99. Laura Luebbert: gget, hunting viruses, and questionable honeybee dances

Laura Luebbert just finished her PhD in computational biology and will soon be a postdoc with Pardis Sabeti, to hunt some viruses. We talk about how she got into biology, how she created a widely-used software project (gget) with no prior coding experience, her recent reports when she discovered questionable data in key papers about honeybee dances, and much more.

BJKS Podcast is a podcast about neuroscience, psychology, and anything vaguely related, hosted by Benjamin James Kuper-Smith.

Support the show: https://geni.us/bjks-patreon

Timestamps
0:00:00: Why Laura studied biology in Leiden/the Netherlands (and the importance of early scientific training)
0:13:41: How Laura ended up doing a PhD at Caltech with Lior Pachter (and how to choose one project if you're interested in many things)
0:22:00: gget: Developing and maintaining a software tool with no prior programming experience
0:54:07: Laura's future postdoc (with Pardis Sabeti): global virus-hunter
0:59:34: Finding and reporting questionable data in published papers about honeybee dances
1:36:43: A book or paper more people should read
1:38:55: Something Laura wishes she'd learnt sooner
1:40:38: Advice for PhD students/postdocs
1:44:02: Bonus: should I learn Catalan?

Podcast links

• Website: https://geni.us/bjks-pod
• Twitter: https://geni.us/bjks-pod-twt

Laura's links

• Website: https://geni.us/luebbert-web
• Google Scholar: https://geni.us/luebbert-scholar
• Twitter: https://geni.us/luebbert-twt

Ben's links

• Website: https://geni.us/bjks-web
• Google Scholar: https://geni.us/bjks-scholar
• Twitter: https://geni.us/bjks-twt

References and links

Episode with Jessica Polka: https://geni.us/bjks-polka
Episode with Elisabeth Bik: https://geni.us/bjks-bik
Episode with Joe Hilgard: https://geni.us/bjks-hilgard

Prototype fund Germany: https://prototypefund.de/en/
PubPeer: https://pubpeer.com/

Aaronovitch (2014-). Rivers of London series.
Frisch (1927). Aus dem Leben der Bienen.
Luebbert, Sullivan, Carilli, Hjörleifsson, Winnett, Chari & Pachter (2023). Efficient and accurate detection of viral sequences at single-cell resolution reveals putative novel viruses perturbing host gene expression. bioRxiv.
Luebbert & Pachter (2023). Efficient querying of genomic reference databases with gget. Bioinformatics.
Luebbert & Pachter (2024). The miscalibration of the honeybee odometer. arXiv.
https://liorpachter.wordpress.com/2024/07/02/the-journal-of-scientific-integrity/

Chapters

• 0:00 Why Laura studied biology in Leiden/the Netherlands (and the importance of early scientific training)
• 13:41 How Laura ended up doing a PhD at Caltech with Lior Pachter (and how to choose one project if you're interested in many things)
• 22:00 gget: Developing and maintaining a software tool with no prior programming experience
• 54:07 Laura's future postdoc (with Pardis Sabeti): global virus-hunter
• 59:34 Finding and reporting questionable data in published papers about honeybee dances
• 1:36:43 A book or paper more people should read
• 1:38:55 Something Laura wishes she'd learnt sooner
• 1:40:38 Advice for PhD students/postdocs
• 1:44:02 Bonus: should I learn Catalan?

Transcript

[This is an automated transcript that contains many errors]

Benjamin James Kuper-Smith: [00:00:00] Given your kind of two country background, I think one of the most German things you could have done is go study in the Netherlands, at least for someone who, I mean, I guess, yeah, we just, we established before we started recording that we grew up somewhat close to each other and the people, I mean, Aachen is right on the border to the Netherlands.

So I don't know, 10 percent of the people I knew studied in the Netherlands or something like that. But why, yeah, why did you study it in? Leiden or in the Netherlands in general and yeah, what kind of, what was your, when you were in school, what made you go there? Leiden

Laura Luebbert: really honest and you have to understand that I was 17 when I finished high school. And so a huge reason why I decided to go to the Netherlands is because I thought that going there to study there and live there would be just like my parents would take me and my sister to the Netherlands.

 On like short trips over the weekends. And obviously that was always a blast. So I thought that everyday life in the Netherlands was like [00:01:00] that and that my life was just going to be that running around the beach catching jellyfish with a little catcher.

So, so that's why I wanted to go to the Netherlands. Um,

Benjamin James Kuper-Smith: The beach, right?

Laura Luebbert: It's very close. So yeah. And it is stunning, so it really, in some sense was very beautiful to live there. But I mean, Leiden in particular when I always wanted to study biology, but I wanted to do working with animals biology.

I wanted to go into the jungle and go find some new animals and new plants. And Leiden is really strong. In their ecology departments and and they actually still have a lot of scientists who literally, that they work very closely together with the Natural History Museum. So they have a lot of scientists who literally go to Borneo mostly every year and go find some new [00:02:00] insects and new plants.

So I wanted to do that. Yeah, it is. Yeah. So that's why I decided to go to the Netherlands, and that's how I picked Leiden. 

Benjamin James Kuper-Smith: Okay, but I don't know, I guess you could have done the same in Cologne or something, right? Or did you also, was going to a different country also part of the appeal of Leaving home and I mean that's why I went also I mean for me there were multiple reasons why I went to England but one of them was also just like leaving not that I had like a bad time at home right doesn't like not like that kind of leaving but

Laura Luebbert: no, I understand yeah, same I wanted to go somewhere new, but I was also 17, so, being able to go to a new country, but still being two hours, yeah, that's 

Benjamin James Kuper-Smith: being able to take a direct train home yeah for 30 euros yeah whatever

Laura Luebbert: exactly.

Benjamin James Kuper-Smith: okay

Laura Luebbert: And I completely ignored the fact that Leiden, at least at the time, taught all of their STEM undergrads exclusively in Dutch.

And I didn't speak any Dutch.

Benjamin James Kuper-Smith: But that's not too difficult, right? I thought you'd just I know some people who took like an intense Dutch course [00:03:00] for a few weeks and then they it

Laura Luebbert: Yeah. I mean, that's what I thought. I definitely struggled the first year. But eventually I got the hang of it. Yeah. And it worked out.

Benjamin James Kuper-Smith: Okay. But uh, so what was the degree? Was it like a, so my sister briefly studied biology and they had all these, what does he had like to these leaf collections that you then dry and did you have to was it all that kind of stuff or learning all the insects and whatever?

Laura Luebbert: Yeah. So what, so in the first year the biology undergrad branches into what they call macro and microbiology. And so microbiology is more bacteria and genetics and macro is basically plants and animals at the plant and animal scale. And the first year is half a year of each, and then you have to decide between the two.

And so I went in thinking I'll definitely do plants and animals. And then the half year of plants and animals hit. And it was all, we had an exam where they would literally give us a bunch of dead animals, [00:04:00] mostly insects and plants, and we would just have to identify them by their Latin names. And that is when I decided that maybe I would go in the microbiology direction after all.

Benjamin James Kuper-Smith: Okay. And in hindsight, was that, do you think there was a decent decision on your part? Or was that too reactionary to the specific environment you were in there? Because I guess that does suck here. That doesn't sound like a fun time.

Laura Luebbert: I think it was, I mean, I don't know what my life would look like right now if I had not done that, if I had gone the other way. But I'm pretty happy where I am now. So, that worked out. And I do think, I mean, there's absolutely more opportunities in the microbiology, especially genetic space there's just so much I can do.

So, I think in that sense, I'm just, I'd like that I'm not limited in my work in the way that I think I would be if I had gone the other route.[00:05:00] 

Benjamin James Kuper-Smith: Yeah. Did you want to do research from the beginning or, okay. So it wasn't just I like animals or whatever it was you want to do research

Laura Luebbert: Yeah, I always wanted to do research. I was really fascinated by this idea that there would be this moment in the lab when I would be the only person knowing something that nobody else knows yet, but I just figured it out, so I know it. So I, I've always really liked that idea.

Benjamin James Kuper-Smith: So how often has that happened? Proportionate to the rest of your working hours? Yeah,

Laura Luebbert: it's been really rewarding. I don't think that moment is as specific. I don't think in the moment you, you even realize that just happened. So, 

Benjamin James Kuper-Smith: Yeah, that's a fair point. It's more like afterwards. You're like, wait a minute, these things I'm piecing together here actually amount to something. Or already amounted to something and I didn't realize it. Yeah. Yeah, I think [00:06:00] there's a lot more. I mean any job right there's more drudgery than the highlight moments.

Laura Luebbert: yeah. There's very few eureka moments in science,

Benjamin James Kuper-Smith: More pipetting than eureka. Yeah.

Laura Luebbert: Yep.

Benjamin James Kuper-Smith: although funnily enough I talked to uh, jessica Polka. She She did a PhD in I don't know something molecular biology. I mean we talked about pre prints and open science and that kind of stuff because she works in that field now, but she actually said she misses pipetting because there's a kind of nice like manual labor to it that she just doesn't have anymore Yeah,

Laura Luebbert: now. I mean, maybe I will in a couple of years.

Benjamin James Kuper-Smith: Yeah Okay, so you wanted to do research from the beginning. Did you yeah, I'm assuming you did some sort of research project for your bachelor's and then later for your master's or Okay.

Laura Luebbert: going to do microbiology, [00:07:00] I had to figure out again what I wanted to do. And mostly because of a single lecture by a professor called Egbert Lacke. He's also a neurologist at the Leiden University Hospital. And he gave a lecture on vision and how visual input is decoded in the brain.

And I was so fascinated by that. I was like, I want to do neuroscience. You'll see throughout my career, I've had a lot of these moments where I make really important decisions. Based on what you would think is not a lot of data, but it always worked out.

Benjamin James Kuper-Smith: Okay.

Laura Luebbert: In the end, it always worked out. But so basically I decided I wanted to do neuroscience.

I was still in a genetics program, but that was fine because there's a lot of need for genetics technologies in neuroscience naming optogenetics just for an example. So, the problem was [00:08:00] that Leiden university basically doesn't do any neuroscience. They do neurology at the hospital, but that's it, they really and they do psychology, but they don't do molecular neuroscience or neurobiology like I wanted to do.

