Jul 08 2021
DNA as an Eyewitness: Phenotyping Applications for Forensic DNA
Like many teenage girls in the 1990s, Dr. Susan Walsh, Associate Professor at IUPUI, was fascinated by Agent Dana Scully’s work as a forensic pathologist on the television series, The X-Files. Recognizing that there was much to be gained in the field of forensics from DNA phenotyping, she wanted to research methodologies that would enable more information to be gained from the DNA taken from crime scene samples.
Dr. Walsh’s earlier work was focused on using SNPs to enhance understanding of the genetics behind determination of human physical appearance and ancestry. This information can be used in a forensic casework context to provide key intelligence information such as eye, hair or skin color to law enforcement. While this information is not comprehensive, it can be used to give direction to law enforcement as an investigative lead in an otherwise tough missing persons case or in a mass disaster situation.
In this interview, we discuss what phenotyping can accurately predict at this point as well as where the field is headed and how forensic laboratories can overcome limitations for using the technology. We also address privacy concerns surrounding the use of DNA phenotyping, and advice she has for those just starting out.
Interested in learning more about DNA phenotyping? Check out our YouTube playlist.
Transcript:
Ann: Today, I am with Susan Walsh. She is a one of the experts in the evolving field of forensic DNA phenotyping, and I’m grateful to have a chance to talk with you about your work; where you feel it’s headed, the challenges associated with it, and so forth. So thank you for taking the time out. I would have loved to have done this in person at ISHI this year, but since that’s not happening, this will have to be the next best thing. Thank God for platforms like Zoom that enable us to at least be able to see each other on camera. It’s not the same as being in the same room, but it’s certainly better than a regular old telephone call, so thanks so much.
So, not everybody is familiar with your work. So, can you tell us a little bit about yourself and how you got into forensic DNA research.
Susan: Sure, I think everybody is intrigued with forensics, mostly because of TV. It started a long time ago that you had some shows like the X-Files and CSI, and I think that kind of sparked an interest in kids growing up. I like science, but I wanted to do something morbid and maybe kind of sexy. I will say that I kind of fell into that realm. I was a pretty focused person that wanted to be in law enforcement when I was younger, and then I got the bug and I enjoyed science (especially biology) and understanding how we worked and functioned. The two merged, and then around that same time when I was in high school, the X-Files was a big show, and I loved the idea of forensic pathology and it basically just morphed into understanding that forensics was something that I wanted to get into. When I decided to go to college, my biology teacher said, “Get yourself a broad degree. Don’t ever limit yourself at the start. Get a broad biology degree and if you still feel that need that this is where you want to go in life, then go that way, but make sure you cover all bases.” I think that’s probably great, great advice and something I’d tell anyone in scientific research. You don’t know until you actually touch upon all of these areas, so you can think on something and want to do it until you actually get in there, and then you’re like, actually, uh oh, I don’t like it and then you’re stuck. So, it’s much better to get yourself a broad degree at the start and then focus. So, that’s what I did.
I did my biochemistry degree in Ireland and then I opted to do a masters in forensics. So it was a forensic profiling masters from the University of Central Lancashire in the UK. It was my first jump out of my home county, and it wasn’t too far over to England, and I just loved it. I loved the area. I loved the lecturers there, they were great and pushed my ideas and research angles and I enjoyed that masters a lot. It was a one year accelerated masters, and it really went well.
Then I took a break and did some travel and kind of worked with the University of Sydney and worked in the sequencing facility, and kind of realized that I wanted to do more. I searched the internet to see if anybody was doing research in phenotyping. So, it was an unusual thing. When I was doing my masters, I came across several papers that were just starting to look at eye color. It was really early days, and I don’t even think Manfred had started to look at eye color yet, so it was really early days. I remember presenting a paper going, “Wow! Could you imagine if we could predict what someone looks like?” and that was back in 2005-2006. Then, when it came time for me to do my PhD, I kind of thought long and hard (and that’s what I say to students as well – don’t just jump into a PhD because it’s 4-5 years of your life). So you need to pull back and make sure the topic is really something that you want to do. So, I literally just Googled “appearance forensics DNA” and up popped Manfred, and the rest is history really.
