Natalie Kuldell (00:02):
Nice to see you today. Thank you for joining for this conversation. How are you doing?
Doug Densmore (00:07):
Good. No, it’s, it’s good to be able to chat with you. It’s great.
Natalie Kuldell (00:10):
I’m very grateful you’re here. You, and I’ve known each other for a little while. I remember many things that you have taught me, but maybe to begin, you can introduce yourself and where you work and what you do.
Doug Densmore (00:22):
Yeah, I’ll, I’m Doug Densmore I’m a professor in electrical and computer engineering at Boston University. But I touch on a lot of different departments. So I’m also involved in something called bioinformatics. I’m also involved in molecular cell biology and biochemistry, biomedical engineering and computer science. So I kind of bridge a lot of different fields. But it’s really about how to bring computing into biology. And I’ve been at Boston University since 2010. So I think almost it’s about 12 years. It’s been a long time but goes by quick.
Natalie Kuldell (00:53):
It sure goes by quick. Yeah. And we’ve known each other probably from the get go, but, you do such interesting work and, and that notion of bringing computers to, into biology is so relevant and so powerful. So so you have students, what kind of students do you have there?
Doug Densmore (01:14):
So I have different types of students. So some students are students that are undergraduate students. So students that are getting their four year degree at Boston University and those students taking their classes, and then they sometimes wanna do a little research or learn a little bit more. So they join my group and they’re able to do some little projects on this side. So I have some undergraduate students, but most of my students are graduate students. So these are students that are working on their master’s degree or they’re PhD. And I have about eight of those students and they are all getting degrees in different places. So some are getting a degree in electrical engineering. Some are getting a degree in biomedical engineering or bioformatics, and then I have some people who’ve already got their PhD and they wanna stay around later and they’re called post-doctoral fellows and they’re working as well. So I’ve got a diverse set of folks, people from all around world and different backgrounds working on computing and biology.
Natalie Kuldell (02:11):
Wow. How, how do you keep them all as a group working together? They’re all working on such different things and at different stages of their academic careers.
Doug Densmore (02:20):
Well, a lot of what we try to do in my group is take your idea of what you wanna make biology do, and get it all the way to actually be in the world. And so to do that, we need people at the front end, making software and computing, and then other people automating robotics and things in the lab and other people testing it. So the people all fall into different steps. Hmm. So the people that do computing are usually up front helping with the computer programs and then the people at the end that are making things actually in the laboratory are probably more biology based. So they all kind of fall in a different role in what they do. And then my job is to try to make the whole story compelling and, and something interesting.
Natalie Kuldell (03:01):
Wow. So I, I won’t ask you for your favorite project, cuz that would be unfair, but do you have an example of a project that has that whole life cycle or something that you’ve been just working on for a while that’s gone through different stages?
Doug Densmore (03:15):
Well, probably the thing that I’m most known for is probably one of my favorites and that’s of a program that was called Cello. That stands for Cell Logic. Like a Cello is a like the, the woodwind, but the, or the, whatever the string instrument, but also Cello like cell logic. And the idea is you sit down in a computer and you write a program in the same language that I would’ve written a program to make a micro process. And it looks like if, if I sent some signal in the environment or I see this toxin then make this protein. So you’re saying if then else, and that gets turned into a series of DNA pieces that then we can then send out of the computer and assemble in a variety of different ways and it actually match the program. And we showed that that worked pretty well.
Doug Densmore (04:01):
And the reason that’s my favorite project is a lot of students see that and they’re like, that’s really cool. You know, that must have taken that past year and was really busy. I’m like, no, I was working on that since 2007 and it came out in 2016. And so it was kind of the vision that, that I, that I had with another professor back when I was finishing my PhD and it took us nine years to kind of put the whole thing together and get it all the work. So it was, it was a lot of, it was a big effort, but it also was that we shared that vision. It was it’s about what, what we could do that we could start at a computer.
Natalie Kuldell (04:38):
Yeah. I mean, it sounds almost like science fiction for so, so I’m not surprised that it took so long to go from like a statement to a DNA circuit that you can then actually build and test. Right. It’s almost like science fiction, but it’s very powerful and very cool. So yeah, I can see that. That’s a good example of all the things that you’re trying to do. Well it’s literally a collaborative group and you’re a very collaborative person, which is a key to doing things that are interdisciplinary for sure. But are you a computer scientist by training? Are you a biologist?
Doug Densmore (05:16):
That’s a good question. I am an engineer. So people often have said I’ve been in scientific communities and they say, they call me a scientist and that’s not fair to scientists. I have never had a hypothesis ever. I’m not really someone who says this is a hypothesis and I go through the scientific method. I’m much more like I’m gonna build something from pieces. Like I don’t. So I’m really, I like to put things together. I like to make physical things I can touch or I can program. So I’m an engineer, I’m, I’m a computer at the end of the day. I am someone who thinks about computing. If biology didn’t compute, I probably would not be interested in it, but biology’s making decisions and sensing and communicating. So it’s a big computer, but we just don’t think of it like that. So I’m a computer engineer, an engineer.
