Skip to content

Jeff W. Lichtman (Harvard University)

Subscribe using your preferred service

About this episode

#10 — From Cajal to Google, networks to solve networks! Jeff is another person who does not stop innovating and exploiting new technology. In this podcast, he talks about the first time he saw the Brainbow image, in a step-by-step manner that blew him away, through his past inspirations and today’s motivations.

New technologies come with larger and larger data sets, and he chats about how he is now working with Google to help with the analysis.

With an equal passion for teaching and research, Jeff does not slow down, although he does have fun outside of work and enjoys the simple pleasures of morning dog walks and good Italian food.

Follow Peter O’Toole, Harvard University and Harvard Brain Science on Twitter!

Learn more about the Lichtman Lab at Harvard University.

Share this to your network:

Sponsored by

Listen now

Watch Now

Please note that this is a Machine Transcription and may not be 100% accurate.

Peter O’Toole (00:00:02):
Hi I’m Peter. Hi Jeff. So today I’m joined by Jeff Lichtman from Harvard and also I believe Professor of Arts and Sciences of the Santiago Ramon y Cajal, I can’t pronounce that at all. A museum, I presume. Yes?

Jeff Lichtman (00:00:20):
No, it’s the faculty of arts and sciences. Yeah. They have a honorific title, Cajal professor. And I have that it’s, it’s impossible to have that and not feel embarrassed. Cause Cajal was of course the greatest imager in the history of biology, probably certainly in the history of neuroscience.

Peter O’Toole (00:00:48):
That’s actually, I, I, I’ve got to say for those who are not so familiar, Jeff, thank you for sending the pictures through. UI think I have a picture actually that you sent through.

Jeff Lichtman (00:01:00):
This is a really interesting picture. This is a picture of Cajal sitting in his workshop where he’s staining. He has a number of microscopes, Zeiss microscope, Zeiss achromatic objectives. And for me, the F my favorite feature of this picture is if you look by the mirror of the microscope he’s using and just look to the right of it, you’ll see a little glass of Sherry. You see that I, I can’t really point on your screen, but right, right on, right. Where you’re right next to your head. There’s a little glass of Sherry suggesting he was drinking while he was working. And, and that’s really dangerous when you look at the chemicals everywhere there, but with that one, proviso, just a truly amazing scientist in every respect.

Peter O’Toole (00:01:53):
So you obviously feel very honored for the position you’ve been given at the school of arts and sciences with his name and your work has moved into neuro science but what got you into it to start with, because we were destined to you as a child. You think it aren’t going to be a neurobiologist, or where do you start?

Jeff Lichtman (00:02:19):
I guess my, my beginnings in thinking about biology generally was thanks to my father. He was a physician and when he went to medical school back in the 1940s, he’s now deceased. Every physician in training, every medical student had their own microscope that they had to buy. So they could learn how to interpret blood smears and your analysis and the like, and when he became a regular full fledged physician he, he was hematologist studied blood, and he got himself a nice microscope. And he gave, he put his medical school Leica a microscope in which was a monocular microscope, but it had oil immersion lenses in the bedroom that my brother and I shared. So from a very early age, I mean, maybe four or five years of age, there was a fancy, really good microscope in our room.

Jeff Lichtman (00:03:26):
And we I I can’t speak for my brother, although he’s a pathologist and an immunologist, so it must have rubbed off on him as well. We just looked at everything, everything you could imagine under that microscope. So I was using a microscope at a reasonably sophisticated level when I was in grade school, you know, looking at paramecium and virtually everything, every aspect of my growing up got scrutinized at the microscopic level, everything, you know, from mucus and other bodily secretions, I’m not going to mention, but everything was looked at with that microscope. So I took it for granted, well, let’s do what,

Peter O’Toole (00:04:09):
But this has been recorded. Yeah, right?

Jeff Lichtman (00:04:12):
Yeah. Maybe this should be deleted, but he said it was, it was it took it, it just, I took it for granted that people wanted to know what things look like when they were small. I, it didn’t seem to be a special trait. But you know, as I got older and began to use microscopes as a undergraduate student in college, I realized many people were much more clumsy with microscopes than I was. And it wasn’t because I had any special facility. It was just that I had one since I was a grade age school. So I knew that whatever I did in science and I was very interested in biology probably would have a component related to microscopes. I, we didn’t use the word imaging back then. I was just looking at things. And then when I went to, undergraduate school, I did a biochemistry, thesis where there was a little bit of imaging, but not much.

Jeff Lichtman (00:05:08):
But then when I went to medical school and began my MD PhD program one of my lecturers actually in my freshmen year in medical school was Dale Purves, who was a young assistant professor at the time. He just opened his lab. And he began talking about, all the ways the nervous system works. And we got to see pictures of neurons. And I said, this is an area that seems really interesting. and he was a great teacher. And I think largely because he was so compelling as a teacher, I got into neuroscience and I never left, even after medical school, I decided just to stay as a neuroscience, researcher, as opposed to taking care of patients.

Peter O’Toole (00:05:52):
Oh, thank you for staying in it, because obviously that’s driven, actually, not just neuroscience. A lot of the things you’ve developed have rubbed off into other worlds of the biological and life science field, always around, a lot of the application side and the development of new applications using microscopy, whether that be from the Brainbow through to now the multi beam SCM. And so the multi beam sectioning, which we can come to in a little bit, but I have to ask, I have to go back that first microscope. I bet you don’t still have it do you?

Jeff Lichtman (00:06:21):
It was passed on to my brother’s children. So it is still it’s. I don’t have it, but it is still being used by young people which is a very appealing, you know, these microscopes basically can last forever, cause they’re just tubes and glass, you know? So if you take care of them there, they’re fine. I suspect it probably needs a good cleaning. I suspect you’ll need a wrench now to move the slide, the stage mover, because the grease has gotten very tight, but, but microscope still works as a one point a three and a oil immersion, 60 X lens, 63 x lens. It was, you know, almost state-of-the-art,uback then. So it was quite good.