Ann: Excellent, so you’re at IUPUI, which is in Indianapolis. So, how did you get there from England and Ireland? It’s quite the distance.
Susan: Yeah, so I went to England, Ireland, Australia, then the Netherlands, and then an opportunity came up to do pigmentation research in primates, so that’s where I did my post-doc. When I was doing my PhD, I actually met my husband there, so that’s what brought me to the United States. IUPUI was interesting. I kind of just stumbled across it again, and I’m a true believe of whatever happens for a reason, and it was a combined position with had a biology home and a forensics lectureship area and I kind of thought, “Well, that’s great”, and it was in a city, which I like being a part of, and located next to hospitals, and I thought, “Hey, let’s give it a go,” and my husband said, “Sure, let’s try Indiana,” and we’re still here.
Ann: Excellent. Well, I have a soft spot for Indianapolis. My son was born there at Methodist Hospital there 19 years ago. We lived there for 6 years. I loved living there. The people are very friendly and it’s a wonderful community. There’s a lot of culture there because of the universities and Roche brings a lot of good things to the city. I love Indianapolis, so I can see why you stayed. Absolutely. It’s a good place to go and raise a family. We were really comfortable there and life takes you different places, but if something came up where we retired there, I wouldn’t be sad at all.
So, you have this whole X-Files vibe and were intrigued on the forensics side of thing, but what really drew you to the phenotyping work and how that’s evolved over the last decade or so. Can you tell me where you got your start and how your research has changed somewhat as well?
Susan: So, it kind of came about with the early days of DNA profiling and the ability to compare profiles, but at the end of the day, you weren’t really getting that much information in terms of appearance from STR profiles. You were just simply matching. Twin studies showed that twins were identical and we knew that they were genetically identical to a certain point, and it just felt right that there should be a niche in forensics that just focused on using that and, not exploiting, but researching what we know exists, we just have to find it. That’s what DNA phenotyping is; the ability to use DNA in an extra special way. Although it won’t be used in court so much, but it’ll still tell you intelligence information that can then bring up what is the gold standard in profiling.
Ann: Yeah, I think that’s a misconception that a lot of folks have. That somehow this is giving away too much information and that it’s what’s being used to convict an individual, and it really truly is an investigative lead. It’s just one piece of that investigative puzzle. It can be an important piece, particularly when there isn’t a whole lot of information available about potential suspects and so forth, but it’s just a lead. So, I’m throwing you a curveball. We didn’t talk about this question, but from the opposite side, what would you say to those folks who bring up privacy and ethical concerns? What’s your counter argument to them about the importance of your work?
Susan: At the moment, we are very good at predicting broad groups. So, we’re not actually predicting an individualized appearance. Technically, I could be predicting several different people, so I don’t feel that it’s truly invading someone’s privacy to say that the person has blue eyes. Multiple people have blue eyes. In that combination, the person has blonde hair. Ok, well, multiple people have blue eyes and blonde hair. But, in the scenario case, you start to limit down numbers. I don’t think it’s something to be fearful of, but also we will get better and we will be able to get more individualized in our prediction, but we’re not there yet. At the end of the day, we’re just trying to give a lead. If you think about techniques that are beyond DNA, so let’s say cameras and CCTV, they’re also doing that and even on a more individualized level. You can capture someone pretty much every minute if you were in the UK (they have the largest number of CCTV cameras). So, I try to squash the fear by saying you’re on camera every day. We’re just giving you an even broader estimate of what you look like so it’s not impinging on your rights at all.
Ann: Yeah, that’s a really good point. Absolutely. I’m always fascinated when I watch British dramas. They always say, “Let’s bring up the CCTV footage.” Holy cow! They have cameras every 15 feet basically, so that’s a good point. For about a decade, even more than that, SNPs and their applications in forensic research have been talked about dozens of times in dozens of excellent papers and such, but where is the focus now? How has it changed from the early days to what you guys are doing with SNPs now in your research?