But it turns out that also worked out because that forced me to do all of my internships outside of Leiden. So I did my first internship at the Max Planck Institute for Neuroscience in Florida, which was awesome because it's right on the

Benjamin James Kuper-Smith: To have a Max Planck outside of Europe. Okay. I didn't know that.

Laura Luebbert: And it's right on the beach. It is absolutely stunning.

I was there for two months and I think I did 60 scuba dives, which is basically a dive a

Benjamin James Kuper-Smith: One a day, yeah.

Laura Luebbert: Um, uh, So I had a blast and that was also my first really serious I mean, I was a first year. But I, I was in the U S. not [00:09:00] that cautious with rules around who is allowed to do what.

So I was allowed, even in the just two months that I was there as a freshman, I was allowed to work with mice. I was allowed to learn mouse surgeries. I was allowed to do all of their cloning. And so, I mean, I was just a lab tech for the postdoc, but I learned so many very valuable things. Laboratory techniques that in the Netherlands, I would not have been allowed to work with mines because I didn't have my bachelor's yet.

So, that was amazing. I learned a ton and that was my first really serious wet lab exposure.

Benjamin James Kuper-Smith: I mean, yeah, so that's also a lot of trust into someone who then, I guess, were you 18 at the time then already? Or were you still 17?

Laura Luebbert: I think I was 19 cause I

Benjamin James Kuper-Smith: okay. I was a bit, okay.

Laura Luebbert: I had to do one year of chemistry and physics. Because there's a mismatch between the Dutch and German STEM system, where in the [00:10:00] Netherlands, you have to have studied biology, chemistry, and physics to study any one of them. But in Germany, you're not allowed to do all three in your Abitur, you have to pick two.

So I didn't do chemistry, so I had to redo chemistry, and then I thought I might as well do physics as well.

Benjamin James Kuper-Smith: Okay. Yeah, but still, I mean, that's a lot of trust to put into someone,

Laura Luebbert: mhm.

Benjamin James Kuper-Smith: who's also going to leave in like three weeks. I mean, yeah.

Laura Luebbert: Yeah. And see, I mean, now that they have a bit more of an understanding of how a lab is run, the amount of resources that I was entrusted with, I am really grateful for that opportunity. So, I don't know if the professor or the postdoc are listening the professor was Hiroki Taniguchi.

If you're listening, thank you. Yeah, it really made a difference. Thanks.

Benjamin James Kuper-Smith: yeah, I mean, it's cool. I mean, I guess I also don't know, like how difficult or complex these things are. I mean, in psychology, when you like a research assistant, you basically your main job is just I mean, it's [00:11:00] like a job. Any 12 year old could do most of the time where it's just, you just take someone, put them in the room and say, here's a computer, the instructions on the screen, and then when they're finished, you pay them.

And if you're vaguely pleasant, then you're doing a good job. Basically most of the time. 

Laura Luebbert: Well, yeah, I was doing in utero electroporation, where you cut open a pregnant mouse, inject DNA into the brain of the embryos while they're still in the uterus. And then you have to put everything back and you have to this is a very delicate surgery because you have to do all of that. If you stress out the mouse too much, they will spontaneously abort.

And so. Yeah I felt so cool. I was like,

Benjamin James Kuper-Smith: but I mean, do you think that's a good system? It's obviously amazing to if you want to be a scientist, it's amazing to learn all these things, but it seems to me it's weird that, especially because you were only there for a short time they would, it's a lot of training resources, a lot of animals and that kind of stuff for someone who, it's going to leave soon [00:12:00] anyway.

I don't know. It's very surprising to me. Yeah.

Laura Luebbert: mean, I will say, I mean, they were not lax about how they handled the animals, so I did receive a lot of very intense one on one training in a short period of time. I think it helps that the Max Planck Institute is very small. So I basically, I had the animal technician for myself and I had the postdoc to myself and I had.

The microscopy facility manager to myself. And so there were just, I could talk to really the experts for every type of procedure that I wanted to do and they had time to train me one on one. So, but yeah, like I said, I am really grateful. The benefit to them was not that big compared to the benefit to me, which is huge.

Yeah.

Benjamin James Kuper-Smith: Yeah. Okay. So you did, so you, you did some surgery, you did some scuba diving,

Laura Luebbert: Yes,

Benjamin James Kuper-Smith: or,

Laura Luebbert: I got back. Yeah, I came back. Finished my bachelor's. I did my bachelor's thesis [00:13:00] at the only place in Leiden where they have anything close to neurobiology, which was the hospital. So I was analyzing some, again, but this was now in the Netherlands, so I wasn't allowed to do anything in the lab, but I was just analyzing the data, which was also great.

I learned a lot. And this was an optogenetics project on trying to understand a very specific type of migraine. Which can cause sudden death in children. So we were trying to figure out the link between the migraine and the sudden death syndrome.

Benjamin James Kuper-Smith: okay. That's pretty cool. It sounds like a pretty good, I mean, yeah, many bachelors projects are not that cool,

Laura Luebbert: Yeah, no. So, yeah, so I loved it. I was like, okay, neuroscience is my thing at the time. And I really enjoyed, going to different places to do my research. So I actually decided to stay in Leiden for my master's, even though, again, Leiden doesn't offer a neuroscience [00:14:00] master's. I again did my master's in genetics, but the reason I stayed with Leiden is because they are extremely flexible.

So I was able to basically not be there for one and a half years of my two year master's in

Benjamin James Kuper-Smith: but what, I mean, but why not just do your master somewhere else to meet? I mean, or was it because you wanted to try lots of different places or,

Laura Luebbert: I, and I really wanted to do a lot of research. And so the flexibility of the program I was able, usually the program is one year classes, one year research, but there's a way that you can shove all of your classes into half a year, which is what I did.

And then I had one and a half years where I could just do whatever research I wanted. And so that flexibility is there. What I wanted is really to have all of this room to do research and do it wherever I am.

Benjamin James Kuper-Smith: So you did some neuroscience?

Laura Luebbert: So I did some neuroscience and I did two projects. So I wrote two [00:15:00] master's thesis, both of them neuroscience projects. And one of them was at a company because I wanted to see a science environment within industry. With brain organoids also very cool. But the other one was at Caltech. In a neuroscience lab studying how antidepressant drugs work or trying to figure out basically we were building sensors for the depression, antidepressant drugs. So we could follow them as they enter our bodies as they enter our cells and see if that localization tells us something about their mechanism of action. 

Benjamin James Kuper-Smith: So one, one of the obvious questions I had was, whenever someone jumps from one place to another, it's oh, well, how did that happen? So that's how Caltech happened, 

Laura Luebbert: Yep.

Benjamin James Kuper-Smith: You liked it there and then just decided to apply to stay, basically, or?

Laura Luebbert: Yeah, pretty much. So I, my internship 10 months. And I was, [00:16:00] I wanted to do a PhD, but I was planning to do my PhD in Europe. And as my 10 month internship was going to an end, this was like a month before applications are due in the U S my professor was like, you should apply. Why not apply? You could do your PhD in the U S you could do it here.

And you have to understand that PhD applications in the U S they're not like PhD applications in Europe. I think in Europe, you write, apply to a professor directly. So you have to adhere to their requirements. 

Benjamin James Kuper-Smith: There's also

Laura Luebbert: guess it

Benjamin James Kuper-Smith: yeah it really depends,

Laura Luebbert: Yeah. Yeah. I think I mean, I think the institutes, yeah, Max Planck they have more defined application cycles, but yeah, in the U S you have to, Take the GRE which is a very painful exam.

Four hours, you're not allowed to go to the bathroom, you're not allowed to drink anything. You just sit there for four hours uh, and have to [00:17:00] take this exam. And then you have to write two or three essays about yourself and about the science you want to do, and then a very specific proposal. But I didn't really know all of that.

So I was like, yeah, I'll apply. And then I realized that I had a month to do all of that. And yeah, I don't know how well I did. I guess good enough. I got offers from Caltech and also from Johns Hopkins. 

Benjamin James Kuper-Smith: Sounds like you did something, right?

Laura Luebbert: Yeah, I did something right, but yeah, the GRE was really painful. So I think now a lot of universities in the U S do not require it anymore.

Which can be a pro and a con for some people. It's a nice way to show their skills. For me with a month of preparation, it was just painful. Um,

Benjamin James Kuper-Smith: Over quickly, right? It's not like this protracted process that you prepare for and blah blah blah, it's just you get it done with and then, I mean when it works it's nice, obviously if it

Laura Luebbert: yeah, exactly. Yeah. In this case, it worked [00:18:00] out. And Yeah, I got the offer from Caltech. Okay. Unexpectedly they flew me out to interview and that went well as well. So when I got the offer, I thought this is one of the best institutions of the world. I can't really say no. So that's how I ended up.

Benjamin James Kuper-Smith: Okay, but you still had like no, I PhD with, Lior Pachter, is that how you pronounce it?

Laura Luebbert: That's right. Yeah.

Benjamin James Kuper-Smith: Yeah so but it wasn't like clear to you that you were gonna work with him or anything like that It was just you know, you like the place and so let's stay here and see what happens that kind of thing. Oh, okay

Laura Luebbert: Yeah. I think throughout my career I mean, I've been doing a lot of hopping around, but I think it's clear now. And so I just had always had a lot of things that I'm interested in and I can get excited about pretty much any project. So when I came to Caltech, I had the same mentality.

I liked that in the US they have this rotation system where in the first year of your [00:19:00] PhD, you do two months in three different labs and then you pick one. I liked that because, yeah, I didn't have any one thing that I was really set on. on doing I had a lot of different skills that could be applied to a lot of different research areas.

So, and I found it really difficult to pick one. So that's, I went into Caltech with the mentality, something will come along and I'll know. Yeah.