Doug Densmore (06:06):
And for the students watching this, my goal when I was, I went to the university of Michigan for undergrad. I’m from Michigan. I’m out outside of little city called Cal Michigan. And when I was this kid, I loved Nintendo. I grew up in the eighties and I said, I wanted to work at Nintendo. And at some point when I was working at, at Michigan, I realized I didn’t really like programming as much. I liked making computer hardware. So I liked making chips. And then I ended up working in Intel and doing other things. And at some point somebody said that same hardware, we are trying to make hardware like that in DNA, we’re making these logic gates. And so I just always have liked the, the substrate, the actual mechanism by which computation happens. And now instead of Silicon it’s DNA. So I’m still someone who just tries to take those principles and put them in DNA.
Doug Densmore (07:03):
But I always like video games and, and I, it came out of that, wanting to do that, but I realized two things. I always tell students that ask,. Why didn’t you do video games? I wanted work in Nintendo. First of all, my Japanese is non-existent. So really being coming immersed in a Japanese culture based company would’ve been tough for me. And also I’m not a particularly good artist and being like doing a lot of video game work is, is almost an artistic and it’s almost as artistic as it is technical. Yeah. So making computer hardware was better for me, so,
Natalie Kuldell (07:34):
Wow. Well Nintendo’s loss is clearly biological engineers gained lucky, lucky us that you have jumped ship. What a great example for how many different parts of your brain you really do need to draw on to be, you know, in these interdisciplinary fields and, and it is a creative endeavor and you do need, you know, communication and all very, very interesting. So no regrets about not going into computer gaming and Nintendo, it sounds like. So and, and then from Michigan, you graduated with an electoral engineering degree.
Doug Densmore (08:11):
A computer engineering degree. And then it was then I, what I had done is I had been fortunate to work at Intel as an intern and what I realized. So I didn’t really like working at a large company or at least I didn’t like working at a large company that had so many junior engineers. I wanted to work for maybe a smaller company or be a senior engineer. So in order to be a designer, I wanted to be a computer architect in order to do that. I need to go get some advanced degree. And so then I had already worked an Intel in California. So I said, I’ll go to UC, I’ll go to Berkeley. I was fortunate to get into Berkeley. I said, I’ll go there. And all my friends are already out in Northern California. But when I, and then when I did that, you could either get into Berkeley grad school and computer science or electrical engineering. There was no computer engineering and computer science required something called the computer science GRE, which is extremely hard. If your background it’s, it’s very diverse. It’s got, I didn’t have the background to take it. Cause I had taken more engineering, less computer science. So I said, I’ll just be an electrical engineer I at Berkeley. And so I got, I ended up being in the electrical engineering department
Natalie Kuldell (09:20):
It funny how some of these decisions get made, right? You’re like, well, I don’t wanna take that test. So I’m gonna take, I’m gonna go in that direction instead. So I think there have been worse reasons that people have made career choices. So I think it’s perfectly sensible. So very, very interesting. So now it’s funny because I had thought, and, and one of my memories, well, I have a lot of memories of working with you, but one of memories I, I have is a speech, a talk you gave where you were talking about math and how math has rules and you can apply those rules all the time. Two plus two is four, right? And, and biology has some but not a simple rule, like two plus do is four. That you can, that you, that you’d have to remember pieces of biology that you couldn’t have a rule that you could then apply to multiple pieces of biology. I think I’ve not done justice to that idea, but I thought math played a bigger role in your background.
Doug Densmore (10:19):
Well, so it, it does more than most people. So what I say often is I’m not very good at math, but I have a PhD in electrical engineering. So, so I’ve probably taken a lot more math. And my, and my PhD was in something called formal verification. So what you would do is you made a model of the system that was a mathematical model. And then you’d say, that’s what I want the system to do. Then you’d make the system and you would check, does that system match what you said it should do? Like if I made a model of a car, I could extract what should happen when I pushed the brakes, then I’d push the brakes and say that there would really happen. Yeah. So I had math in there, but I’m not the most mathematical electrical engineer. There’s people who are more.
Doug Densmore (11:02):
So what I was talking a little bit, maybe with biology, I’m guessing was that there’s probably is. I, I imagine there’s math. Like if I wanted to simulate my body and everything, like on some level that is a bunch of math going on, but that’s extremely hard to model all of that and compute it. And it’s probably inaccurate and you’re making lots of assumptions. And so what I was likely saying is instead of trying to have this rigorous mathematical piece, I suggest often with biology is that we have these heuristics and rules for how we compose parts. And then we just make lots of them and then we can make them cuz making biology is relatively cheap. I can’t make tons of chips. Microprocessors cost more to make than the actual making biology so I can make lots of samples. So I, I kind of, what I propose often is less of a math rigorous mathematical modeling approach and more of a heuristic design build and screening method.
Natalie Kuldell (12:02):
That makes a ton of sense. It really does. And I will say that it sounds an awful lot. Like your PhD sounds an awful lot like Cello, but apply to computers, right? The idea to, figure out what you wanted to build and test that it actually is behaving the way you want it to behave.
Doug Densmore (12:20):
There’s a part of computer science when people who, the people who make computer chips are called computer architects and they sit down and say, these blocks go here. And these circuit blocks go here. Part of the circuit, it is actually doesn’t do anything. It just tests the other blocks because once you get it back, how are you gonna test it? The chip works. Right? So there are parts that test it. So it’s kind of the nerdiest part of computer architecture, cuz they’re not, it’s not exciting. It’s like I make the test blocks and I worked in something that it was called design for test. And I really liked that cuz it was a really kind of internal part. And then the design automation was how do we actually get this to physically be manufactured? So I was in a very niche area and what I liked about synthetic biology instantly when I got involved is when I made a new tool in electronics, five people cared, maybe.
Doug Densmore (13:14):
Yeah, it was a relatively small, you know, it was a very mature space. A lot of the math and algorithms were created in the sixties and seventies when computing was young. So it’s hard to make advances. It was hard to make a lot of impact. But then when I started first talking at this organization called the synthetic biology engineering research center, SynBERC, where I first would’ve met you Natalie I, I could give a talk about the idea of computers and the whole room was interested in it. And everyone was like, no one, nothing like this exists. We like your vision. Why don’t you start doing it? So I just got a lot more satisfaction out of the, the, the, the excitement people had. And I felt like this was a really early stage computing.
Natalie Kuldell (13:57):
Yes.
Doug Densmore (13:58):
I think it is. It is a fun place to be.
Natalie Kuldell (14:00):
Yep. I think there is a lot of opportunity for impact, as you say in this field and you ha you have made a ton of it. And I, I, I remember being part of SynBERC and, and, and learning so much about computers and, and just as, as a scientist, right. Learning a totally different language and a different approach. You know, and, and trying to find the junction where, where those connections could be. We have also gotten to work together through outreach and engagement. So you have some very interesting programs like stem pathways any, any vision that you wanna share about that and how it connects to your experiences or your work?
Doug Densmore (14:43):
Yeah. So, yep. So we have, as I said, I work with undergrads and in this program and I work with other students and I wanted to find a mechanism to get them involved in synthetic biology to get them involved in the research process, also create networks of students. So my vision is figuring out ways to really get people involved in synthetic biology that might not think about it. And that could be for a couple reasons. One, it might be students don’t think, oh, I’m, I’m in computing. I, I, biology’s not for me. Or I’m in the me, I’m a mechanical engineer. I’m not interested in biology. So one is exposing people to synthetic biology and really explaining that it’s not really one thing that lots of people can get involved. And then the other thing is finding students because of either their, their gender or their ethnic background or their other pieces where they may not have seen role models or other people who have that might say these things aren’t for me, or I don’t see people like myself in these situations.
Doug Densmore (15:42):
So I’ve tried to create some things like that, where we can make a big community and a family kind of organization where we can bring people together. So those are some of the things I was excited about. I grew up in a really kinda rural place in Michigan. And I was fortunate that my, that my uncle had a thermoplastics forming company. Wow. And he was really an engineering guy, but I don’t think if I had him around, I mean, he was like, I have a business and I’m an engineer. And I, he always had something remarkable to say, so I was fortunate that I bumped into him. I was fortunate that I bumped into a handful of other people. And so I’m just trying to find places where students can bump into those folks. Yeah. And maybe one or two, it only takes one or two kind of engagements where the people will be like, that’s the kind of person I want to be, or that’s the kind of thing. So I’m just trying to create that environment.
Natalie Kuldell (16:31):
I agree. I agree. Those you know, random collisions with individuals that turn out to be very influential. It’s a lot of luck and, and it’s great to be an architect of that as well. I admire it in, in your work tremendously and very feel, very lucky to be able to be working with you. So yeah. Well, thank you for sharing your path and your interests and all the very cool work you’ve doing. I’m sure there are many students who will find this absolutely fascinating and, and open the perspectives that they have on what is possible. Cause yeah, it’s an exciting time.
Doug Densmore (17:10):
Oh, great. No, thank you. And likewise, thanks for all the work you do with BioBuilder and community building and synthetic biology. So I’m looking forward to working with you in the, in the years to come.