Susan: I think the big boom there has been sequencing and the ability to move towards massively parallel sequencing. Before, we were wanted to understand appearance and what genes and variants are contributing, but we were always so limited by the numbers. When we were doing electrophoreses, we had to contain those numbers in the assays that we designed, and obviously, the more assays you designed, the more you’re using up those precious material (DNA from a crime scene). So, when massively parallel sequencing came, it led us to this open area where the fundamental geneticists can say, “throw everything at it.” We can combine hundreds of thousands of SNPs. So now, we have the freedom to say, “Well, although it’s only introducing a .05% prediction of accuracy, do it, because it does not impact on our assay designs, so just do it.” That’s a wonderful amount of freedom to give us when we’re designing prediction models, because we can really include interactions, minor contributors, and it really helps to push our research, because it’s a technology that’s hindered us for so long, but now it’s open, and I think you’re going to see more and more accurate prediction models made because we’re able to put in more variants. We knew they were contributing, but we just couldn’t see them.
Ann: Can you walk us through what the process is? Some of our readers are folks who are fascinated with forensics and stuff and may not be familiar with the vernacular and that sort of thing. Can you walk us through what the steps are in this workflow, equipment that you’re using, an estimate of how long it takes from start to finish, that sort of thing.
Susan: I like us to kind of slot into an existing routine or procedure. Everyone knows the STR profiling, which is the gold standard, so we like to slot ourselves into post-STR profiles. So, let’s say you have a sample from a crime scene; the technician has gathered the sample and whatever material it is, they’ve done an extraction, they’ve quanitified it, they’ve generated a DNA profile (STR profile). The results they get from that will help us decide what is the next step in terms of intelligence and phenotyping. Obviously, if there are no comparable hits or they have no suspect on file, then we come along with our intelligence methods (and that goes beyond profiling and can be used for ancestry information). The amount of DNA that’s been extracted (standard profile kits involve 750 micrograms to 1 nanogram of DNA) is tons of DNA for us when we’re working with SNPs. Because of technological jump now with MPS, you can either go down the route of capillary electrophoresis, which is the standard for DNA profiling, so we use the same PCR machines, the same CE machine to generate the result. The only difference is that we have a couple of extra clean ups of the DNA, a couple of extra single-based extension at the PCR, but at the end of the day, you’re still looking at peaks that are similar colors to a PCR machine, and the only difference are we have a T, C, A, and G for all the variants that we need. MPS is a similar routine. It’s a little bit more involved and you really have to be sure that you want to go down that line, because it is more expensive, but thankfully, with commercial companies, they are now bringing SNPs on board with their designs, so a lot of times, you can actually generate an STR profile and a SNP profile at the same time. I think that’s going to improve more and more as we go.
Ann: That’s excellent. I can see that duality; being able to use the same sample for your capillary electrophoresis and next gen sequencing really broadening the appeal of the work, because many people associate casework samples as being very precious, so they’re going to go with what they’re comfortable with and what they know, so I think this is going to really open up doors that may have not been before, because of the inability to use them concurrently. That’s terrific. So, when we’re thinking about phenotypic work and so forth, what features are we able to accurately determine now and what are the general success rates for those types of analyses?
Susan: We’re getting better and better. We started off in general with the pigmentation traits, eye, hair, skin, and there are multiple tools. These tools are basically designed in succession, so we basically start with one trait (the eye color) and then, of course, because of pigmentation, we’re going to be including similar variants with a new prediction model. Pigmentation is pretty good. Differences in accuracies down to how continuous the trait is. There are less categories with eye color than there are with hair color and skin color. A big caveat of course is that they are categories, and one day, I hope we’ll move towards continuous categories, which will be more difficult to predict, but will be a more accurate phenotype. So, pigmentation in general has about an 80% accuracy rate. Where there tends to be errors are in the cusps of the categories. It’s not so much that we don’t have the correct variants. We do in the vast majority of cases. Of course with massively parallel sequencing now we can throw everything at it, instead of limiting us with eye color to 6 variants, now we can include 300 to make sure that we can get that difficult to intermediate category, which we’ve always had trouble with. And it’s understanding more of those combinations of colors in admix populations, so that’s what we’re developing now as well. But, in general, it’s about 80%, because it can range from 70-90 with all the different categories in a recurring pigment.