Benjamin James Kuper-Smith: I think Let's maybe talk a little bit more about that because I don't think you're unique in that particular thing. I think that's, I mean, I also ended up doing like lots of different projects getting involved in research here and there and doing all sorts of stuff and yeah, I mean, what I ended up doing my PhD and was nothing I did basically before and so maybe how did you, yeah, how did you basically end up making the decision of, Given that there are, presumably, were quite a few options available to you at Caltech.

Yeah. How did you make that decision? Because [00:20:00] it's Yes, it's a it's a decision. Basically everyone has to make at some point and I think many also struggle with it

Laura Luebbert: so I do want to start. So when I first came to Caltech it was split. There were a lot of people like me like Nelson mentioned, but then there were also a lot of people who had really basically worked in the same lab, the entire undergrad and career after that. So they were really experts in like very particular fields.

And that. It was very intimidating. And they really, I mean, they knew what they wanted to do and they were already really good at it. And so I often felt like I wasn't good enough and I was like, how am I gonna compete if any lab that I consider has one of these people who literally trained for years to join their lab?

But I think that in the end, the flexibility that I had and also the skill. To very quickly [00:21:00] learn a new topic, which I had acquired by just switching around, having to do it that was super, super valuable. Because anytime there was an obstacle, I was able to adapt and work around it and find some interesting angle, even though it wasn't in the original frame of what they had pictured for the project.

And that flexibility I think was harder to obtain for students who had always been in the same lab their entire career. Um, So

Benjamin James Kuper-Smith: to just briefly do a slight know mention some of the stuff We'll talk about later is that for example, I mean the thing you then You what the, what's the G get? Well, what we're talking about a little bit is uh, you know, a software program in genetics and not in neuroscience per se.

Right. So it's as basically it's two different fields that you didn't plan on working on.

Laura Luebbert: completely pivoted again during my PhD. Yeah. And so, yeah, I can talk about that because it's also how I ended up in the Pachter lab. 

Benjamin James Kuper-Smith: Yeah. Perfect. [00:22:00] Yeah.

Laura Luebbert: for my PhD, and that's also why I chose a very technical school like Caltech is I wanted to learn programming.

I didn't know how to code. I think Leiden University does better now, but at the time, they just didn't teach their biology students how to code which is just not acceptable anymore. The data that we're generating today as modern biologists, just, you cannot plug them into Excel anymore and just make a plot in Excel.

They're way too big. So I really wanted to learn how to code and When I started out code tech, I, every class that I could find that I thought I would understand which turned out to be not a lot because even classes that said doesn't require any coding. They always required coding. But I spent a lot of time on YouTube.

I spent a lot of time on code academy. Learning to code is really painful in the beginning, but it is an exponential curve. So anybody who's struggling through [00:23:00] that right now, just. It's very frustrating in the beginning, but then it becomes really fun. So I struggled through that and I found that I really enjoyed coding. And then the pandemic hit, I started my PhD in 2019. So I really only had half a year and then we were locked in at home. So the pandemic hit.

Benjamin James Kuper-Smith: exact same thing. Yeah, exactly.

Laura Luebbert: exactly.

I just couldn't go into the lab. And so I think this is where a lot of students really struggle understandably. I mean, it was a very difficult time to do your PhD. But I decided, well, I just want to code. And now I'm stuck at home anyway, so I'm just gonna do that. And at the time I was working with another student who was generating a single cell on a sequencing data set.

So, very complex, very high dimensional [00:24:00] RNA sequencing of thousands of cells. So you absolutely need coding skills to work with the data type like that. And I was the only person in that lab who knew how to code. So she came to me,

Benjamin James Kuper-Smith: Wait, but how I thought everything was so technical and required coding, how were you the only person who knew how to do it?

Laura Luebbert: it's very technical in terms of classes. So all of the Caltech undergrads know how to code, but the PhD students and postdocs are very international. So. And so a lot of them come from university or universities, especially in the biology labs who are still a little bit behind in teaching,

Benjamin James Kuper-Smith: It's funny that the undergraduates are, in some ways at least, more skilled. The

Laura Luebbert: in coding, I mean, right. Then they lack obviously

Benjamin James Kuper-Smith: That's all other stuff, but just for coding, 

Laura Luebbert: yeah. No, I mean, Caltech does a really great job in teaching them technical skills. They are very good programmers.

Benjamin James Kuper-Smith: Yeah, so just a general question [00:25:00] about, because I'm not from molecular biology and I mean, I guess technically, if you do neuroscience Neuroimaging in humans. I guess that's a kind of biology, but I don't know, biologists probably don't like it when you say that, I could imagine. But the, I'm still a little bit surprised that it's that rare because I feel like if you do any kind of neuroscience in humans fMRI or EEG or anything like that, or even if you do most cognitive science, everyone has to code.

I mean, it's don't know if, I don't know whether I know anyone who didn't have to learn his PhD basically. I mean, I don't like it, it's not my favorite thing. But, yeah, I'm just surprised that like in biology it seems less common.

Laura Luebbert: Yeah, I think it might also be the exact subfield that I was in. Maybe. I mean, I was in a, I was trying to do wet lab neuroscience, which is very heavy in the lab, so you just don't have time to sit down and learn how to code doing a lot of cloning, a lot of cell work and then the main output is [00:26:00] microscopy images.

And so, Fiji is still a very widely used program

Benjamin James Kuper-Smith: Sorry, what is

Laura Luebbert: those images. Oh, sorry. Fiji

uh,

Benjamin James Kuper-Smith: and a, And a water brand.

Laura Luebbert: So there was there was a very popular image analysis program among biologists that generate microscopy images. And that program was called ImageJ. And since it was so popular, They kept maintaining it.

And then they put up a new version called Fiji, which stands for Fiji is just image J. And so anyway, image J is, you can imagine it as the most kind of minimal user interface thing, like nineties computer style. It's just a little toolbar and you basically give it your microscopy image. And then you can actually do a lot of transformation.

So it is [00:27:00] absolutely a very well done, very valuable program. But obviously it's not able to handle high throughput analysis and the way that especially very um, specific individual analysis in the way that you can, if you write your own code. So. But yeah, so everybody in my lab was just throwing their microscopy images into Fiji, a lot of Excel and that kind of worked out, yeah.

But yeah, you really can't do that with sequencing data because now you're talking about millions of reads in thousands of cells across tens of thousands of genes. So obviously you can't do that in Excel. So, yeah, she approached me to do that and I was like, sure, I can do a single cell sequencing analysis.

Uh, [00:28:00] It turned out to be a lot harder than I thought it would be. I think it is absolutely true that analyzing a single cell RNA sequencing data set properly is a PhD project in and of itself, regardless of the data set. Thank you. And so I think that right now the system where usually the biologists themselves are expected to just somehow know how to analyze this like really complex, noisy, high dimensional data is yeah, it's a bit messed up and we're really asking a lot from our biologists right now.

And so I also think that biologists who do that are not getting enough credit. I think bioinformaticians can get very specific and they critique biologists when they make mistakes in their analysis. And I think that the fact that they're doing the analysis at all, it's really impressive. So, 

Benjamin James Kuper-Smith: Yeah, I know, yeah it's, I guess it's similar in the stuff that I did that you, you basically have to do so many different things that like, yeah, of course, you're not going to be great at all of them. 

Laura Luebbert: Yeah. But yeah, so I had absolutely no [00:29:00] idea what to do with this data. So I reached out to Lior because he is the expert at Caltech in RNA sequencing or even, I mean, he's a world renowned expert in genomics. So he was the obvious person to reach out to. And I just sent him an email and I asked, look, I am getting this data set.

It's really cool. I really want to do this, but I have no idea what I'm doing. Can we just meet every couple of weeks and I present to you what I did and you tell me if it makes sense or not. And he said, yes. And so that's how we started working together and I decided that I loved it. I decided that my, or I found out that my deep biological knowledge was actually.

Extremely valuable for this field. I mean, Lior is a mathematician. His lab is mostly mathematicians, computer scientists, physicists some chemistry, he actually has [00:30:00] a wild mix. He also has some biological engineering there, but a lot of more technical degrees. And so then I came in with very little technical computer science knowledge when we started, but really in depth understanding of.

Genetics of neuroscience. So in this case, it was a neuroscience data set. So I really understood which questions to ask and maybe my code to answer them was a little sketchy, but it worked. And if, if you don't know the questions to ask, your code can be beautiful. It's still not really useful.

And so I found that there was really a niche there for me that I just, yeah, really enjoyed

Benjamin James Kuper-Smith: it's, so it seemed like your role switched then. From the, you didn't know that much biologically, but you knew how to code, to suddenly you're like the person who knew nothing about coding or formal analysis, but you have all this biology knowledge.

Laura Luebbert: that's right. That's correct. So, so my [00:31:00] PhD in the end has been completely computational. I have, there's zero wet lab work that I have done myself in my thesis. And I'm really happy with that. I really enjoy the work. I really like being this bridge between biology and bioinformatics. I think there's a lot of need for that. Thanks. Actually, I mean, that's how Gget started

Benjamin James Kuper-Smith: Yeah, I was going to ask, so like how, I mean, maybe also for, I'm assuming that most of my audience, based on the people I've invited so far, don't know that much, or at least are roughly like me in that they don't know that much about Most things, but biology in particular, maybe, or like molecular biology.

So maybe what, yeah, maybe what were you trying to do? What was the problem for which you started, I think as first somewhat accidentally developing and then later a bit more systematically 

Laura Luebbert: yeah, so let me first describe what Gget is maybe so basically it's a software that [00:32:00] allows you to very easily pull specific information from large public databases that contain genomic information. We have a ton of genomic information out there. Think, for example, the entire human genome.

You probably all know that it's sequenced, and you can all look at it. It's all public information. You can literally just go to a website called Ensemble, and you can go look at the human genome and look at all the genes in it. Um,

Benjamin James Kuper-Smith: Read it.

Laura Luebbert: Just read it. Yeah, you can literally just read it.

I don't know if I would recommend it, but you can. But you can imagine that, right, we have amassed in the last 20 years, a ginormous amount of genomic information. We have not just the human genome, but the human, the genomes of. Many more species than that. And even just for the human genome, there's DNA, then there's RNA, which then [00:33:00] encodes proteins, which have different functions and all of that information and also different versions of it, and also genomes created by different groups, all of that information is stored in public databases. And it's extremely important reference information when doing, for example, an RNA sequencing analysis. It's great that the information is public. However, it's not always that easy to actually then access it because every one of these databases has a different website. You get immediately bombarded with a ton of information, a million different options.

So it's actually not that easy to find exactly the information that you need.

Benjamin James Kuper-Smith: so just briefly for context, what kind of information would I be looking for? I mean, so genome, I'm imagining just like a bunch of, C, G, A, and T's or whatever. And like what how do I look for something? What's the, yeah. Okay.

Laura Luebbert: [00:34:00] example is you do your RNA sequencing analysis which means that you have your RNA sequence, which is a bunch of A, C, T's and G's, you align it to the human genome. So, then hopefully you know what your RNA was, what function it had. So getting the human genome that is one thing that we need to do from those websites.

And then also, once you know what your sequences are, once you have the gene names, that doesn't tell you much about its function. It's just a name. So to then go from that to a biological conclusion, you need to learn about the specific genes that you're looking at. Biologists don't just randomly know what all 60, 000 genes in the human genome do, and different combinations of them might have different functions.

And so all of that meta information on the genes that you might be finding in the data, all of that [00:35:00] is in these public databases. And that is the type of information that I think is very important. Most often accessed.

 So you already hinted at it. Yes. All of the information is there, but it is quite difficult to access. So all of these websites have different interfaces. You have to learn them. You get bombarded with different information, 99 percent of which you're not looking for at any given moment.

And so basically what GIGA does is you can give it to a client. The gene name, and it will pull the information about that gene from all databases simultaneously. And it's very specific information. So I really leaned on my knowledge of the information that biologists needed here to make a selection of exactly which information I wanted to empathize and pull out.

And so that's how GGET started. That was the first thing that GGET did. [00:36:00] Facilitate access to these websites so you can very easily pull certain parts of information from all of the websites simultaneously without having to access each website manually and try to learn and navigate and do all of that for hundreds of genes.

Right? That's how we get started. Now we have I think over 16 different. Modules, so sub packages of GGET that each do something else. And we have over 120, 000 downloads all over the world which is really cool to see. So it really hit the spot. I think a lot of biologists were struggling with trying to access this information quickly, and I think that, yeah, specifically for GGET, this knowledge of what biologists need in their analysis.

It's now also widely used in protein sequencing analysis. I wrote a module to run AlphaFold2, which [00:37:00] is a very very famous protein modeling algorithm. Yeah, I really had to think very carefully about how I wanted to design GIGET. So that it wouldn't have the same problem these websites have, but there, there's just too much and that then makes them too difficult to use.

Yeah.

Benjamin James Kuper-Smith: is one of your problems just websites? So it seems to me really that like the, it really is like a software, like it really like first when I heard about it, I thought it was, I mean, it really sounds like a way of using software to automatically just get information that you want from the web.

So is one of your problems that these websites just change their, like the way they set up their structure or something, and then you have to reprogram everything or is that

Laura Luebbert: Yeah. That happens all the time. Yeah. So we have bi weekly tests that run and check every single call that I make to any of the servers to make sure it remains stable. And [00:38:00] I capture problems like I capture bugs and their new releases in my tests all the time. So, Yeah, it definitely has grown out of that.

So, for example, for the AlphaFault 2 module like the company who made AlphaFault DeepMind, they were so proud that they were like, oh, our code is all open source. Anybody can use it. But then you go to the GitHub repository and it's yeah, you can use it. But you need a modern NVIDIA GPU, and you need to be able to work in a Linux environment, and you need two terabytes of disk space, and I'm like, no biologists is going to have that.

At least not the ones that are generating the sequencing datasets. And so the AlphaFold module is not a server call. It actually runs a simplified version. Of a fault locally on your computer. And I wrote all of my G get modules so that they run on any laptop. So they have very minimal requirements for computational [00:39:00] resources.

So I just wanted to

Benjamin James Kuper-Smith: did you realize how big of a project this was going to be when you started, because I mean, this is, I mean, yeah, like maybe. Yeah, maybe because I don't like the program in particular, this seems like an enormous task to me. But it probably also actually is. Um,

Laura Luebbert: no, no, no. I 

Did not set out to build this at all. Uh,

Benjamin James Kuper-Smith: it's,

It's, I mean, it seems like you're almost like a software developer now with a biology background.

Laura Luebbert: yeah, I mean, yeah, certainly G get, I think it was, it's becoming more of a software development challenge. Though, yeah, you still need a lot of biological understanding to even know which problems you need to solve. But yeah, I did not set out to build to get it started because I was getting really annoyed and tired of having to constantly look up different genes and different gene IDs that are completely meaningless they're just a bunch of letters and numbers.

Manually on these websites. It [00:40:00] just frustrated me so much. It was a huge, I thought waste of time that I had to keep doing it. And I also it's very error prone because you're manually copy pasting information. And so that's how I started building Gget and I built this first module, two modules uh, Gget search and Gget info.

They basically do exactly that, just finding information about genes, combining information from different databases. And it just came up in a conversation with Lior and he was like, you should make this public. Other people might use it. And we were like, yeah, maybe there will be like two or three other people who will find this useful.

We'll make it public, we'll post about it. And then it went completely viral and now it's become a huge thing. Yeah.

Benjamin James Kuper-Smith: I mean, obviously I only have hindsight bias now, but it does seem obvious that if there's something that, that replaces a time consuming and error [00:41:00] prone manual search automatic yeah, yeah, people would like that.

Laura Luebbert: Yeah, no, I did not realize how many people had this problem. And also that there wasn't really a good solution out there for this problem. And I think a huge part is that nobody who really have this very deep biological understanding and also an understanding of these biologists, they obviously know how to code because they're analyzing this data, but they don't necessarily know how to make an API call.

They don't necessarily know the. Intricacies of bash and Python functions. And so you, I really wanted something very easy, one line of code, I think is one of our ethos that you get is every module solves a problem in one line of code in the simplest form, and then you can build from that, but yeah, that was really important to me.

And it turns out other people had the same. The same problem,

Benjamin James Kuper-Smith: Yeah. But I guess it's a [00:42:00] lucky coincidence in that sense that, as you said, because you have to have this As I said first, the biological knowledge for what you even need to do and how you would like it to be done. But also, yes, you know, that's also what I meant with like, it's a big software project where like, I know how to code.

Like I can do my analysis, I can run and psychology, my stimuli, I can do that. But I don't want to have to deal with all the stuff you just mentioned. That's, I've heard of these things, but but I guess you found a really nice combination where I guess stuff that other people, they would not like one or the other.

You like both of them.

Laura Luebbert: exactly. Yeah, no. Yeah. I love it. I really enjoyed working on G get I mean, like I said I was building it for myself. So I was the first version. I was just, when I had time on the weekend, I was sitting at the cafe and just working on it. If, for me, it was also, I was still teaching myself how to code on Python.

So it was a great problem to do that. All right. Every once in a while, I visit some of the old modules and I'm like, Oh God, I need to replace all of this code. This is terrible. [00:43:00] We've done a pretty good job of that. We've made it a lot more efficient. But you know, the first version worked and it was correct and it was super useful.

And actually, I mean, I always tell this to young biologists who are like why would I bother to learn how to code? There's all of these people out there who are so much better than me. And I always say that, yeah, they might write better code, but they don't necessarily know which problems to tackle. And you can write beautiful code. If it's tackling the wrong problem, then it totally doesn't matter. So

Benjamin James Kuper-Smith: Yeah. And I mean, I think the standard advice for like people who want to be entrepreneurs or that kind of thing is often take a problem that you have

Laura Luebbert: yes, that too. Yeah. Yeah.

Benjamin James Kuper-Smith: you did, right?

Laura Luebbert: exactly. It was just a problem I had. And I think next time, I hope next time I will have more confidence to know that if I have this problem. Probably other people do too and it's probably worth [00:44:00] solving.

Benjamin James Kuper-Smith: I mean, let's, yeah, I mean, obviously it's also lucky that you did solve a problem that lots of people wanted. There is also an alternate universe where you just put the same amount and like, no one really cares.

Laura Luebbert: no yeah, I, a hundred percent, I would say luck was also involved. Um,

Benjamin James Kuper-Smith: Like the, the, the regular updating and checking whether it works and all that kind of stuff. So from what I understand, you're going to start a postdoc soon somewhere else. So what's the kind of Plan there for maintaining gget is it or is it just gonna follow you now?

This is your child. It's gonna be with you for your life

Laura Luebbert: yeah, I haven't quite figured it out because yeah, I mean, Jiget was supposed to be a side project. It is not by PhD project my PhD project. PhD project was on detecting viral sequences and RNA sequencing data. And GGET is really a helping tool in all of my work. But maintaining it [00:45:00] is technically a side project which has just invaded a lot of my life.

So I would love to get a more sustainable. Long term environment where you get. So, so one of the things that I've been working on when I got funding from the prototype fund in Germany, one of the things I really worked on a lot was community building and that worked quite well, which basically entails that I cleaned up the structure of and both very detailed instructions.

Such that biologists and other experts can now, when they want a new module, they have a problem that I haven't solved yet, or maybe that I don't understand well enough. I can guide them to write it themselves. And so that is something that at least, and it's happened already there. We already have four or five community contributed modules.

Which is also great for [00:46:00] them because they can now put it on their CV, they wrote code that is very widely used and it's part of this very widely used coding environment. So that is something that I've been working on and I would love to see that more. I would love to see more community involvement.

Yeah I also, I mean, I set up a GitHub sponsorship page. Um,

Benjamin James Kuper-Smith: Is that

Laura Luebbert: Oh, it's basically, you can get sponsors if people are fans of your work.

Benjamin James Kuper-Smith: Give you money basically, oh,

Laura Luebbert: Basically, yeah, it's like a buy me a coffee. I think it's a similar page. Exactly. Yeah. It's the exact same thing. So I set up one of those. It's always, nice to get a little bit of financial support for your work because yeah, I am just doing it.

In my free time, unpaid if I'm being honest, I don't see myself stopping anytime soon. I feel a huge responsibility, like whether or not I get financial support. I don't think I'm going to stop. I do feel a huge responsibility towards my users. There's a huge list of in [00:47:00] GitHub issues, there's a huge list of requests for new modules.

So that is something that I need to be very careful of whether or not I have time for that. But whenever something breaks or there's a bug, that gets addressed immediately. So I do make sure that everything that is in the DGET environment is correct and reliable. Because yeah, I definitely feel that responsibility to

Benjamin James Kuper-Smith: It's good to see that you have the same not particularly sophisticated Negotiation strategy that I have of saying Hey, you can give me money for it, but we'll do it for free.

Laura Luebbert: Yeah. Yeah.

Benjamin James Kuper-Smith: I'll do it anyway, even if you don't pay me.

Laura Luebbert: Yeah. If I'm being entirely honest yeah, I would be very disappointed in myself if gig didn't work anymore in the future. So,

Benjamin James Kuper-Smith: Yeah. I mean, yeah, I have a patron for the podcast, but like number one, I've basically never mentioned it. And number two it's, yeah, I mean, I'll do it anyway. It's just, it doesn't depend on that. Yeah. Yeah, I'm not sure it's, yeah, I'm not sure it's the best negotiation strategy, [00:48:00] but I don't know, it makes life easy.

Um, 

Laura Luebbert: I think we should get paid for the work we do. I think we're contributing to the scientific community. But I think part of why you have so many listeners and I have so many users is because we are doing it because we really care. So I think partly because we are going to do it anyways, it is safe.

It's so useful. So, there's a bit of a

Benjamin James Kuper-Smith: and I mean, in, in your case also more than in mine you're also becoming maybe not indispensable part of the scientific community, but like it is very much, it seems like it's a tool that people really use and makes people's life a lot easier. So I guess in some sense it is for you also some sort of not job guarantee, but I would imagine it does help you quite a lot in

Laura Luebbert: Absolutely. Yeah. Career wise it is really great to show that I'm able to produce something that people love and use and rely on.

Benjamin James Kuper-Smith: Yeah, I mean, how many people can say if you hire me, you hire the person who's responsible [00:49:00] for this thing that all of you use? Not exactly like that, but, that's what you can say, right? 

Laura Luebbert: I've benefited a lot from GIGA's success. So, yeah. Absolutely.

Benjamin James Kuper-Smith: yeah, but yeah, I mean, it's obviously still cool that you're doing it. I mean, yeah But okay, so so you're So this is something I wasn't quite sure about so your main research is still something, not completely separate, but it's separate from Gget and that's just a part of what you did and now it's more maintaining and 

Laura Luebbert: Yeah. I will say GGET's success has tinted all of my PhD projects in that think a lot more carefully about this niche that I try to be in between biology and bioinformatics. So I really try to make all of my projects, everything I do, very accessible like I recently published a paper on finding viral sequences in RNA sequencing.

And it's a big paper with a lot of figures, but every single [00:50:00] figure has something called a Google Colab notebook. I don't know if you're familiar with it.

Benjamin James Kuper-Smith: Vaguely but not really i've seen it and used it once but yeah, not really

Laura Luebbert: It's great. It's basically a Google Doc, but for Python code, in that it is an entirely contained coding environment. And you can just press play and the analysis runs, you don't need to install anything locally on your computer.

And it's really nice because you can really see how this figure was produced and you can even change things along the way and see how that affects the results. And so every figure of my paper has a notebook like that, that anybody can just run, all you need is a browser and you can reproduce that figure.

And then also that you can reuse my code if you want to make the same figure but with your data. So yeah, in that sense, gget has become a huge part of my thesis. But I've actually done a lot of work outside of gget.

Benjamin James Kuper-Smith: Yeah, I [00:51:00] really like the whole what you mentioned with you know being able to share And I mean, I guess the data you use is public from what I understand. But yeah, it's basically I'm always torn between, so I learned coding in MATLAB because that's what most cognitive neuroscientists use.

And I just don't like coding. So I've wanted to switch to Python for ages, but I was like, it's the part I like the least. The problem with MATLAB is I mean, in cognitive or computation neuroscience with humans, like lots of people use it. But there's this point where you share all of it and people are just like, yeah, I don't use MATLAB.

I can't read it. It's oh, it is it's technically, it's all opened, but if lots of people don't, can't use it, then how open is it really? So there's a bit of a, ah, but I just can't bring myself quite to, to learn a whole new programming language.

Laura Luebbert: I mean, yeah, I think that's I mean, Python in general, obviously I think it has the most users. Of any programming language, at least of the ones we use in biology and in my field of biology. But that's why Google [00:52:00] call apps I think are so powerful because you literally just hit run and you don't, I mean, MATLAB is also a problem, right?

Because it's not free. So, it's, is it really open source if you need to pay for a program to actually then run the code? 

Benjamin James Kuper-Smith: Yeah, exactly. It's open in the sense that anyone can evaluate But yeah it's not, I mean, to be fair, like the yeah, yeah no, no, you're right. Yeah. Just my aversion to programming is strong enough.

Laura Luebbert: I can recommend Python. I love it. I,

Benjamin James Kuper-Smith: I know. I mean, I've decided to switch in the beginning of my PhD and the beginning of my postdoc. And each time I was just like, but I know how the other works. And then it's just, it's every. It's like learning how to walk again. We're like, of course you know in principle how to do it, but every step is what's that function called?

And all that kind of stuff.

Laura Luebbert: Yeah. I had to so for my main PhD project, I had to write a lot of code in C and that [00:53:00] was so painful.

Benjamin James Kuper-Smith: Yeah.

Laura Luebbert: yeah, so I feel you.

Benjamin James Kuper-Smith: Yeah. I mean, I already have it with R because it seems like in lots of stuff I do, you don't get around R. But

Laura Luebbert: That's

Benjamin James Kuper-Smith: yeah. Yeah. I mean, I think one day I'll switch, but by that time, I don't know, people probably be not using Python anymore. It's probably, by the time I switch, it's not going to be cool anymore. It's going to be some whole new programming language none of us know

Laura Luebbert: is up and coming, and a lot of people that I know swear on Rust, so, maybe you can get a head start.

Benjamin James Kuper-Smith: Yeah, I'll get a head start on that. Yeah. Or people don't use that and I've got a head start on nothing. Yeah. We'll see for now. I'm just, I'm just, I've got enough to learn while keeping the programming constant.

Laura Luebbert: I mean, yeah, I mean, that's, again, that's the problem, right? If you were the one generating the data, if you were the experimenter.

Benjamin James Kuper-Smith: yeah, exactly.

Laura Luebbert: difficult to also be able to learn these very technical skills. I mean, I spent my entire PhD learning this full time. I don't know how anybody expects somebody who does [00:54:00] wet lab work full time to also have the same level of knowledge.

Benjamin James Kuper-Smith: Yeah.

Laura Luebbert: So,

Benjamin James Kuper-Smith: Yeah. Okay. But your postdoc , That's, what are you doing there? Because that's gonna start what did it say, like two

Laura Luebbert: uh, September. Yeah. Yeah. So my postdoc I'm joining a lab at the Broad Institute in Cambridge. Yeah. And I'm joining the lab of Professor Pardis Sabeti, and she is just an incredible person. I think I think it's fair to say that part, in part, thanks to her we never had to worry about Ebola outside of Africa.

So she has done a lot of work on Ebola virus during the outbreak. Her lab has It's done a lot of sequencing of genomes, a lot of rapid test development to very quickly react to new pandemics and keep them contained. So like I said, my, my latest and big [00:55:00] project was on detecting viral sequences in sequencing data specifically new viruses.

So even trying to recognize viruses that we've never seen before. Because with RNA sequencing data, you sequence all the RNA, but then the way it's analyzed is you compare it to a reference, and that's how you figure out what is what. So anything that is not in the reference just gets thrown out, and so I try to focus on that part that is thrown out, and I built an algorithm that can basically learn what looks like it's a viral sequence.

And then pull those out, pull out those leads and then also try to figure out which virus it came from, whether it's a new virus or something we've seen before. And then I can also see if so I'm mostly working with human cells I can see how the human cell response to the virus being there.

And [00:56:00] so, my preprint is out describing all of that. Putting in some

Benjamin James Kuper-Smith: it in the description.

Laura Luebbert: wonderful. It's under review right now. But that was just a first step because this was, again, entirely computational. I think I have potentially identified some new virus like sequences that have an effect on cells that they are infecting or that they are present in.

But there's not much else I can do with that right now. And so, I want to continue that work in her lab. And basically what I'm going to be doing is analyze a lot of really amazing data. I mean, the Broad Institute they are the epicenter for genetics, genomics and sequencing data. And so I'm gonna, I'm going to be hunting for viruses and I'm going to see if I can actually.

Extract them characterize them I've been also doing a lot of work with [00:57:00] cancer. So we know that a lot of cancers have viral causes. So, yeah I'm trying to just see if there's some correlations that perhaps we have missed. And obviously if a cancer has a viral cause and you don't know it like that might be.

A relatively straightforward treatment route and also early detection and early treatment route that we're not aware of.

Benjamin James Kuper-Smith: Virus Hunter is a pretty cool title. Yeah.

Laura Luebbert: I do want to say that any viruses that I would hunt would already be out there in humans. So,

Benjamin James Kuper-Smith: Yeah, you're not creating them.

Laura Luebbert: yeah, I'm not creating them.

Benjamin James Kuper-Smith: yet. Yeah.

Laura Luebbert: Yeah, yeah, I would love to continue that work and just really apply it at, honestly, a global scale and see if we can so, so Pardis and her lab, they are extremely passionate about global surveillance of viruses, not accepting that, oh, [00:58:00] I just have something, the flu or this virus, maybe, I don't care just not accepting that anymore and just really getting better at keeping track of what viruses are doing, where they are, how they're spreading.

And so I'm super excited to, to be part of that.

Benjamin James Kuper-Smith: I mean, it also sounds like quite an administrative task, right? Because you have to coordinate all the different places and collect all the things centrally or something like that.

Laura Luebbert: Absolutely. Yeah. So they have a lot of collab collaborations with local institutions in Africa. And so, and they also work with the CDC and the US government. And one of the things I'm really looking forward to learning from them is how do you do all of that? How does that look? How do you even set up a global viral surveillance system?

Where do you even start?

Benjamin James Kuper-Smith: It's funny. I, so on your website, you mentioned, that you do a postdoc there soon. So I had a brief look at the website and I was like, this is not a lab. This is an institute because it's like, she has what, like 10 people working in [00:59:00] administrative stuff. And I was like, what are they doing there?

But now that you've explained it a bit more, it makes a bit more sense why there's so much, not just I mean, there's lots and lots like science, like postdocs, like herself and PhD students, but like postdocs. Now I understand like why there's so many like of those

Laura Luebbert: no, yeah. Her lab is a huge part of her lab is basically run like a company and they do very immediate reactions to outbreaks and very rapid test development. And so she has that infrastructure to do that at work. Relatively large scale in her lab.

Benjamin James Kuper-Smith: it all makes sense now. Yeah, yeah. Okay. So let's talk about a little journal club you did a few years ago. What happened?

Laura Luebbert: Yeah. Okay. So this is completely unrelated to any of my work. During my rotations at Caltech, one of my rotations was in an insect behavior lab. And for a German club in that lab I was assigned to very fundamental papers to [01:00:00] the field of the honeybee ragdoll dance. So we'll see. Funny weather events.

I'm sorry, that was your part.

Benjamin James Kuper-Smith: You can do both. Yeah. You can interview yourself if you want. I'm happy to just listen. I'll just, I'll edit it and publish it. You can do the rest.

Laura Luebbert: No, so, well, so,

Benjamin James Kuper-Smith: What is the honeybee wiggle dance?

Laura Luebbert: Yes. Good question.

Benjamin James Kuper-Smith: Thank you.

Laura Luebbert: Um, So we've known for a long time that honeybees dance to communicate the location of food. And I should disclose again, insect behavior is absolutely not my expertise. So I'm just reporting on my understanding of it. But we've known for a long time that honeybees used to ride a dance.

to communicate the direction and distance of food sources to other honeybees so that they can then find the food source [01:01:00] as well. And so the papers that I was reading were trying to figure out exactly how honeybees track the distance. Do they remember certain monuments along the way?

Do they track their energy consumptions? Do they track their wingbeats? There were a lot of different theories on how honeybees might do this. And the two papers that I was reading and this was the fundamental work of the first author on those papers and it was very fundamental to the field, is that we found that honeybees do that based on visual cues.

So basically visual motion is how they track distance flown. And then he also tried to make in these two papers a very specific calibration of this many seconds of dance mean this many kilometers in distance or meters in distance to the food.

Benjamin James Kuper-Smith: So just it's funny that like [01:02:00] I've done lots of episodes on the neural basis of spatial navigation and only just now it occurred to me that this kind of also fits into that but yeah sorry which is usually just like in rodents or in mammals but yeah I guess that actually is that's pretty much the same question yeah

Laura Luebbert: I mean, it's super interesting. And I quite enjoyed reading about the experimental designs actually. But well, okay. Why am I telling you about the Hornaby dance as a computational philanthesis? Basically what happened is that I was reading these papers. I had no idea about the Hornaby dance.

So I started reading other papers by the same author. I realized that I was looking at the same data over and over again. Basically when, as I was trying to really understand what was going on, I was also trying to redo some calculations. So the specific calibration number, I was trying to understand it by actually redoing the analysis.

And I [01:03:00] just kept getting a different number. And so this is really how it all started. And then from there I started very carefully analyzing the data. And I found that several of the figures were showing identical data, but for different experiments with different numbers of replicates.

Benjamin James Kuper-Smith: Okay. So, I mean, so you found some, yeah. So basically what did you do? This is not what you would expect to find. 

Laura Luebbert: Yeah, I was pretty concerned really concerned about those findings and I had no idea what to do. So first I presented my findings at the journal club because I also wanted the feedback from actual experts in the field. And that is really where this most shocking response happened that stuck with me for my entire PhD afterwards, which is that.

Nobody cared. The response was just Oh yeah, that sucks. Oh, well. [01:04:00] And I reached out specifically to the professor of the lab and he said, Oh he basically said he didn't know what to do. Which is a little bit better than not caring entirely, but also he didn't try very hard.

And also. You're a tenured professor at Caltech. What do you mean you don't know what to do about the data integrity issue? So I reached out to another professor at Caltech. This is the person who taught the ethics class that year to all the first year students. So I just had to listen to him for a month preach to us about data ethics and it's so important.

And if you make any mistake, you're horrible. And he just basically told me, Oh yeah, this happens. Science is messy. Do you really want to waste your time on this? And I got really frustrated. And at the time, I mean, I was a first year student. I wasn't going to cause any trouble on my own.

Well, I did a little bit cause I did tweet about it. And then I got some help from [01:05:00] Dr. Elisabeth Bik. Who's obviously amazing. She's done some really important work in the field of data integrity. And so she helped me find the papers on PubPear, and at least flag the issues that I found on PubPear where they have remained ignored until pretty much this month.

Benjamin James Kuper-Smith: yeah. So yeah, there's an entire episode with Elizabeth that I'll also link in the show notes where we talk about all sorts of stuff I mean related to this particular. Um, Yeah, I mean, so maybe just to I guess we've set the scene a little bit, to You know motivate a little bit why we're why we're talking about this is that You I mean, so I, as I've had basically two episodes before about this kind of topic.

The first with Joe Hilgard and the second with Elisabeth Bik. And basically the first one I had with Joe in large part was because I also found some irregularities in a paper. Where I basically thought two figures were mutually exclusive. Like there was a specific set of like findings where like, wait a minute, these can't both be correct.

[01:06:00] And I just had no, I mean, I presented it in our lab meeting just at the in the lab. And Basically, everyone's yeah, that doesn't look good, but like just no one knew what to do, right? Like you're not like trained to if this happens then this is what you do, right? And so I like one of the reasons I invited Joe to talk about it is because I was just like, what do I?

Joe, what are you doing?

Laura Luebbert: Yeah.

Benjamin James Kuper-Smith: I mean interestingly he so he is in particular I interviewed him at the time because he found this like ridiculous fraud case in his own field, which is how he got into it because someone in china reported studies With like thousands of people where usually you just have a hundred or something So it like messed with his entire meta analysis.

He was running Anyway, so we talked about that and then I asked him also what I should do and he you know As someone who has spent lots of time Reporting these things and getting papers retracted. And that kind of thing Basically also told me like, you know what it's probably not worth it it takes a lot of time.

Yeah, you have no idea how the people are going to react and Yeah, like it's potentially a lot of time that [01:07:00] just goes to waste. And then, I didn't do anything during my PhD, also because, you have other stuff to do. But then when I talked to Elizabeth a few months ago, I actually Because this was like between postdoc, I had a bit of time.

So I actually wrote it up, contacted the original people. And it looks like they are the best case scenario where they basically said they responded within two days. They said yeah, this basically we're taking it seriously. And they already contacted the journal. To ask like what to do exactly.

I mean, so yeah, it's funny because Elizabeth, I asked her and Joe, like what to do. And both of them were like, I mean, you almost don't need to bother contacting the authors because they'd never respond and that kind of stuff. Anyway. Sorry. I just didn't want to interject that much.

But the point was that I mean, this is a very, I mean, it's not an unusual, it's not what I said, a rare situation, but it's a situation that it seems like the entire system isn't exactly prepared for. And yeah, we're probably not the only people who have found something like this.

Laura Luebbert: No. I mean, [01:08:00] I have gotten so many messages from especially early career scientists telling me that the exact same thing happened to them.

Benjamin James Kuper-Smith: Yeah, and so I guess the idea is to just share your story and to some extent continue my story in little bits over the years now, In a year, I'll do another episode then probably finish it or something like that yeah to just give people like some idea of what to do and some sort of Suggestion or guidance or something like that?

 So maybe going back to your story, so what Okay, so you did nothing. Okay, cool story.

Laura Luebbert: Yeah, I did not think that's , the story does continue. I still haven't quite decided yet if it's a happy end or not, so I don't even know how much it's really helpful in giving advice or if it

Benjamin James Kuper-Smith: Well, it's at least sharing examples because it seems most people don't haven't even heard of examples, right? That's I mean, it's basically it seems like I have the examples of Joe and Elizabeth who've done this a lot especially in Elizabeth's case but it's also [01:09:00] very different if you are You know, I mean so if you're I mean in your case, I guess, you know one one Interesting commonality to me is and maybe you can tell me whether this is a coincidence or not.

It's that both of us have You In my case right now, I've only mailed them, and it's behind closed doors for now, but it, they're probably, there will be some sort of public statement at some point soon. Yours is then public, but both of ours are cases where the field or the papers we're criticizing were from a field that was adjacent but not our own.

And I have to admit one thing I, so I contacted them myself. I just wrote them an email by myself and it's scary. Like you write this report and you're basically saying in my case, it's three relatives it seems like fairly senior, well respected professors. And you're going like, Hey, you three professors, I'm a PhD student.

I think you're wrong.

Laura Luebbert: yeah, no, it absolutely. And I think having it be in an adjacent field makes it a lot easier because you already mentioned, you never [01:10:00] know how the authors are going to react and I mean, that is something that I was told a lot is that do I really want to jeopardize my career? I think in the end that was really bad advice.

No. Young scientists should ever hear that in the response of them doing their duty as a scientist and ensuring data integrity. 

Benjamin James Kuper-Smith: I mean, but don't you think it's also, I mean, you don't

Laura Luebbert: are a lot of, that's true. Yeah, certainly if it's in your field unfortunately, even if you're right it can really hurt, especially younger scientists if they anger the wrong person.

Benjamin James Kuper-Smith: Yeah, exactly. So, I mean, in my case, it was basically, I mean, it's interesting because in my case, I didn't have access to the data. I just found some irregularities and where I thought there was stuff that was mutually exclusive. So they've, I respond, I got an email from them like a week ago where they said we've took a while to get the data, which makes sense.

It was like a while ago that they published it and that kind of stuff. But they basically said like, okay, we got all the data again. We reanalyzed and we think your initial point [01:11:00] was actually not correct. Okay. There's some other stuff, and we are going to issue a correction, like there is some stuff, like some of the stuff you correct is correct, but like the big point you had is, we think that's actually true the way we have it in the paper, and so, I made, I mean, I'm still really surprised by that and I'm looking forward to getting the date and seeing how both how it can be true but you know you make in my case I basically just made sure showed it to some people the draft I wrote what you think of it that kind of stuff and then I Sent it thinking they're not from my field.

What's the worst that can happen? That's basically it right? But yeah, if it's exactly my field

Laura Luebbert: It's scary. It's very scary. You actually, you also mentioned a great point. I'm skipping ahead a little bit here, but yeah, so, so, I mean, my story ends with eventually me and my current advisor, we wrote a report and we figured out that there's nowhere to publish it and there's certainly nowhere to publish it with a peer review.

We really wanted a second pair of eyes on this report as well.[01:12:00] 

Benjamin James Kuper-Smith: Hmm. 

Laura Luebbert: Very confident in our conclusions. Obviously now we've had a lot of peer review because it has gotten a lot of attention, which is great that this is exactly what we wanted.

Benjamin James Kuper-Smith: But have you gotten so have you gotten actually good peer review or is it more like You know a lot of noise and people react to it in more like a social commentary style rather than actually You know there's a difference between you like knowing you can anonymously give feedback on this thing and Spending time with it versus seeing something tweeted and then checking of it quickly

Laura Luebbert: Yeah, none of the issues we brought forward have been clarified. I'm not entirely surprised by that because there's not the issues that we found. I just not really debatable in that way. The data is not very similar. It is identical and it is for different experiments. So I think in that way, there's just no way in which you can be like, Oh yeah, we looked at the underlying data and it actually looked at that.

I was like, no [01:13:00] but we did get some helpful comments Initially, actually when we submitted to the Journal of Experimental Biology they rejected the manuscript but one of the scientists working for them contacted us separately and said off the record here are some suggestions that I found in the report.

And that actually was super useful just in the phrasing and how to explain. What we did, cause obviously we know very well what we did. So sometimes it's just really nice to have a second pair of eyes, especially in that field. And so that was super helpful.

Benjamin James Kuper-Smith: So what's the current situation? I mean, I know you, you published the preprint it got some attention including from the original author. So what is, I mean, I think you said he was going to respond

Laura Luebbert: Yeah. He promised a detailed response. Yeah, I think, I mean, you were talking about the best case scenario. I think we have one of the worst case scenarios, unfortunately. [01:14:00] It, it was quite bizarre because. For example, the Journal of Experimental Biology who, they published four of the problematic papers, two of them with very obvious data duplications.

And the journal actually posted a correction in which they list all of the duplications. So, they say themselves, This data is duplicated here for this experiment with this number of replicates and then this data is duplicated here and it's it's a whole list. It just goes on and on, but then they conclude with the author takes full responsibility.

The data is not available anymore, but the conclusions are not changed. The conclusions that are drawn. From the data, which we have just established, at least one of the two sets is obviously wrong because it can't be identical. And so that is really bizarre to me. So, I think the other journals are still investigating.

I know [01:15:00] from, so science corporate stories story, and I know that their internal investigation is ongoing. So I'm looking forward to, to see how other journals are

Benjamin James Kuper-Smith: because one of the papers was also published in science, right?

Laura Luebbert: Yes, exactly. Yeah. Yeah. I mean, I think it's like seven or eight different journals that are impacted by these papers. Yeah, I'm very disappointed in the handling.

I mean, I'm now only talking about the journal responses, but yeah, the Journal of Experimental Biology. It also looks from the responses of both the author and the journal that they were aware of these issues a year ago. The author himself has commented on our blog post that they were contacted by the journal a year ago.

That wasn't us. That was way before we even started writing our report. And so it sounds like both him and the journal just decided to not it, not publish any sort of [01:16:00] response or just making the readers aware that this is happening in these papers. we published our report and then a month later, it sounds like the journal published the responses that the author gave a year before to only those issues.

Benjamin James Kuper-Smith: So it isn't actually. Okay, so you actually is it just to get the timeline correct then so you published the preprint, Well, but you did contact the journals first, right? So so basically the question is is the comments that the journal officially made Is that in response to your stuff or is that?

independent of that Yeah,

Laura Luebbert: 2020, in the first year of my PhD, that resulted in those two pub peer comments, which were completely ignored by the author and the journal. Okay. And then in the last year of my PhD so that was like October last year this came up again in a discussion with Lior actually, because we were talking about what I wanted to [01:17:00] do after my PhD and I was still so frustrated by this thing that happened in my first year that I was like, I don't know if I believe in science anymore, like that just like goes against everything I've been taught. Um, and, And you as a kind of person who will not let something like that slide, which I really had my eye on. I think we need more people in science like him. But so, so. We wrote our report, and we sent it to the journal of experimental biology in March because several of their papers are affected because we wanted a peer review process.

And, and we thought that they might be interested in publishing this kind of report on work that has been published mostly in their journal. They rejected the preprint. They rejected the manuscript without sending it out to peer review said that they would start an internal [01:18:00] investigation.

Benjamin James Kuper-Smith: So just briefly, so it was just wrong type of publication, basically, or what was their, it's just yeah, what was their?

Laura Luebbert: yeah, they said that they didn't want to publish anything that includes the criticisms of works that are published in other journals. And also generally, I mean, they really just wanted to focus on the research finding on this calibration number. and not talk about any of the duplicated images.

And we said the whole point is that this happened over and over again. It needs to stay together. If you look at any one piece of a story, it's not meaningful. It's the whole of the sum that is greater. Um, So the,

Benjamin James Kuper-Smith: yeah, sorry I interrupted you the so they started internal investigation.

Laura Luebbert: Yeah, so they started, they rejected the manuscript, but told us that they started an internal investigation. Nothing happened. We [01:19:00] said, okay, whatever, we'll publish our manuscript on the archive so that at least it's out there and we just know that this happened. And then a month after we published our archive manuscript, they published their expressions of concerns.

But now, based on what the author has commented on our blog post and wondering whether they just posted the responses that they actually got a year ago. And, I mean, this is entirely, completely speculation, but my guess is that they knew those responses were pretty bad. And so they just rather not talk about it, hoped that nothing else would happen.

And then when we published our manuscript, we basically, through that, forced them to acknowledge the problem. And that's really disappointing, right? The fact that in this case, we had to go public like that to get any sort of response from [01:20:00] even just one of the journals and the author that, that was, that's really disappointing and scary.

Benjamin James Kuper-Smith: Yeah. Yeah, I mean as you said, like some of this was maybe a little bit of speculation, but one question I had is basically about contacting the author. So in my case, I basically I was like, okay, I think. I mean, first of all, my case seemed a lot like it seemed less severe than what you found and it seemed like there was a solid chance that there was just an error in the analysis and it wasn't done that carefully.

That's basically it was very possible. And also I, Didn't read any other papers by the people, did do any of that I had it just looked like the one paper. But in my case, I basically I was I initially thought so I asked I think elizabeth you know what should I do? And I think she said basically just contact the journal Put a comment on PubPeer Okay basically what you did, right?

But then for me, I don't know it didn't quite sit right and not let the people know before I did this So I sent them an all an email included them all in it and then you know I mean I say it's the best case [01:21:00] scenario so far. Nothing has actually happened They just say I mean, I don't want to be too distrustful But like I haven't actually seen gotten the date or anything like that, right?

um, we're We're still a bit behind some of the things but If, let's say, in my case, it turns out that, their initial re analysis is correct, then I'm quite glad that I went to them first and they can correct it. Yeah, I'm curious it seems like you didn't do that, right?

Laura Luebbert: That's correct. Yeah. We didn't. Contact the author directly part of that was because of the nature of the problems that we found, there was no really debating the issues that we were finding that said, obviously, like I said, I very much encouraged discussion on the preprint that we put out, we really tried to describe in detail what we found and where, and any analysis that we did do is there and is reproducible.

. But yeah, I wanted to be able to stay objective in this and giving the [01:22:00] of the issues. I didn't think I could do that. I, and I also don't think, I didn't see how meeting the author and talking to the author would add anything to it. And in addition to that those PubPeer comments had been sitting on PubPeer for four years.

And SW was, well, the author himself has, confirmed that he has known about the issues, even if he didn't see the popular comments and popular dusts and authors and email when the comments up here on the papers, but even if he didn't see those, he has admitted himself, but he has known about these issues for a year.

And so he has had a lot of time to address these issues to do the right thing. And he didn't. And so I think yeah I don't see how us. I don't know

Benjamin James Kuper-Smith: Yeah, right, I see it's yeah, it's a different kind of problem in that sense, yeah if you have If you think you've found systematic problems across several publications compared to [01:23:00] Me having looked at one paper and going hey, I don't think this is right. Yeah, that's obviously Okay. Yeah, I mean, I always don't want to say this because it might give away some stuff but for but one interesting thing that Elizabeth said is that you can basically from the response of the people tell what happened.

So she said basically, if The people who actually made an honest mistake will respond immediately and be very friendly and grateful to you Which is pretty much what the people did in my case And the people who maybe did something more serious tend to start attacking the people and questioning why they even did it in the first place and there was a little bit of that in the comment on the blog

Laura Luebbert: Oh, very much. Yeah, no, I mean, there has been zero addressing of the issues that we found, and a lot of personal attacks on Leo and Dior. I mean, he literally called us unprofessional. And if you read our manuscript, we're not even mentioning any one person by name. We are just [01:24:00] Correctly citing papers and listing the issues we found in each paper.

Benjamin James Kuper-Smith: But I mean to be fair like it is I know you don't, I mean,

Laura Luebbert: obviously, yeah, it happens that most of the papers have a common first and last author. But we are just listing the papers and listing the problems that we found. So, yeah I have been asked how I would respond to that and it's really difficult because I honestly don't know what we did that was unprofessional or unscientific.

I think we wrote a scientific paper, we made it very transparent very clear, we invited uh, comments. This is why we wrote the blog post is we wanted a platform where people could respond. And we're inviting, definitely inviting experts, including the author to comment on our technical analyses.

But that has not, yeah, happened at all. 

Benjamin James Kuper-Smith: It's really weird, like I can kind of understand why people feel attacked by these kind of things, even if it's done in a completely [01:25:00] professional way. Like I think it's a completely natural response, but, I think, yeah, I mean, I think this is also part of the bigger problem, right, where it's just somehow it's never done, right, it just happens so rarely.

And, I mean, there might be people like Elizabeth who've stepped aside, basically, out of academia itself, and do this as their thing, and who, who seem to have a thick skin and don't mind being attacked by people. So, 

Laura Luebbert: And look what she has been through. I mean, she gets attacked all the time, and it's completely unprofessional and, yeah.

Benjamin James Kuper-Smith: Yeah, but it's so weird how I mean, I think it's just because yeah, because it happens so rarely once when someone does it, I guess it's often seen like a personal attack or something.

Laura Luebbert: Yeah, and honestly, I think that's part of the problem. I think that, retractions shouldn't have as much of a stigma as they do. Like everybody makes mistakes or sometimes we just simply get it wrong. A huge part of science is just sometimes you just get it wrong just because we are [01:26:00] totally fine with getting it wrong until the paper is published.

Then all of a sudden it has to be holy.

Benjamin James Kuper-Smith: Yeah.

Laura Luebbert: And it's really weird, like we should be able to, what we did was peer review. And for some reason, the Journal of Experimental Biology just didn't accept it as peer review because it happened after the paper was published. And I think that is very detrimental to correcting scientific literature.

And I think part of that is, is the fault of us as scientists, as a scientific community. I think. Obviously, again, our issues are of a very difficult nature, but there are a lot of restrictions where, the science just doesn't hold any more based on new evidence and then it should be totally fine to retract the record. 

Benjamin James Kuper-Smith: Yeah, I like your comment about peer review because it's I've had a few situations where I'm always surprised that people think peer review is the only thing that happens when you submit it to a journal and you get invited [01:27:00] by the journal. Everything else is somehow not counted as peer review, which is funny to

Laura Luebbert: on puppy, like comments on pup here, a peer review comments on bio archive, a really valuable peer review.

Benjamin James Kuper-Smith: Yeah, I mean, I've mentioned this example once on the podcast before when We were reviewing a paper and they basically made a claim that no one had done something We said well, there's three preprints on this it doesn't really matter like it's not But just mention them and they basically said like we don't cite preprints like we want to go through peer review and it's like you are a peer Like you can read it and do your own peer review You can do it, you know

Laura Luebbert: Yeah, yeah,

yeah. Yeah. I, yeah. I think it's a huge fault

Benjamin James Kuper-Smith: Okay

 Yeah, I mean, so, I guess one kind of thing that's I mean, not unfortunate, it's maybe the wrong word, but we don't have the official response by the author yet, who knows what he's going to say. So, this, again, this isn't about, I hope our discussion isn't about this particular person or even the Honeybee Papers, but [01:28:00] it's, as we said earlier, I hope it's more that, other people who find this have some sort of something to grasp onto or some examples.

I'm curious what would you, I mean, would you do it again? What would you recommend yourself a few years ago or?

Laura Luebbert: I think it's a little bit early to say that I think that going in. So I want to just very clearly say my goal here was. to correct the scientific record. There were very obvious problems and I felt strongly that people should know about them. And that was my only goal. And I think that we got, I got really lucky and that I was giving a platform via Twitter, via Lyor blog to be able to get that message out and inform the readership.

Our findings were very widely shared within the community of insect behavior and [01:29:00] etymology. And that's great. I'm very happy to see that. That was my goal, is to let them know. So I think in that sense my goal has been achieved. I think while doing that, we realize that I shouldn't have to have gotten lucky to achieve that goal.

There should have been a platform to do that. And there should be some, there should be some incentive for young scientists to, to. To talk about their findings. Again, I've had so many people reach out to me with similar stories and, who knows what they found? Who knows which findings we're building on that at least several people in the world already know are not actually correct.

Like that's that for science everywhere like that is lethal. So literally lives depend on science and on science being. Especially foundational papers like these, we need to be able to build on them. So in that sense, I've achieved my [01:30:00] goal. I'm very happy about that. I hope that moving forward it will, this issue of there not being any defined way of getting word out like this will get some attention and maybe we can even work on solutions.

To it Leo and I write about the journal of scientific integrity now, if we can get that and then also give it the same status of nature as nature, that would be great. Um, but, But yeah, this kind of work, it requires, it's really difficult work. It requires a lot of very careful analysis and right now there's just no incentive to do that.

So I would love to see that. be solved in some way in the future and, if this was a step in that direction then, yeah, absolutely, I would do it again.

Benjamin James Kuper-Smith: Okay. Yeah, it's such a difficult, I mean, for anyone who's in, in that kind of situation, I find it, it's I still don't have a good answer [01:31:00] to what to do, because it's I think in our case, we both maybe got lucky in different ways, to some extent assuming that, my case continues the way, I mean, they, they even said we we told the journal we're going to write a correction and we want to acknowledge you.

For letting us know about this, right? So like for me, it's great I can put this on my cv and say you know I was acknowledged for this in the corrections and whatever because it was a lot of work right because like You do check your work five times before you tell people that you think they're wrong and The But yeah, I mean, but the probability that someone does the same and it's just taught to fuck off or ignored is yeah And then you've just spent lots of times and someone just shouted at you.

Yeah, great.

Laura Luebbert: Yes,

Benjamin James Kuper-Smith: I mean

Laura Luebbert: Yeah. Yeah. I mean, I spent weeks on this initial analysis

Benjamin James Kuper-Smith: Yeah, yours was much more extensive, 

Laura Luebbert: so, so the preprint then but I meant even like the initial, just a few issues raised on top here. I mean, I was a first year student. I really went through those papers over and over [01:32:00] again.

Because I really wanted to make sure that it wasn't just me,

Benjamin James Kuper-Smith: yeah 

Laura Luebbert: A mistake, of course. So, yeah. And that, all of that work, just for it to be completely ignored for four years right then and.

Benjamin James Kuper-Smith: Yeah. Yeah, I guess you did have that case actually. Yeah

Laura Luebbert: Now.

Benjamin James Kuper-Smith: Yeah, 

Laura Luebbert: And yeah.

Benjamin James Kuper-Smith: But, so what's the status now? You said you couldn't find a journal, what's the

Laura Luebbert: Uh, So actually now that there's been so much attention for the issues that we face, we've had several journals either interact on Twitter or actually reach out to us. So for example, there's a, apparently there is a journal of trial and error. I have not really looked into it in detail. Yeah. It's.

It's a relatively young journal. I think, I think the founder is literally a grad student. Uh,

Benjamin James Kuper-Smith: Okay,

Laura Luebbert: I really hope, I hope I'm not getting this wrong. I have not spent that much time on this journal but there are options. So we are thinking about it again. [01:33:00] Our goal was to get the word out and get peer review.

So, we've achieved those goals. So I don't really see how getting it into

Benjamin James Kuper-Smith: I see, yeah.

Laura Luebbert: adds much except that maybe it will appease. Some like random people who are like, anything that's in a preprint doesn't exist.

Benjamin James Kuper-Smith: Yeah, and I guess the No, I guess yeah, if you have it a preprint then it will be findable, but I guess it's still you know It makes a difference if you get the actual Oh, I guess you'd have to get the actual journals to do something about it for it to like permanently be Visible for people who search the paper or something like

Laura Luebbert: Well, actually. So, I mean, if you put out a preprint, it has a DOI, so it is a permanent record.

Benjamin James Kuper-Smith: Yeah, Yeah, but I mean like if I let's say I want to learn about bumble bumblebees honeybees and I Or bumblebees but let's say I want to learn about honeybees. And I you know find those papers You I mean, there's always a problem, right? Like, how do you find out that there are corrections or retractions or anything like that?

But having it on the actual journal is, I guess, the biggest [01:34:00] step.

Laura Luebbert: exactly. Yeah. I would, I'm really looking forward to hopefully seeing some more responses from the journals. But yeah, that will take a

Benjamin James Kuper-Smith: Yes, especially because I mean, now there was a lot of attention, but like, you know, give it a year. How many people who are then starting off as grad students will know about this, right? I mean, it will

Laura Luebbert: Yes. Yeah, exactly. Yeah. Yeah. No I really hope to see and I think that, if the responses from the authors remain the same then a retraction is absolutely the correct way to go because we simply don't have any data supporting the conclusions made. So I, yeah, I would love to see more responses from the journals and I'm really looking forward to them.

Benjamin James Kuper-Smith: Okay maybe a final question on this point. How has this changed your view now on science and academia? Is it too early to tell still, or?

Laura Luebbert: Well, okay. I think that honestly what [01:35:00] changed my view the most was you were telling me that I can join Academia and I can be just because academia is not perfect, I don't have to, act like a bad actor. So Leo was like. can absolutely continue your path in academia and we can just write about this.

We can just get the word out. We'll just do it. And you don't have to, be like, Oh, I want to be in academia. So I have to be quiet about issues of scientific integrity, yeah. So, so that was very inspiring

Benjamin James Kuper-Smith: He basically did the old Gandhi, be the change you want to see in the world, basically.

Laura Luebbert: yeah. Yeah. Yeah. So we'll see how it, how it works out longterm. But I think, yeah, to just have a very senior, but supportive and optimistic mentor like that, who's not just, [01:36:00] Oh, just, keep your head down, shut up. That was very motivating and important to me, and that really made the biggest impact on how I see science in that, Lior can't be the only

Benjamin James Kuper-Smith: Yeah.

Laura Luebbert: scientist.

I've worked with a lot of great scientists, so I know that he isn't. And so, yeah, that and again, and then, the responses that we got, especially on Twitter from scientists with a lot of concern, a lot of care. And so that was also really amazing to see, and that restored a lot of my hope for science.

I would say. