Dr. Barkha Yadav-Samudrala: 0:06

Hello, everyone to another episode of nerd RX podcast and I'm your host Barkha. Today we will cover a genomic topic called high C. And honestly, I have no clue what that is. So we will hear from our guest, Alexis Stutzman welcome Alexis to the show.

Alexis Stutzman: 0:27

Hi, thank you so much.

Dr. Barkha Yadav-Samudrala: 0:30

Thank you so much for being here and talking about Hi, see. So before we go into the topic, we would love to know more about you. And how did you get involved in high C and research in general?

Alexis Stutzman: 0:46

That's a really good question. So I'm actually originally from an Amish community in northern Indiana. So there's not a lot of science going on there. And my family there, they left the Amish A while ago, I'm mixed. So there's some Amish ancestry floating around in my family tree. And my great grandfather left the Amish and started a chicken business. But none of that really appealed to me. I didn't want to sell chickens. So when I was growing up, I found a lot of excitement in research. So when I was in third grade, my grandfather had an experimental medical trial where, so he had a kidney transplant and with a rejection. And so the experimental trial was to take care of his rejection. And basically, his team was led by a bunch of PhD scientists, he was he was an experiment. And so I just thought that like the science of biology and medicine was really cool and appealed to me a lot more than practicing. I thought I wanted to be a doctor like every young child does, at some point. So yeah, I just really was really, like, I thought that the science of what was happening was really cool. And then I also saw that that science can be applied to save someone's life. And I thought, this is the place for me, that's, that's perfect. So when I was in high school, I got really involved in my biology classes. And that was sort of the the time that I decided I wanted to go to graduate school for real. And I remember sitting in 2012, when I was watching the Nobel prize being announced, it was Shinya Yamanaka, who won for the Nobel Prize in Physiology or Medicine. And it was for inducing poor potency and stem cells. And I just thought, Whoa, this is like really cool, you can reverse development. And so it was like this video there. And I ended up knowing I wanted to go to graduate school ultimately. So I went to the University of Chicago for my undergrad and got involved in research right away, I started during my first year, the winter there. And my first lab was actually a genomics lab. We did a lot of chip seq experiments, which I'm not talking about today. But this is sort of how I came to genomics, I did a lot of chip sequencing and mammalian cells. And we had this really cool system where you have like two versions of a protein, they're both tagged with different tags. And if you do chip seek for both of those tags of over a time course, you can get a sense for whether the one epitope of the protein is staying stably bound, or if it's switching on and off, because you would detect both of those tags during the time course at a single genomic locus. And so this was like perfect for me, it was putting together all these logical sort of puzzles, it was in a million contexts. And then I sort of left my passion for disease behind and really focused in on the genomics. And then I got involved. So that was kind of my first project. And then the next one, I moved labs and was studying the structure of the nuclear lamina. So the lamina is like this meshwork, this protein meshwork that gives physical support to the nucleus. And we found that this nuclear envelope protein may be a transcription factor. And I just thought that was so cool. And I studied a lot of heterochromatin. So then, when I came to UNC, I had all these questions about like what shapes chromosomes make, how that has an implication for actual transcription or silencing, maybe even heterochromatin requires structure. And so I had all of these really big questions and sort of had my own research project mapped out like the perfect one for me. And I was really just kind of trying to find a lab that would allow me to explore these questions on my own. And that's how I came to UNC for my PhD. And now I'm doing exactly that I'm studying the gene the shapes that chromosomes make in three dimensional All space. And in particular, I'm focusing on heterochromatin, the regions of the genome that are silent.

Dr. Barkha Yadav-Samudrala: 5:07

Wow, I am so, so much in all that you had your roadmap planned, while in undergrad, I don't even want to think what I was doing in my undergrad. Oh, my God, I had not figured out anything. So let's talk about Hi See, then I'm really intrigued by this topic, because I was reading some of it. And it was still like, I could not wrap my head around it. So let's hear it from you. So what is high C? And why is it important? And why would you use it?

Alexis Stutzman: 5:47

Yeah, so from a very basic level, high C measures DNA DNA contacts. So I just talked for a second about the shapes of chromosomes make, I think that everybody in their mind thinks about chromosomes. It's like the X type of structure used in karyotype. But that's actually only what a chromosome looks like when it's going through mitosis when a cell is going through mitosis. So during the rest of the cell cycle, the majority of a chromosomes lifetime is actually spent in a non random shape somehow in the nucleus. So in mammals, they keep their chromosomes and like these territories, so they occupy different like domains of the nucleus, you could say they have like their own little spot. But then there are certain parts of the chromosomes that are unwound, so that gene expression can happen, and certain parts that are compressed, and we think that that's important for keeping those genes silenced. So in order to actually make all these assessments and measurements, people had to have an assay that was capable of measuring those DNA DNA contacts in three dimensional space. So that's where high C is coming in. High C is, is built on another assay, where basically, you're able to probe one particular genomic locus and ask what other contacts it makes with other places. And then genome Heisey is a little more complicated because it's looking at the whole genome at one time. So any to DNA DNA, or any two loci of the genome might be physically together in the three dimensional genome, but very far apart from each other on the linear genome. I mean, you can even think of circumstances where you have an enhancer on one chromosome and a promoter on the other chromosome. And maybe they have to physically be in three dimensional space next to each other for that gene to be expressed. That's some of the questions that lead I think, to the development of high C and sort of a lot of the technologies in this space. So from a basic standpoint, right high C is measuring the frequency by which to local or to DNA loci are interacting with each other. It does that by basically fragmenting the genome in some way. So you just cut it up with some kind of enzyme. We interestingly, have not seen sonication being used in this step. And I think that's probably because people have tried it and it didn't work. I personally have not done it, but you can use some enzyme to fragment the genome, and then you stick biotin on all of the free ends of your DNA, and ligate, all of those biotinylated free ends, so that anything that was physically together, three dimensional space now has a biotin, and is making a little DNA DNA circle, you could say, because we've now ligated, all those free ends. And if you use a strep Avidin bead, which has a high affinity for biotin, you can pull out all of those junctions that are now between two far away DNA loci and sequence them. And so you have to do paradin sequencing, where the direction of your DNA sequence actually matters, because when you're mapping it back to the genome, you expect the different ends to map to different places in the genome. And then you can just take the read ID from where that read originally was in order to like mash them together again. So this lets you look at all of the potential contacts that could be happening in a single genome and one single experiment. There have been some adaptations to make this like single cell and or you can do from a population, which is what I have done. And so yeah, once you have that information, then there's a lot of computational analysis to actually go back through all of those reads, and assign them to each other. And then from that, the the best way we've come about with visualizing this is by making a matrix where the x and y coordinates are just different genomic locations and then the the Pixel where they meet has a is a reflection of how often those two loci are interacting. So it's like this really big heat map is what it looks like. And the diagonal is really dark because two loci that are together on the linear genome are more likely to found be found next to each other. And then as you get further away, the pixels that are enriched are more meaningful, because if you get a contact between two things that are far away, they've sort of like broken, the most likely probability of like things close together will be like it. So that's the general gist of Heisey. Yeah, you're you're looking for DNA DNA contacts.

Dr. Barkha Yadav-Samudrala: 10:43

And by the way, does high C stands for something.

Alexis Stutzman: 10:48

Um, it probably does. So high C comes from another assay that I mentioned, it's called three C, there was a development of three C. So in three C, you use primers for these two genomic loci are specifically interested in. And you see, basically, you, you know, you fragment your genome, you do some ligation, and then you see, if they were interacting with each other in 3d space by PCR based assay, then there was foresee that was three C, then there was four C, which is an improvement upon that, we're now you just pick one locus, and you want to ask all the, or you have like an idea of many other contexts it could be. And so you put together a list of your primers that you think could possibly be involved in this interaction. And then you can look at one locus versus several loci, then there was five ce o might, or you could look at one locus versus any locus and the entire genome. And then we got to high C, where it just measures all possible contacts, and you don't have to be focused on one specific spot. And so I think that's where the name comes from, rather than an abbreviation for some, okay, okay. Wow. And three, C, I can tell you three, C stands for chromatin, conformation capture. So there were three C's to get the name. And then people just kept building and building and building and they're actually a lot of like, really quirky names. Like, there's some there's like one software package called cooler, because it's supposed to be a cooler to keep all of your high C in if it is a drink. Yes. juice box is another one. You can do strokes. Companies use? Yeah, these are all the juice box strong juicer is another one. These are all high C tools to measure IC data.

Dr. Barkha Yadav-Samudrala: 12:49

Wow, that's funny. So let's talk about the steps involved here. I know you kind of gave an overview. But for example, you are in a lab, you're setting up a new experiment, how would you begin? And just like an overview, what are the steps you would perform? And in like approximation, how long does an experiment take?

Alexis Stutzman: 13:16

Yeah, that's a really good question. So a single experiment from, you know, culturing the cells all the way to having the data in your hand, or I guess, giving the data to giving a tube of DNA to the sequencing core, maybe that's a better measurement. That actually can vary a lot. I think, on my fastest days, I got this to work in two to three days. But generally, it does take a very long time. So when I do these experiments, I actually adapted the protocol to work in Drosophila tissues. So my primary, my primary starting point, actually is after a lot of work, because I've had to do a bunch of fly genetics already to get my genotype of interest, culture all of those flies, get them to the right developmental time point, I guess I study larvae or maggots, instead of national fly but you so that, that that part can take days. Or if you know, you're in cell culture, it's just a matter of growing up your cells, getting them to your count. And, you know, the cell count is about like, one to 5 million cells per experiment or per replicant, depending on you know, how, how big your genome is. So, you know, it can it can take a while to get your starting point. And so that kind of varies for every situation in person, but the actual protocol itself, I think, takes about three days. So I break it up where I do the actual high C reaction itself on day one And then on day two, I do some library prepping. And then on the third day, it's like some cleanup of that library prep. So it's like two and a half days, I guess. So the actual steps that are involved, you start first by preparing your tissue and every in any way that it needs to be prepared, and that preparation does involve fixation. So you want to have a pretty nicely fixed, so you would treat with formaldehyde, as well, I do. There have been some protocols where they switch out formaldehyde for like DHS or something, which get is a longer crosslinker. So you may get DNA contacts that were further apart from each other in 3d space. And that can, you know, have lots of implications for how you interpret things later. But in general, people start with formaldehyde you fix for like, 20 minutes for my tissue, and then you quench that fixation, and now you're ready to move on to the high C experiment. Step one is some kind of enzymatic digestion. So there is high C, and then there's it's very similar brotherly experiment called micro C. And they only differ in the enzyme that actually goes into the first step. So in high C, you use restriction enzymes for my protocol, I use a kit actually, and it says enzyme A one and enzyme A two. I know that there are two restriction enzymes, and I know what the sequences they recognize. So I can figure out, you know, at least sort of what they might be. But yeah, you so you, you break up the genome with some kind of enzymes. So for high seeds, restriction enzymes, and for Microsoft is micrococcal nuclease, a m&s and so you just fragment your genome, and it doesn't require a DNA sequence in order to make that cut. And so Microsoft has done a lot to actually improve resolution of your high C matrix in the end, the next step is just treating. So first you make the cuts, then you need to treat the ends to make sure that they can be biotinylated. And then you add your biotinylated adapters. So I think the way it works in my protocol, I don't know this for certain because it's proprietary, it's commercial, and proprietary. But I believe the way it works is there's like one or two nucleotides that have a stripped out or have a biotin added to that couple of nucleotides. And so I'm actually like gating and biotin, deleting all in the same single step. So I do the ligation, or I do the biotinylation. And then I do the ligation. And now I have a whole bunch of biotinylated circles, where the actual DNA to make that circle may have come from too far away, low sigh. And that's about the end of day one. There's some QC stuff that's involved there, to make sure that my biotin experiment worked. And then I can actually have a high recovery. So that's one point where you can say, I'm not going to continue, I'm just going to start over. And that has happened to me plenty of times. So that's right. So that's sort of step one, or day one. And then day two, is, is then taking that biotin in that library of your biotin, elated circles. And you you cut all of them up, so that you can put your experiment into a sequencing experiment. So you cut up all of your circles, you do some size selection. There's also a biotin enrichment step in there, where you actually use the streptavidin beads to pull out your biotinylated junctions. And then yeah, then after that, it's a very basic library prep like you would do for any other genomics experiment. So you fix the owns you add on your Illumina adapters, and then you basically clean up the library, maybe do a size selection, I also do a tape station run so I can see the distribution of my fragment sizes, and make sure that it matches all of my size selections. And then after that, at the end of that after I have I guess partway through library prep. So right before I'm about to amplify my library, I do a quick qPCR experiment actually, to predict the number of amplification cycles I need in my library prep. That way I don't over amplify and have a bunch of PCR duplicates that just get tossed and I wasted all the sequencing data. And then yeah, then I do my amplification Question with those predicted number of cycles, another cleanup and then I submit for sequencing. So, yeah, I think in general that that can take two days, but I usually break it up into three. Okay, especially because getting my tissue is pretty laborious for me I'm bisecting from larvae. So it takes a while.

Dr. Barkha Yadav-Samudrala: 20:22

So you can get you your cell lines for these experiments.

Alexis Stutzman: 20:26

Yes, you definitely can. I think in my so I'm jointly mentored by two labs, I'm the only person that I know in my building, who is doing Hi, see from tissue like this, most people would use cell lines for this type of experiment. And the preparation is pretty easy for my lab mates who do that. You just get your, you know, appropriate cell number counted up, spin down your cells treat with formaldehyde, pellet them again, quench you also have to quench that formaldehyde. But it's really straightforward. And I actually think this kit was designed for cell lines,

Dr. Barkha Yadav-Samudrala: 21:05

okay. Okay. And these kits are like you just buy the kits, you don't have to make any reagents and thinks

Alexis Stutzman: 21:13

I'll come No, I'm not not for this kit that I'm using. There are certainly people who prefer to use their own reagents. But I personally, there's so much going on that it's just so much easier if I can just order it and not have to think about it anymore.

Dr. Barkha Yadav-Samudrala: 21:29

Exactly. Yeah. Well, so thank you for that great overview and explanation. So let me ask you, are there any like I know you mentioned two C threes, no, three C, four, C and five, C and Mike? Yes. So are there any alternative techniques that would give you same results as high C?

Alexis Stutzman: 21:54

Yes. So right, I mentioned, as you said, the other C techniques, and micro C, but the big problem with all of these is that they're all focused on proximity based litigation. So that means you're just cutting up the genome hoping that all of your stuff is fixed enough to stay in its place. So you can't you know, you can't disrupt the nucleus at all, which can be I mean, that's fine. But you can't disrupt the nucleus at all. And the bigger problem is that the actual readout that you get assumes that only two DNA loci are involved in that DNA DNA contact. So in the nucleus, there's no reason to expect that only an enhancer and a promoter would touch each other, it's possible that they have a whole bunch of they're a whole bunch of other regulatory DNase that are involved in a single regulatory event. And so the assumption that it's only region A and region B, and you exclude region C, because in a if you are like gating, and depending on that ligation, you can only have two fragments involved in that. So there have been other techniques, too, that have been developed to try and address this. And the best one, and I guess the only one that's coming to mind, but it's pretty fantastic. It's called Sprite. So all these names, oh, they all have these like dream names. It's pretty great. So right, so sprite overcomes this by not depending on like proximity based ligation whatsoever. So the whole concept is you still frag or you still are fixing your genome. So in sprite, you actually use a stronger crosslinker, you double crosslink than you would in a high C experiment. So you're making sure everything is really rigid. And then you Sonic eight, your whole genome. So you don't need any restriction enzymes at all. You just fragment the whole thing. You want to make sure that you're keeping the proteins in your sample. So you don't do any kind of like RNA. Well, I guess you can do an RNA degradation depending on your context, but you don't really degrade RNA, you don't really degrade the proteins. You just let your whole chromatin say how it is. You fragment that chromatin and then you get a 96 well plate and you pipette your your fragment and chromatin into all of the 96 wells on your 96 well plate. And then you add barcodes to each well in that 96 well plate and the paper that outlined sprite described how you can do this with like only six barcodes or something like that. They tried to minimize the cost of the barcoding. So you add all these barcodes to these wells, and then you pull everything again. So you take everything from that 96 well plate, put it into a single tube. Now you take that single tube and you divide it into a 90 Next, we'll play it again. And you add barcodes again. And then you pool. And after several cycles of this splitting and pooling is what they call it. Any DNA fragments that were traveling together in those 96? Well plates, you can assume are going to have a the same barcode and B must have been together in three dimensional space in order to be traveling together. Wow. So the the analysis for sprite is a lot more complicated. The visualization for sprite is not perfect. I've seen, I'm seeing things similar to like the high C matrix, where they're basically still telling us these two loci interact, but you don't know about the last locus. But but they demonstrated really clearly in this paper that, you know, these, these contacts that we think only happen between two loci can happen for you know, they they measured the nucleus and nucleolus. And they saw that there were n of 1000 contacts there at low side that were engaged in a single contact. So you know, the picture that we're getting from high C is far from complete, something like Sprite is a lot is a lot more powerful to actually detect those many contacts. And the other thing you can do with sprite is, if you don't add a RNA an RNA is you can actually measure DNA RNA interactions by also barcoding barcoding the RNAs too. And so there's a lot of potential and power that's, that's sort of embedded in sprite. And I think that's probably where people are going to turn to in the field pretty soon. But I think the analysis is still a little bit complicated. So we're not quite there yet.

Dr. Barkha Yadav-Samudrala: 26:44

Correct. So the sprite is, is it readily available right now?

Alexis Stutzman: 26:49

It is readily available, although there's not a kit to do so. Or at least I haven't seen one. But I wouldn't be surprised if that's something that people are working on.

Dr. Barkha Yadav-Samudrala: 26:58

Right. Okay. And it's Hi see user friendly, or does it require a long training? I'm assuming it does

Alexis Stutzman: 27:09

it. So I would say that, so. So I did high C for the first time in a tissue, so I had a lot of optimization to do. And that was just right away. During my first year of graduate school, I did that during my rotation. So there were certainly some learning experiences along the way. But I think, you know, a first year grad student could reasonably figure out how to do the Hi, C protocol. Okay, I do not think a first year graduate student, particularly if you haven't had computational experience before, is equipped to handle the analysis of high C, that is the part that I think does require expertise, it is possible to develop that expertise. And there are plenty of analysis tools out there to help people who are not experts, such as me, I'm totally self taught on a computer. But it took me a good two to three years just to be able to, like, make sense of the data I was looking at. And another year on top of that, to have numbers to back up what I was looking at. So so the actual computational side, I think that does take an expert, but the analysis side or the actual generating of your Heisey data set that's following a kit. And that's something that I think people can do pretty reasonably.

Dr. Barkha Yadav-Samudrala: 28:32

Okay. So let's talk about advantages and disadvantages of high C.

Alexis Stutzman: 28:41

Yeah, so I think the biggest disadvantage for any sequencing any genomics experiment really is the cost of DNA sequencing, it's quite expensive to do any high C experiment in the first place. The kit itself is actually pretty expensive, and you only get eight reactions out of a single kit. So, you know, you're working with very precious materials when you're doing KYC. So I think that's, that's probably the biggest disadvantage is the cost. Advantages. I mean, you know, if we compare high C to its predecessors, these three C, four C, five C type of experiments, high C is a lot, you're getting a lot more bang for your buck out of a single experiment, because you're not restricted to just looking at one particular locus. There are other experiments I can think of, and like ways to build on high seas. So there's another very similar experiment called Hi chip, which basically combines chromatin immunoprecipitation and high C. So if you're only interested in DNA contacts that involve a protein of interest, you can basically do your Heisey experiment, but add a step where you are immune precipitating like from the beginning you immunoprecipitate your protein And, and then you do all the biotinylation and ligation and all that stuff. So you can see just the contacts that are that are mediated by your protein of interest. This is actually one way that you can get around the cost disadvantage. By restricting your view of the genome, just two things that are mediated by a particular protein. So I guess that would be Yeah, I guess that would be one way around it. And then yeah, advantages, like I said, you, you can see the whole genome at one time, and that actually is become extremely powerful.

Dr. Barkha Yadav-Samudrala: 30:34

Right, yeah. My next question was cost. And you mentioned the kids are expensive, and they only give you eight reactions. So yes, that's expensive. But apart from the kids, what are the other costs involved here?

Alexis Stutzman: 30:51

Yeah, so I mean, the kit is going to give you your home genomics experiment, basically, there are a couple of reagents that you need outside of the kit like crosslinker, you would have to get yourself any tissue or cells that you're wanting to get those can, you know, obviously have costs associated with getting there if you especially if you're doing DNA editing to your cell types. And then after that the big costs is sequencing. And that really is even more expensive than the experiment. Sometimes it can be 1000s of dollars, just for one lane. And then typically Heisey, you really want a lot of reads. So I'm in the Drosophila genome, which is, it's something like 20 times smaller than the mouse genome, I think it's like 16 times smaller than the mouse genome. So we only have four chromosomes, it means it's very easy to sequence deeply. But I still need half a lane on a high seek 4000 In order to get that read depth. So one, single replicas that I generally try to shoot for 150 million reads for that one replicate. So if you multiply that across a experiment, where you have two conditions, you're doing this in triplicate, you're quickly getting, let's six times 15. That's six times 150 million. That's a lot of reads. And that that is extremely expensive.

Dr. Barkha Yadav-Samudrala: 32:16

Right? Wow. Okay, and let's come to my favorite question. Are there any fun facts about high C? I love this question, because you generally like people talk about, so things that are like not generally known by public. So is there anything about high C?

Alexis Stutzman: 32:39

So high C is pretty new, I think it hasn't had its opportunity to get this kind of thing. I mean, hi, I think the first ice experiment was in like 2009 2014. Those are like the two Hallmark early papers. So we haven't had the opportunity, I think, to get much fun facts, but I can think of a fun application. So when you're doing a high C experiment, you're basically measuring the entire genome at a single time. And I have seen experiment or papers, where they have done a high C experiment, in order to assemble the genome of a new animal. So in this one paper I was reading, I think it was a mosquito strain or something. And they basically demonstrated, like, they did the, the DNA sequencing to like, you know, the classic, like assemble the genome with just long read sequencing. And then they did the high C experiment. And basically, because contexts that are too low sigh that are close together are more likely to be ligated. Together. So that data is already in your high C experiment. And so they came up with this paper I was reading came up with this package to basically make a de novo genome assembly all completely on your own, just from your high C library. And so this allows you to be able to measure two things about an organism at the same time. One is what that ACTG sequences in the first place. The second is what their chromosomes are looking for their chromosomes look like in three dimensional space. And that's pretty cool. If you can get out of a single experiment.

Dr. Barkha Yadav-Samudrala: 34:17

Wow. Yeah, that is a fun fact. Wow. And my last question for you would be could you suggest some, like your favorite are any papers that you really like and enjoy reading so that I can link that down in the description for our listeners?

Alexis Stutzman: 34:38

Absolutely. So I think I would be remiss if I did not plug my own work. So yes. The first is a review article that I wrote with my PI's that are about basically everything we've talked about in this conversation is called advancements in mapping 3d genome architecture. And it goes through a sort of like mentions, this is where Heisey came from with these other A three C 14 C five c experiments. And then here are three or two other assays that do the same thing as high C, one of them I talked about already is sprite and so we get really into the details of Sprite. And the second one is track looping, where instead of relying on fixed genome restriction enzymes, all of that, they instead use a TN five transpose ace to measure their three dimensional contacts. And that's pretty, that's a pretty fun assay as well. So all of these details, the comparisons between the assays, and how they function is in our review, it's called advancements in mapping 3d genome architecture. And that came out in methods a couple of years ago. And then an article that gives a lot of really cool science that I think is pretty fun, is called chromatin architecture emerges during psychotic genome activation independent of transcription. So basically, there's been this question of like, do we even care about the structure of chromosomes in 3d space? Or is that like, not at all relevant to how transcription happens? And so in this paper, they look at early Drosophila development. So in fruit flies, they actually do not express any genome, the genome is not expressed by a given animal until a certain time in development. So your genome is totally silent. There's no transcription happening. Instead, it depends on maternal deposition, like these maternally deposited things in order to survive for the first couple hours. So what this group did was they added a transcription inhibitor. And they asked, if you don't have transcription happening, do you still get the DNA structures that come out? And you see an icy experiment? And the answer is, yes, the genome architecture is completely independent of transcription in the early Drosophila embryo. And what transcription was needed for is actually stabilizing those contacts over time. So even though you can make the contacts you can't change the contacts, and you can't keep them without transcription happening. And so there's this really cool interplay between like chromosome structure and transcription that we're still not totally sure about. But this paper is the first one that I feel like really answered, No, transcription is not the reason that we have these these, you know, big G DNA structures.

Dr. Barkha Yadav-Samudrala: 37:27

Wow. Thank you so much, Alexis. That was a fun one. Early enjoyed our conversation. And this has taught me something totally out of my norm. So thank you so much for that.

Alexis Stutzman: 37:42

I'm so glad and thank you for having me. This was such a great opportunity to geek out and this truly is my favorite experiment.

Dr. Barkha Yadav-Samudrala: 37:49

Oh, wow. Thank you so much, Alexis, and listeners, I will catch you next week on another episode of the podcast and in meanwhile, if you have any suggestions, or if you would like to come on the podcast and discuss a technique with me, please email me at Barkha at Nerd RX podcast.com. And remember, it's good to be a nerd by