Peter O’Toole (00:07:05):
Okay. Let’s think about the oil immersion bit, but I suppose it’s, I presume it was an upright yeah. An upright sure. Yeah,

Jeff Lichtman (00:07:11):
No fluorescence, you know, this is a monocular upright microscope.

Peter O’Toole (00:07:16):
Yeah. I’m sure we can get an led and a, a bit of colored foil and they can probably do fluorescence on it today as well. That’d be like if your brother’s children are inventive enough with it. I think from, from that. So that’s what inspires you to go to move into neurobiology cells. I was going to ask you what your first microscope was, but that’s patently obvious exactly what that was now. I’ve got to say, I, I, my first microscopes were dreadful. I really wanted to look down a microscope. I think most children are inquisitive and want to see what’s going on. But there was so poor quality after the price range that my parents could afford. And even when I went to university, the microscopes at that level was so poor. I actually hated microscopes. So I just, I couldn’t tolerate them. I could never make, see what I wanted to see. And then I realized it was just a quality that microscopes can make a big difference.

Jeff Lichtman (00:08:07):
Yeah. So I, I definitely lucked out and I don’t even know that my father had any intention here, but it turned out to have impacted both my brother and my future careers. Basically we both use imaging in our daily work.

Peter O’Toole (00:08:24):
So throughout that, so, so obviously you’ve, you’ve reached yeah. I, I’m not going to say you at the peak of your career because you never know where you’re going to end up. It can still go up, but yeah, but you’re, you’re the very highest in your area for science, but it’s never easy to get there so, well, if you go, what has been the biggest challenge that you’ve come across in all those years to get to the pinnacle of your,

Jeff Lichtman (00:08:52):
You know, I think people who, who are in science for the long game are people who really can’t do anything else. It’s it’s not that I’m making strategic decisions about what to do in order to climb a rung on a ladder to get tenure, to get a paper published in a particular journal. I don’t think about it that way. I think most scientists who are my age, who are still scientists have to have some innate pleasure that they get out of doing science. And you have to, of course be curious, you have to, if you run out of curiosity, basically, that’s the end. And I, I, I think if I have any particular gift, it’s that I’m curious, I’m really interested about things. So it’s not been hard. I wouldn’t say every step of the way has been easy. And in fact, almost everything we do seems technically hard.

Jeff Lichtman (00:09:46):
And that’s why there are, you know, a lot of things that we’ve tried to do differently, new technologies, and you’ve mentioned some that are successful or somewhat successful, but there are many other technologies we have invented that you’ve never heard of because they, they didn’t have legs. They didn’t go anywhere. They weren’t really that good of an idea. But the, the fun is in the journey. And I think if a person enjoys this kind of work, then it, it’s not really that hard to keep moving. You just have to have a thick skin for you know, what patient, occasionally a paper you think is good. It doesn’t get accepted in the journal. You think it should be accepted in, or occasionally a really great experiment that you’re hoping is going to work. It doesn’t work for technical reasons and you have to figure out why, but if you are willing to bounce back from those kinds of disappointments I th I think it is a pretty straight arrow to just staying in the game and you end up okay,

Peter O’Toole (00:10:52):
Which actually comes to I’m shooting a heading where I was going to go ask him of these things. We talked about getting some of the work published, and sometimes he doesn’t get published where you expect to get, well, hope it would get published. He knocks down and for whatever reason, which can be fairly frustrating sometimes because some of the, some of the best work can be some of the low perceived, lower ranking journals. Certainly. But what is your favorite publication that you’ve authored or co-authored, I’ll give you two, if you can’t think of it, if you can’t pick one.

Jeff Lichtman (00:11:23):
Well, I mean, it’s a no brainer for me. I think if I think about this and that is, I had a very fortunate experience as a graduate student. I worked in the lab of a very hardworking scientist Dale Purves. And he and I participated in a number of experiments that were part of his main mission in the lab. And I helped with, and while I was working on that he suggested, I look at a particular ganglion that piece of the autonomic nervous system, no one had ever looked at before. And I decided, yeah, I’ll take a look at this, guy named [Inaudiable], a very famous scientist from Europe had suggested to Dale Purves in a letter that somebody should look at this collection of nerve cells because they’re so exposed and accessible. I began looking at them and I saw something there that I thought was quite interesting, which was that they were innovated by only one axon and adults.

Jeff Lichtman (00:12:25):
And I knew that muscle fibers are also only contacted by one axon in adults. So I suggested to him that maybe I could look at whether they are contacted by multiple axons in development as is known to be the case in the muscle. That was a recent finding in muscle. And, I looked and found that that was the case. And so this was for me, the first example of a neuron undergoing a rather radical change in its wiring diagram in early postnatal life. And, I think, this was not his agenda. It was my agenda. And when I wrote up the paper, he said, why don’t you publish this by yourself?  as a graduate student in the class, I said, is that because you think this is ? He said, no. He said it, wasn’t my idea. And, you know, we’ve, you’ve been very helpful.

Jeff Lichtman (00:13:18):
You know, we’ve got six or seven papers together, so why don’t you just publish this by yourself? So I published, I actually sent it to nature at his suggestion. And over Christmas, I sent it just before the Christmas break, he called me and said they sent it right back, that it was too specialized for nature. So they were not interested. They didn’t even review it. So I, it was the first triaged paper that I knew I had, I’d never heard of a paper being sent back without review. And it happened to me. And so I was really despondent for a few minutes and he said, send it to the journal of physiology. So send it to journal physiology and got very good reviews. And I then went to a meeting and described that result. And there was a nature reporter at the meeting who then wrote a, like a news and views about this, my result.

Jeff Lichtman (00:14:10):
And when that was published in nature, I said, why was the original not interesting enough report of it? So it all ended up very nicely, but I am extremely grateful to my mentor for this, but the idea of having my thesis published and I wrote two papers, both internal physiology that were, I was so lost there as a graduate student. It’s unique. I mean, I, it’s hard for me to imagine doing, allowing that to happen to one of my graduate students working with me. So it’s even more impressive to see the generosity of my mentor back then. And so that paper has a very special place for me. That that’s a story as well.

Peter O’Toole (00:14:53):
Just, just to the rejection and then how it’s been well before, once it was out there, it was interesting. The sole authorship question, because that’s less and less the case. Now. I think, you know, I, I think there’s an acceptance that teamwork and very few, maybe a few publications come from one person doing absolutely everything from concept to maturing it and optimizing it to actually publishing it. And quite often I think it is team science now, but what’s your feelings.

Jeff Lichtman (00:15:20):
Yeah. I mean, I would say it’s, it’s not just one person. It’s rarely even the case, at least in my field, that one laboratory has the expertise to generate a paper that’s acceptable to journals that have a pretty high bar of what’s acceptable, that you need more expertise than exists in a lab. No lab is really good at everything, but most papers now require, everything they require maybe a molecular analysis and animal behavior in the same paper, but behaviorists are not molecular biologists. So you can’t do that in one lab easily. So I think this has changed a lot, this idea of kind of collaborative science. And you probably know that it’s not just that there’s one first author. Now, sometimes there are two or three first authors and the fight is who is the first, first author. And at the other end, the senior authors, not just one, but there are several senior authors and they’re also have to be put in the right order.

Jeff Lichtman (00:16:25):
And in between there’s a host of people who did small parts of these projects. So I think this is the reality and it, it makes for a kind of science that is somewhat unsettling in the sense that no one who writes or is a participant in a study like this can attest to the quality of every single thing done in that paper. Whereas when I was the sole author of a paper published in journal of physiology, the buck started and stopped with me. I, I knew everything. There was no one else to blame for anything that was wrong. Now it’s very complicated. And none of these papers are fully understood by everyone who is a co-author. And that is a weird situation.

Peter O’Toole (00:17:08):
That’s my next question. Now I don’t want to go too serious, but I think it’s quite an important point. I think team science is really important. I think that is what we’re doing. And I think that is the right way to do it. But also when you see papers rejected because they’ve missed an area out, or could you now go on to do this? Well, that’s kind of the next research project, but of asking for it to be done for that publication. I what’s your feelings on that.

Jeff Lichtman (00:17:34):
Again, I got very strong feelings about this. I think what used to be a career worth of experiments is now often jam packed and single papers. And these papers have to have 10 or 15 supplementary figures to cover all the ground necessary, to get into a journal where that’s necessary. And, and, and I think the reason for this can kind of fascinating. I think it’s a kind of sociological phenomenon that the reviewers who are just people like me and you. Yeah. We just, we always take as our benchmark of quality, how we were treated. So as soon as some reviewer asks us to do extra, then when we review papers, we assume that’s now the common level. And so there’s only one direction. Things go, they get increasingly impossible.

Peter O’Toole (00:18:29):
How do we go back to where good science gets published? And you don’t have to have answered everything. You can leave a question for someone else to answer.

Jeff Lichtman (00:18:38):
So I th you know, one possibility is that the democratization of publishing that is there are all sorts of places. One can publish that are not just, the name brand high visibility journals means that papers don’t have to be everything to everyone to still be published. Also a remarkable changes BioRxiv, or biologists, where we are now. And I think most labs I know are submitting. They’re not submitting putting their papers before they submit, and they get a lot of feedback. And those papers often that feedback can be used, in the letter to the editor of how much interest this paper in its present form, has engendered, it doesn’t guarantee anything, but it gives you some sense of confidence that the people have read it already say, this paper is good, but it would be more useful if you did this. And so they’re reviewers who really, they’re not reviewing it because they’re trying to get the paper rejected. They’re reviewing it because they’re interested in the subject. So BioRxiv, I think, has been a force for good. Definitely.

Peter O’Toole (00:19:47):
Yeah, I will. I will caveat both are feeling. I completely agree. But I would also caveat, I think there are some publications I can think of things like nature methods that don’t ask the world. It needs to be bullet proof, but ir doesn’t ask for absolutely everything. If you invented the method, I like those, the new journals that have come out to address just that concern, which is great for the technology side, but maybe when you actually solve a bigger scientific question, they’re the ones that have the, the higher barrier, maybe the high bar to get at that point. So that was your favorite publication. What are your career highlight?

Jeff Lichtman (00:20:27):
My career highlights.

Peter O’Toole (00:20:32):
So one point that to me, once in my career highlight was probably my gray hair, the side, but I disagree with them. This is why

Jeff Lichtman (00:20:39):
I think this is maybe a psychological phenomenon that it always seems to me. The next thing, the thing we’re trying to get out now is the ultimate. That’s the thing that is finally going to really do it. It’s going to make a case for something. And so I, I, don’t maybe in a few more years when I can’t do science, because my brain is melted, I’ll be looking backwards. But at the moment, my highlight is that next paper whatever it is. And right now, you know, we’re working up what you probably don’t know, we’re working on a very, very large dataset from human brain that came out of that multi-beam microscope to 2000 terabytes of data. And I’m really looking forward to that paper. And I’m thinking that will be the highlight, but you know, when Brainbow came out, that was the highlight. And when I first saw synapse elimination as a graduate student, that was the highlight. It’s all, it’s always whatever I’m doing. I think that’s the highlight,

Peter O’Toole (00:21:42):
Which is good. If you’re always looking back, that’s not good. Yeah. Was looking forward 2000, 2000 terabyte. Did you say 2000 terabytes? Yeah. So, okay. Now we’re going to talk about teamworking. Surely that’s going to be someone like Google that you’re working with.

Jeff Lichtman (00:21:59):
We’re working with Google with [inaudible] remarkable team there. And don’t ask me why Google is so generous. I think it’s partly because it’s hard and they like hard problems. But they have really, really stepped up in a dataset that requires an enormous amount of effort. And it’s it’s miraculous what they’ve done. It’s truly amazing to see all these neurons and synapses completely wired up. It’s like the golgi stain, or like Brainbow, but at four nanometer lateral resolution.

Peter O’Toole (00:22:39):
So I’ll come back to this in a moment, but just thinking about the neurons I can see on your screen behind you, that you’ve got neurons. However, that, that looks like an Apple PC. And I’m guessing that isn’t your office.

Jeff Lichtman (00:22:58):
You’re right. I don’t use Apple products mostly because many of the programs we have don’t run on the Mac operating system. This is not my office. I don’t know whose office this, is I envy this person who has an office in the corner. If you look behind me, you see it’s a corner office overlooking a river. I think maybe the Hudson river. I know I’m in a dingy little bedroom right now, but I couldn’t, but the room didn’t seem like my own office until I Photoshop some neurons. Those are actually human neurons from this project we’re doing with Google on the screen. I also put a diet Coke back there too. I I’m addicted to diet Coke. So I feel comfortable in this office. And so, so we think

Peter O’Toole (00:23:46):
Is this you off. Is this your true office?

Jeff Lichtman (00:23:48):
Yes, that’s my true office. Not this time of year, but probably yeah. In February or January year or two ago.

Peter O’Toole (00:23:55):
Of course, that does look yeah, that the yes. And you get proper snow. We’ve got to be very, very ready to get proper snow these days. And

Jeff Lichtman (00:24:05):
Yeah, we used to get proper snow. We’ll see, as the climate changes, whether we’ll continue to get it. The thing in that office is that I, as you notice, I’m facing away from the window and that’s for the obvious reason that if I was facing the other way, I’d get nothing done. It’s just really a beautiful view. So I’d face away and people who visit me, which is now never no one ever comes to my office, including me. I haven’t been in my office for six months, thanks to COVID, face out. And I face inward to a much more boring scene.

Peter O’Toole (00:24:41):
Like I said, that looks like an Apple as well.

Jeff Lichtman (00:24:45):
Yeah, that was an Apple computer. I see. There’s no Dell sign on it, but there was a time where I, I needed about two, I needed a solid state disc with two terabytes, at least. And I needed 64 gigabytes of Ram to make a laptop run powerfully. And there was no, laptop that was not, that didn’t weight 20 pounds, that would do that. That was a PC. So I, I took, h Macintosh laptop and I stripped out the operating system and turned it into a windows machine. But now I’m back to a PCs because PCs have, have the power that I needed.

Peter O’Toole (00:25:28):
And you said you had a Dell at home. I’d love to meet her down. I think her music is brilliant. Sorry, Jeff. That was really bad. Wasn’t it? Yes. Well, ah, yeah, so, so

Jeff Lichtman (00:25:47):
This is a, the most amazing animal I know maybe even including human beings that is Minnie actually Minerva is her real name. She’s a Glen of Imaal terrier and you’ve probably never heard of a Glen of Imaal. I M A A A L it’s an Irish terrier breed, pretty uncommon because they look so funny. You can already tell, just looking at that dog that its head is gigantic. It’s got a yeah, really, really big head. It actually looks a lot like an Irish wolfhound. And if you know anything about Irish, wolfhounds, they’re the size of small horses and Minnie is a Glen of Imaal is just like an Irish wolfhound, except its legs are only this big. So it’s a gigantic dog on very, very small legs. And what that means it’s really interesting. I think from a neuro-biological standpoint, it has this gigantic brain, but very little amount of body to use it for.

Jeff Lichtman (00:26:53):
So that brain is being used for other things. And I think it’s mainly not being used to plan for the future. It’s not a, I wouldn’t call it an intellectual dog in that sense, but it’s emotional valences, are very, very large. It has every possible human emotion and a bunch of emotions that I don’t think humans have. And it’s very vocal. So it’s constantly making weird noises of satisfaction, anger, boredom does a lot of sighing, you know, also, yawning, whining. It’s just constantly making noise and it’s completely taken over our household. My wife and I spend much of our food budget, keeping the dog fed because it’s very food interested and yeah, we spend a lot of time walking it and just taking care of it.

Peter O’Toole (00:27:47):
What sort of places do you go walking with Minnie

Jeff Lichtman (00:27:52):
Cambridge where we live Cambridge United States? There’s a dog park near where I live just a half a block away. So every morning for her whole life, we got her when she was nine or 10 weeks of age. By we didn’t know anything about this breed. It’s not like I’m a dog fancier or anything. So I go to this dog park and she’s gone every, almost every day, her whole life. And she’s now nine years old. So she owns that park. It’s a dog park. It’s just a regular park. It’s not a dog park, but dogs are allowed off leash in the morning between six and 9:00 AM. And so she and her friends play around used to most of her friends who were older than her are now dead. It’s a very sad thing about dogs because they, they grow up so much faster than human beings. But so now she goes to the park and chews on her squeaky ball. While the puppies run around her a lot, she’ll occasionally run, but she’s not nearly as running as she used to be where she would get covered with mud and play every day.

Peter O’Toole (00:28:57):
They said she’s she, she likes her food or very interesting food. What about yourself? What sort of food do you like? Huh?

Jeff Lichtman (00:29:05):
I love pasta, which is not good because I could eat a pound of pasta in one sitting, but I don’t anymore. Ubut that’s the only food I can actually cook reasonably. Well. Anyone who knows anything about cooking knows cooking pasta is not that hard making

Peter O’Toole (00:29:23):
Pasta. You

Jeff Lichtman (00:29:25):
Can overdo it with cooking, but I am an al dente type of guy. So I under cook the pasta slightly and I love pasta. My wife is a Italian descent. So she’s tolerates pasta but she thinks I overdo it as does everyone else who knows me and you shouldn’t eat that much pasta, but I now basically I eat almost nothing all day and then I have pasta for dinner.

Peter O’Toole (00:29:52):
So, so here’s the question. Would you rather eat in or would you rather eat out?

Jeff Lichtman (00:29:57):
Well, it’s not really much of a question right now, I guess in normal times we rarely eat out, but in normal times we used to almost every week have food delivered, take in and we haven’t even done that. It’s terrible. Cause I know these restaurants are suffering terribly, but we’re now cooking for ourselves. My wife is fortunately a truly masterful cook. But I’m, I’m not, you know, Chinese food or pizza. I’d be happy to have somebody deliver it in. But I used to get to do that only when she was too busy. She couldn’t cook.

Peter O’Toole (00:30:42):
I think pizza is probably my my favorite food. So going back to your, your role at the moment at Harvard, you obviously still do a lot of teaching as well, so it’s not just research. So what do you prefer now? Do you prefer the teaching or the research or is that an unfair question? Cause you love both.

Jeff Lichtman (00:31:03):
I think the latter is the truth. Maybe more than as good for me as a researcher, I really do love to teach. I love to explain things and it’s hard to explain things. I teach the introductory neuro course in the fall which has this year 178 students in it. And it’s it’s a hard course to teach cause I have to teach synaptic potentials, resting potential action potentials, and I don’t dumb it down. I teach them everything there is to know about equilibrium potentials the Goldman Hodgkin Katz equation and so on and so forth. All of the details of conduction of action, potentials cable properties of neurons. And I tried to teach it without mathematics, which makes it even more challenging to describe and explain these things. And this year even more challenging because I’m not teaching it live.

Jeff Lichtman (00:32:00):
The students are spread out over the whole world. And so there’s no time zone where the entire class can listen at the same time. So I’ve recorded these lectures and then we have a synchronous Friday. We had it this morning, actually where the students bring questions. They ha they get credit for bringing questions. So there were 177 questions about cable properties today and getting students to understand membrane capacitance who have never, who don’t know what a capacitor is, is a true challenge. But I love trying I use analogies like leaky hoses to try to teach people how capacitors work and how leak channels and membranes work

Peter O’Toole (00:32:42):
Actually on the it’s just purely out of interest. Do they have their cameras on when they talk to you or do they keep their cameras off?

Jeff Lichtman (00:32:51):
We, we suggest that if they can, they put their camera on. And usually when they ask a question, they turn their camera on, but there are a number of students who for one reason or another, bandwidth is one of the reasons. And another reason is maybe where they are. They would prefer not to have the camera on. The virtual background in zoom really does help. But there’s still situations where just the lighting of the room prevents a virtual background from working. So there’s no requirement that students have cameras on. There, there are sections, we have 15 sections. So the heads, the teaching fellows and teaching assistants who do the sections in those smaller rooms. I think the encouragement is stronger that the students actually show themselves so that they can have a conversation, but in a class my size, you know, with 180 plus students, what differences does it make.

Peter O’Toole (00:33:46):
Yeah. I guess they ask you a question then that then they’ll pop up. And right

Jeff Lichtman (00:33:50):
When they ask a question, they almost always show up in person.

Peter O’Toole (00:33:53):
So how many of these students have you have a citizen science project as well? I think, or have been involved in some citizen science in helping process the data through EyeWire. I think I read somewhere.

Jeff Lichtman (00:34:09):
Eyewire is mainly the brain child of Sebastian Seung who used to be at MIT and is now at Princeton, I think has just actually moved temporarily at least, or maybe forever to Samsung in Korea where he directs research. So that was his idea, sort of crowdsourcing of tracing. And that’s been very successful and we’ve participated in that a little bit, but I can’t take any credit for that. That was his idea.

Peter O’Toole (00:34:35):
If any students participated or helped or go along with it.

Jeff Lichtman (00:34:40):
You know, for us, it’s absolutely essential in order to get images like the one you’re seeing there where every in that case, it’s all the axons and a little volume. That’s what it says up in the corner. That’s a little volume where all the axons, all the dendrites, all the glial cells were labeled, to get that done. it requires a lot of human tracing and, and once you have good human tracing, you can use this as classifiers for automatic segmentation. But when that picture was taken, we didn’t have very good machine learning yet. and so you have to have ways of getting human beings to participate and it’s time-consuming, and it takes a lot of effort and we have taken advantage in my lab, not so much citizen science as undergraduate, pre-med student science, where, students help us, participate in these things, sometimes generate, a thesis for their, undergraduate thesis. And all we can do in return basically is write letters of recommendation to get help them get into, postgraduate schools like medical school, which for Harvard students, you know, this this helps, but it’s never necessary. These are all good students, pretty much

Peter O’Toole (00:35:56):
The, the image we just saw and this sort of image. So tell us if this is one of the famous Brainbow images itself. I always, I’m curious, actually, I’d be interested to see what your thought is on this as well. I think a lot of undergraduates when they’re coming through 18, 19 years of age 20, they’re seeing these sorts of images, but they’re so used to virtual reality. They’re so used to everything being computer generated, whether they realize that this is a real biological image, or whether it’s just a computer generated, you know, sort of image itself. Do you have similar thoughts or concerns? I get concerned about it. Cause I, I think they come to us, They, when it comes to their PhD, they have no idea what they can do with a microscope. They still think it’s black and white drawing under a microscope.

Jeff Lichtman (00:36:47):
One of the highlights of my career, you were asking about the highlight, but one of them was the first time Jean Livet who developed brainbow, was showing me on the confocal, a sample of brain. It could have been a hippocampus like this, I think actually was cerebral cortex, but it’s very similar where he first scanned the section with a laser light that activated the red fluorescent protein. So you got an image that was red of neurons. And then he turned on the green fluorescent protein laser light, which is blue color light. And then superimpose on top of the red were a green cells and a certain number of cells that were yellow were various shades of yellow, where there was green plus blue. And then he turned on the blue laser light and out came this third channel and suddenly this amazing, unbelievable picture with no processing, just three channels, superimposed generated this myriad of colors.

Jeff Lichtman (00:37:49):
And that was really wonderful to see. And I do, I remember that moment quite clearly because it was the first time I’d ever seen anything that looked like that. This was like quite astonishing to me. Now I think we’re a little jaded because computer science can take, for example, electron microscopic images and do exactly the same thing post hoc through machine learning. And so you get exactly the same effect, only better in the sense that we have a data set now with 42,000 cells in it. And every cell has a unique color. Basically everyone, it has its own signature. So, whereas with Brainbow, maybe you have 50 colors, you know, to have 42,000 colors. And every cell is a uniquely identified object and all the synapses are there and it’s just, it’s, it makes the highlight reel of Brainbow seem a little unimpressive actually, when it starts seeing what electron microscopy can do for you,

Peter O’Toole (00:39:00):
Maybe a bit harsh. But so the EMS that’s using the multi beam set up that’s that I’ve got to ask. I think last time you had 32 or 64 beams, is it still set at that or have you got more beams.

Jeff Lichtman (00:39:15):
We have 61 beams. There is a 91 beam out there with one more it’s a hexagonal array of beams. And so if you add one more array of beams around the outside, you get to 91 beams. And as soon as we can figure out a way to afford the extraordinary expense, such a device, we definitely want one. use that will speed our image acquisition, which is our bottleneck

Peter O’Toole (00:39:43):
Right now. It says the acquisition, that’s the bottleneck, not the analysis. It used to be,

Jeff Lichtman (00:39:49):
Oh, you know, no matter how slowly you image it, it’s going to take you years to segment it and stuff. But the segmentation, thanks to the teams like Aaron James team at Google and Sebastian Seung’s team at Princeton have figured out really, really powerful and efficient algorithms that if you give it enough training data, they can do this job quite quickly. So the stitching alignment, segmentation synopsis identification, which were almost impossible to imagine that they could be done by machine are now automated. And I would say, it’s not like it’s not bumpy. There’s still issues, but those issues are evaporating very quickly. Whereas the image acquisition is right limited by the number of machines we have because we can sort of disperse our data on many machines. Oh, if I had 20 multi beams, you know, projects that take six months, it could be done in a couple of days.

Peter O’Toole (00:40:53):
So what I think what’s scary for me is you’d have this expensive microscope. We talking millions of dollars to, to get a 91 beam system in chances are, it would only be used probably by your own group because that’s the rate limiting step is the image acquisition. So unlike many like, like the cryo electron microscopes are really a big, expensive microscopes, but they’re used by lots of researchers to look at different questions. In this case, this is one giant Nutcracker to solve one nut. In this case, obviously it can be used for many other applications actually. Um

Jeff Lichtman (00:41:29):
It’s not exactly that in the sense that we received from the Brain Initiative of the NIH, a grant it’s U 24 grant that allowed us to open our facility to other users. If there’s spare capacity, between times we’re doing big runs and we’ve already run 20 different laboratories samples through our pipeline, Mo most people’s samples are much smaller than the samples we run. So they go in and out reasonably quickly. And we’re doing, we’re continuing to do that. We have lots of samples. So I have everything from octopus to hypothalamus, spinal cord worms, you know, things that are not in my own personal research center that we’re running through these machines. Although we only have one machine. If I had more machine, we could do more good in that realm. And I’d love to be able to convince the NIH. It would be in their best interest because if you just, delivered a 91 beam machine to a lab that has no experience with it, they’re not going to get anything out of it.

Jeff Lichtman (00:42:44):
It’s just so there’s so many steps still that have it’s hard. These are very sophisticated machines and very different in that sense, perhaps then cryo EM machines, which are more like standard confocals now, you know, you need a certain amount of expertise, but it’s the tissue preparation is learnable here. Every sample needs its own staining. You have to figure out how to cut it thin. You’ve got to make for the multibeam the way we do it, at least wafers. And you gotta put the sections on wafers. There’s all these technical steps that make it challenging. So I think one way to do this is to use a facility like ours. And not as a collaborator, basically it’s the government is paying for their use of the facility.

Peter O’Toole (00:43:28):
So we spoke about this looking like art and this, I’m not sure if you’ve been seeing this, I believe this might be a picture of what, from one of your students,

Jeff Lichtman (00:43:36):
What was that? I don’t even know who that is. Who is that? Yeah.

Peter O’Toole (00:43:41):
Get away, man.

Jeff Lichtman (00:43:42):
I don’t really, I don’t see that as looking like me. I just don’t look

Jeff Lichtman (00:43:46):
Well. So Suzanne sent me this study today and said, yeah, you got to put this up. And it was one of the last minute. So it’s kinda cool, but this is art, but you also, so this is in the gallery, is that correct?

Jeff Lichtman (00:44:02):
This is at the MIT museum when Ramon y Cajal’s traveling exhibit came to Boston and it was fantastic. I, I had seen many of these virtually every one of these images in books and things, but to see the originals and see how fine he must have used a microscope to draw them. I mean, the fineness of the pen he was using and the control, every dendritic spine of a, of a Purkinje cell, there’s a very famous Purkinje cell that he drew that everybody knows. And you, you assume it must be this gigantic drawing. It’s only this big, you almost need a magnifying glass. In fact, I don’t think you can see them there, but they handed out magnifying glass. We could look at his drawings with a magnifier because he draw drew such fine lines, just fantastic

Peter O’Toole (00:44:57):
Truly, microscopic drawings.

Jeff Lichtman (00:45:00):
It was almost real size

Peter O’Toole (00:45:03):
Well with it and this image.

Jeff Lichtman (00:45:06):
So that is Daniel Berger. I’m standing there with, and this is serial electron microscopy where every synaptic vesicle in every nerve terminal is labeled butts, those white dots. And you’re seeing a whole bunch of axons making synapses on a pyramidal apical dendrite in red. So the red thing is the dendrite and right in the center, there there’s a green blob filled with vesicles, making a synapse right on a dendritic spine. And Daniel is a master. He made some of the tracing tools. We use a software called Vass that’s used widely and is also just a master at rendering. So he got enough transparency in here. So these objects look like sort of stained glass that have transparent. So you could see through the membranes it would these light colors into see the synaptic vesicles and even see through them to see the Dendrites just fantastically beautiful drawing. It’s not a drawing, a computer rendering thing. And it was in the, and, and the crazy thing is you have all these true masterpieces of Cajal. And then at the end, they throw in a couple of modern pictures like this one. And, you know, for us, that’s deeply, it made us feel really great. But, we know where Cajal stands in the pantheon where we stand, where are amateurs by his by comparison?