There are also tools for hair structure, balding, male pattern baldness. There are many, many studies now to understand the fundamental genetics behind phenotypes. So, we’ve now know that height, though it’s been studied the longest, is the most difficult. There are about 700 variants to get at least 70% accuracy, and even at that you’re talking millimeters to centimeters different, so it’s very difficult to predict height. You can get in the extremes – very tall or very small, but at the end of the day you’re investing many markers for something that at the end of the day I don’t think we’re going to see much improvement on. So, the key no where the wins are. You’re not going to be good at everything. You’re not going to be able to understand the environmental impact on many physical appearance traits, but some there are less of an impact, like eye color, which doesn’t have much of an environmental impact such as height would and diet.
Other traits that are being developed are age; it’s slightly different in the methodology, because you’re looking at methydation, so you do actually have to treat the DNA before you begin to look at variants in the prediction models. But, it will change how we predict the traits when we know an age, because, especially for pigment, it will impact on hair (for instance). Every single trait that we add on is building up our portfolio, and I think with these additions it will certainly improve, and the variants (with the technology), we are just getting better and better.
Ann: It sounds to me that the last five years in particular, as MPS has become more widely adopted within the forensics community, that’s when those big advancements have been made, so that’s really exciting. It’s amazing to think what could happen in the next five years as well if we continue on this path. It’s been over 10 years muddling through things and then you hit this sweet spot and the work improves, so I’m sure that’s also very rewarding for you as well having been in the field for a while to see how rapidly those improvements are happening and recognizing the success rates, which (if I remember correctly) in the early days were not so terrific, but 80%, that’s nothing to sneeze at for sure. Certainly in the context of helping an investigation.
Can you kind of give an overview on how forensic DNA phenotypic results can help investigations, and is there a specific case or two where forensic DNA phenotyping played a role in helping solve an investigation?
Susan: I’ll do the latter one first. I wish I knew! We’ve helped law enforcement, and they’ve come back to us saying it’s still under investigation, but the phenotype predictions you’ve given us are great, and that’s all I really hear. So, I’d love to know more, and if anyone has information, that’d be great. That’s what pushes us as researchers and that’s what gets me up the morning. I’ve seen that I can help with this. Let’s push myself, and let’s push my students to make it better and to get beyond that 80% and to get even more realistic as opposed to just categories. So, I think that’s what I would love to see and I wish there was more casework done. That is certainly something that I try to push for. We, as researchers only love working with practitioners. I don’t think that they know that, that they can just email us or call us and say, “Hey, I have this case. What do you think?” I’ve done that a few times, and I wish more people would do it, because without practitioners, we can’t do our job. We can’t make something unless you tell us what you need, and I think that relationship should be built a little more and the trust that goes between a practitioner and a researcher should be much closer. We are not outside the realm, and that’s why I love conferences like ISHI where you go and you talk to practitioners and you say, “How do you do that in a lab right now, and why couldn’t you just put in a little extra step here?” Or, “If I could make something, how would you see it fitting into your lab or your routine and is there anything that I can do differently to make things easier for you to understand?” That is why that conference is great, and I like going, because it helps me go back to the drawing board and go, “What do we need to make so they’ll use it?” Because at the end of the day, that’s it, right? And I think that’s what pushes us along.
What was the first part? How could it help? Well, one of the pushes for me to get into this area was if there were remains, they’re kind of lost. You have a bone, and they’re not anything. They’re bones and their structures, and if you were an anatomist, then you would appreciate that, but not if you’re a human being who wants to be able to see the physical form of a person. I think that when you can give that back to skeletal remains in any possible way, whether it’s just a simple category, or a simple hair structure element, you’re trying to paint their picture. You’re trying to give them back their life, and I think that if there’s a missing persons investigation. If there’s a mass disaster investigation that requires that, phenotyping can help. It can give you the ability to say, here we are looking for this broad classification phenotype, and let’s focus in on these types of people. Let’s try to gather as much information and put this other group to the side. Not get rid of them, because I’d never say that with phenotyping. We may be inaccurate. We may be wrong. So, what we do is we rank them. We say, “This is the top phenotype. This is what I believe it is. This is what I do not believe it is from the genetics, but just don’t discount the second part, but certainly you have start.” So, I think mass disaster and missing persons that are something that are close to my heart because you give them their voice back.
