. 

 
17:46 
You know, one of the really interesting things about all this that we, we just keep learning all this incredible information. 

 
17:52 
We didn't know anything about the, you know, the genome at first. 

 
17:55 
And then once we started getting some of the information from humans and plants, you know, the predictions were that plants are going to have 10,000 genes and the humans would have 100,000 because they are so much more complex. 

 
18:05 
Turns out they have about the same, you know, plants are more complicated biochemically. 

 
18:09 
They can't run away from things. 

 
18:11 
So they have to be able to do more. 

 
18:12 
And understanding all that, really being able to decode it is, you know, is really 97% of the genome. 

 
18:23 
You know, only 3% of the genome actually has coding genes. 

 
18:26 
So that discrepancy between the amount of information that is there and what we can actually read that actually is a codable element. 

 
18:33 
That's really interesting to me. 

 
18:35 
You know, what's the operating system? 

 
18:38 
It's in there, it's operating system for the cell, but it's also the operating system for how you grow in whole organism. 

 
18:47 
So that's really the frontier of omics and multi omics is through these types of integrated approaches. 

 
18:55 
And that's why I've always - from the beginning - stressed integrated omics, not multi omics, because integrated omics means you run one experiment and from that one experiment you measure transcriptomics, proteomics, metabolomics, and you can integrate across that one experiment. 

 
19:10 
And you have to actually, at some point we have to calibrate across these different omics channels so that we can be able to really read that information, understand it better. 

 
19:22 
I see us ultimately being able to. 

 
19:24 
You know, the way that I approach sensing and the way that we use biosensing experimentally in the lab in terms of physiology, I see us being able to do - use RNA-seq that way one day where you can have real time read out of what genes are being expressed and how those gene expression patterns are changing. 

 
19:44 
So you can really introduce stress and then see those pathways, and see the other coordinating elements that are part of that actual operating system. 

 
19:55 
Fantastic, is there anything else that you're happy to share with us in terms of upcoming projects or experimental missions that are building on the foundations on GeneLab? 

 
20:04 
So one of my main projects right now is the NASA LEAF experiment. 

 
20:11 
It's actually a mission-class payload experiment for Artemis 3. 

 
20:18 
We're building a plant growth experiment, which will be a payload that would be deployed from the Artemis 3 Lander on the lunar surface as part of the Artemis 3 mission. 

 
20:30 
It is a standalone plant growth experiment that would operate on the lunar surface, it will be activated by a crew member and it will run automatically and grow plants on the lunar surface. 

 
20:40 
And then at the end of the experiment, it will actually prepare a sample in return that we're going to be able to get RNA samples from. 

 
20:50 
So basically have an RNA-seq sample coming from the lunar experiment, potentially from three different species of plants that we're working with and that will be integrated and analysed through NASA GeneLab. 

 
21:05 
And ultimately, after the team has a chance to analyse it, those data sets are going to be made available to the broader research community through NASA GeneLab as part of the broader space biology collective research data. 

 
21:22 
So, you know, the vision of multiomics and precision medicine in space really is moving forward. 

 
21:30 
It is a foundational platform for what we're doing now and in the future. 

 
21:37 
We see omics as being a major tool for integrating humans and agricultural systems in terms of ecological life support. 

 
21:44 
That's really my main focus for my, you know, for my career - is developing regenerative systems that integrate plants, animals, fungi, microbes in terms of an ecological life support system. 

 
21:56 
And ultimately, I see omics as being the main controller for that system because you're going to have to connect the human, human microbiome, human nutrition and needs to the soil. 

 
22:10 
But we're, we're just part of a continuum, the microbiome, our microbiome that comes back to us from the plants we eat. 

 
22:17 
We’re just part of that continuum. 

 
22:19 
And really what you know, omics would be is the ultimate tool to be able to integrate all that. 

 
22:24 
When you start looking at hyperspectral sequencing and being able to look at the genome of the soil and of the gut and of the human and have that all in a closed integrated system. 

 
22:36 
Those are the types of test beds we're moving towards for NASA, but those are also the ultimate test beds to drive omics forward because you're going to actually have a closed system. 

 
22:46 
We can integrate and calibrate omics at the ecosystems level, potentially with using as part of this advancement of human habitation systems. 

 
22:57 
And so that that ultimately gives us the power to control and close our ecosystem needs here on the planet. 

 
23:06 
So if you think about what we need to be able to do to live in space off planet with a closed system, that's really what we need to develop here so that we can solve our own ecological imbalance problems on this planet. 

 
23:19 
So we're going to have to figure out how to save this planet before we can actually leave it. 

 
23:25 
Thank you very much. I think with that, we'll close today's interview. 

 
23:31 
Thank you so much again to Dr Marshall Porterfield for your time today and for sharing such fascinating insights into your work. 

 
23:41 
So thank you very much again and take care. 

 
23:46 
Thank you.