Peter O’Toole (00:46:36):
Yeah. I, I think that’s harsh. I think he had an open field and, and, and developed it really well, which inspired generations beyond that. Yeah. But then it’s even harder to make your name and make your career because it’s so much more competitive. So I wouldn’t tend to do yourself. So this is this your group research group or interns, or

Jeff Lichtman (00:47:00):
That’s a group of interns. Yeah. As I said, this tracing is a, a central feature of doing a lot of our work. And many of these interns were helping us with tracing the person right in the middle, Alesse par Verano. Who’s got a, a light blue shirt on is now a graduate student in the program in neuroscience at Harvard. So she had a nice landing from, she was a master’s student from Germany, I think Italian by birth. And now she’s in the program in neuroscience here. Yeah. They’re all terrific. I mean, a number of these people have really transformed what we did because while they’re tracing, they keep their minds open. They see things that we’ve never seen before and they bring them to our attention. And that has really changed the way we think about lots of this, a lot, a lot of the things we’re doing. So we’re extraordinarily grateful.

Peter O’Toole (00:47:55):
So thinking about the lab itself, can you think of a, maybe the funniest moment you’ve had at work, whether it be in the lab or conference or wherever, but work related?

Jeff Lichtman (00:48:05):
I mean, it’s yeah, I can think of something that was deeply embarrassing. But I didn’t know at the time, until in retrospect, there was a, I w I was teaching in a summer course at Marine Biological Laboratory in Woods Hole. And they were preparing for the 4th of July. They take a time off in the morning and have a float related E each, each part of the, of the community there gets involved in the 4th of July parade. 4Th of July is another big American holiday. And they said, I should dress up as GFP. And I said, Oh, that sounds funny. Since this was a Brainbow time. So I wore a sort of green green scrubs, and someone put a green shower cap on my head, and I don’t know what else was green, but I looked absolutely ridiculous and terrible. And I didn’t know, cause I wasn’t, I didn’t look in a mirror or anything and they took pictures of me and a lot of people have this picture. I hope to God, Suzanne, didn’t send you

Peter O’Toole (00:49:17):
Well, I haven’t got that picture. I’m so happy actually, Jeff, no, wait a minute. No, I’m joking, but I haven’t been,

Jeff Lichtman (00:49:28):
My heart stopped for a second. I just look awful. I mean, really, it looked like a crazy person and I look kind of serious in the picture. Like I thought I was really GFP or something terrible. So yeah, I, I think that was maybe my most embarrassing

Peter O’Toole (00:49:49):
Sounds like good fun though. Thinking of some quick questions for you, your fondest is time. Would it have been being an undergraduate PhD post doc or been a, a group leader?

Jeff Lichtman (00:50:05):
I think without question, the most exciting time in my life was when I was in a laboratory as a graduate student,

Peter O’Toole (00:50:13):
I had a feeling, you got to say that

Jeff Lichtman (00:50:14):
It was just you know, the only thing that held me back was the number of hours in a day. And that was you know, I had very few complications to work hard. Nowadays, you know, I gotta really find time to do things, but back then time was, I could use as much time as I wanted and yeah, it was productive because of that. And I just, it turned me into something that I was not, when I entered graduate school

Peter O’Toole (00:50:44):
Slapped me down. If I’m completely wrong, I’ll be guessing also that you don’t spend much time in the lab anymore. And most of your time is actually writing, getting the grants and directing and leading in many other areas and other administration parts as well. Is that correct?

Jeff Lichtman (00:51:01):
I mean, the sad truth of the matter is that many of the manually challenging steps in the kind of work I do, I don’t do anymore. And it’s not that I don’t want to. I just don’t for one reason or another, but I am, fixated on seeing the data. And, and for me, the data is the raw results. When students give me a histogram of a result, but don’t show me the data on which it’s based. I have to, I have to understand for myself what these images are, and it’s very different from the good old days when I was a student, you know, you did a lot of hunting and pecking in the microscope itself. You, you didn’t even have a scanning microscope. And we built a spinning disc confocal. That was real time. I did that just so I could watch, as I was moving the stage around to find things that were interesting.

Jeff Lichtman (00:51:58):
And there was a lot of bias based on what you’re looking for, something, and you move around until you find it. Nowadays, of course, the modern confocal microscopes and the EM microscopes, you were not doing that at all. We do this kind of shoot first and ask questions later as Josh Morgan. And I said in a paper as opposed to looking for something interesting to take a picture of and shoot. And so now we have a much more unbiased data set to look at, and I want to look at those data sets. So when students generate data, I want to see it. So I spend a lot of time looking at sharing the students, share with me now with zoom, share it on the screen. And we play through stacks of images. Let me zoom up and lab meetings. We have, we just have one person per lab meeting present, and they present for two hours. And during those two hours, I expect that they will show the raw data. And we discussed the quality of the data. And I, if I, if I don’t do that, I’m not happy. I don’t want to see just graphs. I want to see the original data before the graphs.

Peter O’Toole (00:53:08):
So you are interesting, you have the lab meeting, but I, so I meet with my PhD student once a week try to meet once a week, but I really like it when they just drop in with a new result rather than waiting. I do like that come and tell me as you go along, just, just drip, feed it and come with the excitement. So I guess my office door was open at that point rather than closed, even though there’s things going on. And sometimes you can’t see them. What about yourself? Are you a open office or a please book a time type office.

Jeff Lichtman (00:53:41):
I completely agree with you. Nothing is more pleasurable than someone interrupting to say, you’ve got to see this and they come in and show you something. Again, the reality is that because I teach and I’m on committees and et cetera, sometimes a student will come to the door and they can’t come in because someone else is in there, I’m on the phone. And that really bothers me. And I, I think that’s why I said before teaching, maybe I like it too much for my own good in the sense that it’s sapping time away from what is, I think by primary mission, Harvard would probably disagree. They say, you know, they’re paying me to be a teacher and a researcher, not in a particular order. Both of those things are equally important or perhaps nine months of teaching. And three months, summer, months of research is what they would like.

Jeff Lichtman (00:54:32):
But most of us biologists, you know, are full-time researchers. And we put the teaching in at the same time, as much as possible. And I, I love teaching, but it does make it harder for students to be very spontaneous, even though I don’t have, there’s nothing on my door. I say, in fact, I have a thanks to Suzanne, my lab manager. I have a, my door is glass, so a student can stand there and just stare at me while I’m on the phone. And then I say, I have to get off the phone because somebody is staring at me. So I can’t hide from my students.

Peter O’Toole (00:55:06):
No blind on the door. Nope.

Jeff Lichtman (00:55:09):
Well, there is a blind, but it’s only down when I have a grant that’s due or something, but now it’s irrelevant with covid.

Peter O’Toole (00:55:15):
Yeah, no, it is. So, okay. Coming back to outside of work life are you a book or TV person?

Jeff Lichtman (00:55:25):
I am a TV person. Absolutely. I, I don’t read. I read, you know, almost nothing. I read some things, but almost nothing. Yeah.

Peter O’Toole (00:55:35):
I’m with you on that. Are you a TV or film person?

Jeff Lichtman (00:55:40):
What I’ve discovered is that these long format TV shows that are where the episodes form a very long story. There are Dickens novels that are now Dickens, novels were serialized. You know, they basically came out in pieces. There’s a lot of television now like that. And I love that form. So it’s like movies, but these are movies that last 18 hours. So you really get to know the characters well that if it’s on a good story including Dickens and George Elliot and sort of classic things like that, but modern ones, we just finished 40 hours of detective in Italian with subtitles. Thank God. Which was really fascinating, really wonderful. I got to see what Sicily looks like. And you know, after 40 hours in Sicily, you really know the place.

Peter O’Toole (00:56:37):
If you like Dickens you need to come to York because we have the medieval Victorian streets still intact. It doesn’t, it does just look like Dickens as you walk through the streets. So not all of them, I don’t live in a dickens’ house, but there’s a few streets still in the main city, South for its city, self itself. When it comes to TV, what’s your vice? What’s the trashiest TV. You watch them.

Jeff Lichtman (00:57:02):
Well we just finished yet. Last night, we just finished five years of a show called Friday night lights. I don’t know if you’ve ever heard of it. So it’s about football in Texas high school football in Texas. And it’s, it’s just melodramatic. But yeah, it was good. It was, it, my advice is a things that resonate emotionally. I don’t like even though I’m a naturalist, I don’t watch too many nature shows. I don’t watch history shows as I got to be manipulated by the director and by good actors.

Peter O’Toole (00:57:50):
So he kept coming back. So we were nearly at the, at the hour, Mark, I fear I’ve got two more points to come to. The first point is on a serious note. What’s the next big challenge. What’s the next big technology developmental, an unmet need that you see [inaudible],

Jeff Lichtman (00:58:12):
I’m going to say something that I think many people would disagree with that we are leaving the age of understanding and entering the age of information. And I see understanding and information as antagonists, that when we had very little data, it was easy to have an understanding or an idea of how something works. And the biggest enemy for that kind of understanding is actual big data because it, it provides you with so much more than you ever imagined. Biology could provide, and scientists have to come to grips with the fact you can’t just verbally explain the connectome of a whole animal, a mouse or a worm. You know, we’re trying to help promote the idea of doing a whole mouse, his brains connect to them. That’s not going to be understood in any sense of the word it could be described, but it can’t be understood. And most scientists begin and end every talk with what we’re trying to understand is, and what we now understand. I think that’s going to be outdated language. So that’s, I think the biggest challenge is what do you replace understanding with when you just have big data sets? What is the alternative to understanding,

Peter O’Toole (00:59:32):
But do you not think that that will create the next part of understanding once you’ve got all that information and you’ve got it there, do you not think people delve down into that information to find new questions to solve the answers to within that dataset?

Jeff Lichtman (00:59:46):
I think the problem is that the complexity of the things we’re trying to understand, transcend the thought processes of a human brain. I think that’s the problem. It’s, it’s not that there really is understanding there it’s that the complexity of the system is beyond understanding. If I say to you, do you understand London? You say, what do you mean by that? That’s crazy. There’s just so much going on in London. What do you mean? There’s millions of people. There’s all these streets and every different kind of house. If you think that’s complicated to think of a brain, it’s way more complicated than that. You can’t understand it. You know, it’s just, it’s imposed to say, can you understand it? You can’t understand London. You’re never going to understand the human brain, but you can

Peter O’Toole (01:00:32):
End this down in parts of London,

Jeff Lichtman (01:00:35):
But connectomics and other fields like genomics, which tried to give you entire datasets Omix style datasets are not there to understand the three-dimensional structure of one molecule and how it interacts with another or how one synapse works. It’s to try to understand how you get an emergent properly property, like behavior out of this vastly complicated network. I don’t know how you get there by this reductive approach. I think you have to confront the data on the level at which it exists. It was designed to work as a network. So you have to think about it as a network, but it’s not a network that’s trivial. It’s a network with insane numbers of parts that do many different things simultaneously all over the brain. So,

Peter O’Toole (01:01:21):
Hmm. Quite literally mindblowing, I guess. And on that pun, I’ve got to ask you what is your best science joke? And if you don’t have a good science joke, what is your best joke? Come on, give it to us.

Jeff Lichtman (01:01:34):
Ah, yeah. Yeah. You know, I think the jokes on me really here in the sense that I often give lectures and people say you’re so funny. And I say, but I didn’t tell any jokes. They said it was filled with jokes. I said, I was serious. I don’t actually, I’m not a funny person by nature. I’m dead serious. But for some reason, people think I’m telling jokes all the time. I’ve never, I hardly ever told a joke in my whole life. I really I’m. Sorry. I can’t give you a good answer.

Peter O’Toole (01:02:12):
Actually be warned here because when people get cross with me, sometimes I sometimes think they’re just joking and that’s my defense mechanisms. Maybe on these people thinking maybe you were really close with them. It’s just their defense thinking. I was just joking. That’s quite funny.

Jeff Lichtman (01:02:26):
That’s what it’s

Peter O’Toole (01:02:29):
Jeff, thank you very much for taking your time with me today to talk. It was a

Jeff Lichtman (01:02:34):
Pleasure, Peter,

Peter O’Toole (01:02:37):
And I hope everyone really enjoys this in a big way, because this has been a thrill. Thanks.

Jeff Lichtman (01:02:42):
Thank you, Peter. Good luck stay well. Bye-Bye thank you.

Scroll To Top