When it comes to current casework, at the end of the day, it’s a time thing. You know that when you go into a case, whether it’s a murder or a crime, it’s all about time. And suddenly when you push and push and push to get that rapid DNA profile and it’s done nothing, do the same thing for a rapid phenotype profile. Go out and canvas the area, because you can actually do the same thing. You can generate a phenotype profile just as rapidly as you can generate a DNA profile. Currently, if you do this in a nice system, you can get that added information and go out and tell your police investigators, this may be the person’s profile. Don’t just go for that only, but certainly keep an eye out for this potential profile. Perhaps we can pick up some potential suspects who are in the vicinity of the crime when it took place and question them. Not arrest them, just question them. These are all areas that can help, so time is a big one. And sometimes cases have nothing. Cold cases can have nothing, and thankfully genetic genealogy has come on the scene now which could give them that lead, but still, you have hundreds of scenarios that you could take with genealogy. You have hundreds of families that you could go down and waste weeks on, because you didn’t have some of the extra information. If you simply classify them or run a quick test, now they’re looking at sequencing information and extracting it using our tools and other tools and getting even simple eye color. Now you’ve got one family with one specific eye color and let’s start with that. And that’s all that phenotyping will ever be is intelligence, whether it’s applied to STR profiles or whether it’s applied to genetic genealogy.
Ann: Yeah, you bring up a very good point. The genetic genealogy, thanks to TV, leads people to believe that this can be done relatively quickly. But there’s some significant parameters that need to be met in order to have a successful investigation. Once you get into the third cousins and fourth cousins, it’s gigantic numbers of folks to take a look at, so I love that point that you brought up about how phenotyping can really be a tool to help in those types of investigations and can guide you where right now you may look at this gigantic puzzle to solve. It may tell you which direction to look in and let you know to focus in over here. That’s terrific – thank you for that. What do you think the biggest hurdles for wider adoption of SNP analysis and phenotyping are in casework laboratories? What’s holding them back from embracing this and running with it?
Susan: Commercial kits. It needs to be in a nice commercial kit that’s validated, secure, reliable. I like to say that we make those in our assays, but we’re a research lab. So, we need the backing of commercial entities to mass produce them reliably, consistently, and then I think it’ll go. It’s kind of happening now with Illumina/Verogen, and also ThermoFisher who are starting to include assays that have phenotyping SNPs and ancestry SNPs. We’re getting there, but it’s all about making it consistent and reliable.
Ann: Yeah, I mean there’s criteria that they need. Many of the laboratories simply don’t have R&D departments to work on these, so yeah, your point is definitely valid, and I’ll be sure to share that with our R&D department at Promega as well.
If there are laboratories who are interested in having phenotyping done to help with their cold cases, who should they be reaching out to? Obviously, people in your group, but are there other organizations that they should be reaching out to and saying, “help, we could use a hand here?”
Susan: That’s the thing. You can reach me anytime – just send me an email or anyone I work with. There are individuals who are happy to help, because that’s why we do the research. But I don’t think there’s any institutes… It could be nice if there was one entity that people could ask and they could support and guide practitioners, because a lot of them don’t know what’s available. They don’t know what they can do or what they have the ability to do, because we’re ever changing. We’re developing and developing and developing, and short of reading papers and going to conferences, they’re not really up to speed on what’s possible. I find a lot of times that I would talk to practitioners and I would say, “You know you could have done this, this, and this, and it would only take .5 microliters of your sample.” And they’re like “WHAT?” Yeah, we don’t need that much. So, I think there’s a little bit of a disconnect between the researchers and the practitioners, and we need one entity that could be the buffer between the two.
Ann: Yeah, I think that would be great. I don’t think there’s a consortium or there’s not a place like UNT or NIST or some of these other institutions that are doing very specific work on phenotyping, but we’ll make sure to put some contact information in our article in the ISHI Report.
So, I know that your lab is also working on facial morphology and prediction. Can you talk through what that looks like and how that differs from the phenotypic work that you’re doing?
Susan: Facial morphology is so, so interesting. It’s something that I’ve come to work on that is an unknown area. We have so much potential in terms of prediction, but we have to stop ourselves from getting a little to eager and understand the genetics. With pigment, as I said, the groundwork for pigment prediction had been done 10-15 years previously. There had been years of fundamental research where we tried to understand what the genes are doing, how are they contributing to a pathway, how they differ between populations, and we have to do the same. We have to do all the same with respect to facial morphology, and it’s such a new area, that it’s really only been focused on in the latter 2010 and up. We have to start there, and try to combine the forensic side of research and what we want to predict with also the medical side of research. There is an overlap there, and that’s the group that I’m working with now. Thankfully, they’re genuinely interested in understanding the genetics of facial morphology. What is contributing to the structure and how many genes are in this pathway affecting bone structure? It’s so many elements that are very different to pigment, which is a singular entity. There are so many things that are responsible for the face. Our group is basically doing that. We’re trying to gather individuals so we can do these large studies. We’re trying to establish and prove that these are essential genes, and from there we will hopefully try to pick out the variants that are causal variants that are responsible for contributing to facial morphology. Another interesting aspect of the face is that it’s not flat. When we think of pigment, it can be a continuous distribution, though we’re not there yet. The face is a 3D structure. So, trying to predict not only one dimension, but trying to predict several dimensions. The work of Peter Klaus is truly revolutionary. He’s able to take his engineering research and apply it to the face. He uses a data driven approach to understand the face. He uses the face’s variation of multiple individuals to tell us what structure exists within the face. From there, we try to understand, well, what is a gene doing in this part only of the face? When you start to break down the face into several segments, it’s only then that you start to see what it’s doing in there; what kind of variants actually exist in there between individuals. There’s so many things to think of and to be careful of, but so exciting.
Ann: It sounds like something right out of a science fiction mystery. You’re absolutely right, the face is very complex. I remember sitting in Indianapolis and listening to your talk, and the smallest details change your face. The shape of your nose, the curve of your lip, all of those things. And there are predictors in all of that, so this sounds like a lifetime of work ahead. Hopefully things advance at a much faster pace, but the complexities alone make me think that this is years in the making.
Susan: It certainly is. I think we’re getting there, but at the same time, we’re encouraged by what we’re finding. We have to acknowledge that we won’t get it all. We won’t be able to predict an entire face. We cannot predict the composites that are on the side of the face. Yes, we may be able to predict the bone, but contribution from body mass index and age will alter that tissue that’s deposited on the bone. So, we have to appreciate that there’s some things that we can’t do. If we go into that with that mindset, then we’ll do great, but if we go into it thinking that we’re going to make this exact individualized face of a person and a singular person at that, that’s the incorrect way to think about this research. It’s literally going to be that we will be able to predict a nasal structure. We probably will be able to predict the depth or dimension of a forehead, perhaps of a chin area, and the lip area, but we will not predict one face, because of the variation of the contribution of environment, diet… We will actually have to predict several faces, and then it’s up to you to use the multiple faces that are predicted to give you the impression of what someone may look like. I think that is kind of the approach that we’re going for, but that’s still a long ways away. First, let’s find the genes, and then we can see what we will do next, and what we will and won’t be able to do in terms of prediction.
Ann: Fascinating! Where do you see DNA forensics in five years? Where do you see most labs? What types of research do you see going on? Where do you see the field evolving to in the next five years?
Susan: I think in terms, of a practitioner lab, that genetic genealogy is going to be pushed among a lot labs. I do think that there will be a lot of procedures and standards put in place before that happens, because I know that we’ve been a bit rushed into that area. If that gets to a place where we’ve standardized as much as we do DNA profiling itself, then I think it’s a great area. It’s really opening the doors to so many cases. It’s fascinating. The science behind it is not difficult. Yes, there are techniques and methods where family trees and documentation required, and that is an expertise in itself. In terms of the lab work, it’s not exactly difficult. You’re generating SNPs, you’re generating variants, and then you’re generating algorithms for matching potential relationships. That will definitely go down well in terms of cases, but as you said already, it can only do so much. If you only have a tiny bit of material at a crime scene, you will not be able to generate those SNP arrays. Then you may need commercial companies, such as Promega, and Illumina, and those to generate more highly sensitive sequencing kits, and that is a technology stop. In terms of research for us, I think even going beyond forensics, everyone’s interested in appearance, right? I love working in pigments not only for forensics, but also for areas of cancer research. We need to understand why there are breakdowns in pigmentation. In terms of facial morphology, diseases such as cleft-lip palate. Why are they forming and how are they forming? So, we will always bring back appearance to medical reasons and areas, so I think that we will always work in that area, and I think that appearance is so unique and interesting. I think the area of ancestry inference and microhaplotypes and just basically inferring ancestry is cool and niche and goes alongside phenotyping. There are two individualized areas. You don’t really see too much overlap in the SNP variants, but they certainly can help each other and help intelligence. I love the idea of methylation. It just fascinates me. I think I think I will incorporate methylation into my own work here. I have done some of it, but I just want to do more, because there are so many unanswered questions as to why we’re seeing so many different types of phenotypes when the genotypes don’t dictate that, and methylation is key to that, so anyone working in methylation will be a hot spot. Again, the technology needs to come with that. We’re kind of limited in our treatment of DNA before and going off to do sequencing, and fine tuning that would be an excellent area to get into for some commercial entity to help us with. There’s so much to do and so little time.
Ann: Definitely. I wanted to follow up on something that you said relating to the lab. A lot of the reasons for hesitancy for brining on MPS to a casework lab is this feeling that we need a bioinformatics guru in our lab to be able to interpret this data. Has that also changed? Has there been improvements in the software or other things that have been provided to help with this data analysis hurdle and make it a little bit shorter than this huge, daunting mountain in front of folks?
Susan: Yeah, I think now, as we’ve seen with research, we’ve seen that, yes, not everyone likes computers, right? I know when we were working on HPS and MPS and were working on those platforms for Verogen and Illumina, I kind of went into the mindset that I want my student who’s never touched a computer to click four buttons and get a result, and that’s how we’ve written our pipelines, and I know the commercial companies are going towards that route as well. As long as you can be open and tell me exactly what you’re doing when you’re pressing all of those buttons, and here’s the script, and you can see all of those things, but still make it easy, then I’m all for it. I don’t like it when bioinformatics is used in a hidden way, and it’s not shown how something is generated or run. This can be a big problem with machine learning, where it’s like a black box and you don’t really know what it’s doing, so I have a big problem with that. I don’t think we’ll move towards machine learning applications until that is more understood, and there are researchers where that’s their whole game and they try to understand more about machine learning. So, I think for now, very easy pipelines, very open source pipelines, and the ability to talk with the engineer or the commercial entity that is honest and tells what they’re doing and what they’re running and making it easy for them. I know that when my students come in, they hate computers at first, and when they leave I’m like, “You now have a skill that every job wants.” I tell high school students when they ask me what do I think is the best piece of advice, I say, “Go on the internet and look up a course on programming or language and pick the language you like and just go with it. You can’t make a mistake. It’s a computer. Just learn.” I think we’re all going to be good in bioinformatics at the end of the day.
Ann: Yeah, some of us kicking and screaming, but yes. I’m just grateful that I can operate my phone most days, so I’m right there with some of those folks. So, let’s switch gears a little bit. What’s life been like for you personally since the COVID-19 pandemic hit? What’s life been like? How has it effected how you’re getting your work done, how you’re collaborating with others, etcetera. Have there been big changes in how you’ve had to work?
Susan: Oh yeah. As a mom of two boys who are quite young – 3 and 5, like other moms that are working moms, it’s very difficult to do both. I’ve always said about my mom, who was a stay at home mom, “You have and you did the most difficult job in the world.” And I have the greatest respect for my mom and every mom at home, because I cannot do it. I’ve had several months of being at home and trying to work and I just can’t do both. My work did suffer in that time. My students have been great, and that’s my saving grace, because I can send a quick email and ask how they’re doing, and I love my students for that, because they will just continue to work from home. Thankfully, we haven’t stalled to the point of nothing, but I have not been back to the same level of research that I was before all of this began. I think, to look at the positive side of things, which is difficult sometimes in this scenario, you do start to realize that there are jumps in time when you have to optimize your time efficiently, and when I look back at my life, the biggest time was before I had kids. Before that, I could be like, “I have all the time in the world to do this.” And then when I had children, it was like, “I have one hour. I have to get this done.” And now, I think I became more relaxed, and then COVID-19 came, and I’m back to “I have to get this done in an hour!” So, I’m back to where I have to be very focused and get something done or else it’s not going to get done. So, I think maybe that’s helped for us working moms. We will really utilize that one hour to the best of our ability.
Ann: I totally agree. I think that someone who has 2 or 3 kids and is working from home should be giving TED talks on how to manage your time, because they know how to do it. I remember when my son was younger and you’re absolutely right. I’ve got 30 minutes while he’s napping before he’s going to wake up, and here’s what I need to get done. Now, he’s off to university and the pressure is very different. I totally agree and definitely sympathize with having to juggle that and recognize that it isn’t possible to do it all in the middle of a pandemic. So, hopefully those things change and evolve and can get back to some semblance of normalcy. We’ll see. What do you think the biggest challenges are for women in forensic science?
Susan: I don’t know if there are any. I think if you have a goal, and if you have drive, then this is probably the area of research where we are seen as the leaders and the workers and the people making differences. Of course, you have that time that has to go as a young researcher and over time you’ll see more female researchers in panels and editor boards. It can’t happen overnight. But I do think that forensic science over other sciences, we’re the leaders in that. We are really going to push for equal numbers and are going to have women in higher positions. I think of Nicki Prince. I think that she’s one of the leaders and has been at the top of ISFG and pushing for female researchers to go and go and go, so I like to see us as being the leaders.
Ann: Yeah, I definitely think that we are several levels above other sciences, whether it be physical sciences, mathematics, or things like that. I love going to ISHI or other conferences and meeting Technical Leaders and Lab Directors and they’re female, and that’s really inspiring and I love to see that happening, and for the most part, when I talked to some scientists last year for an article I did for the magazine, they had very supportive male mentors and supervisors who really encouraged them, and that is definitely not the case in other fields. So, I do think we’re going to lead the way and will be on more editorial boards and in large organizations leading the way.
Susan: I will say that Manfred Kayser has been a particularly great mentor in that area. Even at the last ISFG, I was astounded at the amount he thanked me for the research that we did. For me, I was just a student and was just doing my job, but he sees you as an equal and it’s a change in not only the females wanting it, but the males going, “Yeah, come on, let’s go!”
Ann: You don’t see that everywhere else. So, we will lead the way for sure. What advice would you give for a young woman who’s considering a career in DNA research?
Susan: Probably the advice that my biology teacher gave me. “Don’t focus yet. Get a broad degree.” The beautiful thing in forensics is that we see the big picture and then go, “that thing that they’re doing in mathematics right now for developing algorithms, why don’t we do that here in forensics?” So if you get that broad exposure to many disciplines in science then you can bring that with you to forensics when you decide to focus. So, a broad biology or chemistry or whatever you’re into, do it in a broad undergraduate degree and then start to get narrow in your graduate studies, and then wait until you are absolutely sure in your PhD. Too many people go, “Oh there’s a position open. It’s mine if I want it.” Well, that’s not how you should go into your PhD. You should have passion from day one and you should be the one that goes in on the weekends and say, “I don’t go in because I’m told to. I go in because I want to be the one to figure it out.”
Ann: That’s great advice for everybody. I’m going to give that advice to my 19-year-old. My last question is one of my favorite questions. If you could have one superpower, what would it be?
Susan: I think that goes back to what I just said about time management. If I could control time, and just slow it down a little bit during that hour, then the world would be my oyster.
Ann: Just think what you could get done. That’s a great superpower, I love it. If you could then market that and share that ability with a bunch of other folks, especially those with young children, they’d be eternally grateful. I loved catching up with you. Such a delight talking with you. I want to thank you for your time today, for the passion that comes through about your work. Thank you and I wish you continued success.
WOULD YOU LIKE TO SEE MORE ARTICLES LIKE THIS? SUBSCRIBE TO THE ISHI BLOG BELOW!
SUBSCRIBE NOW!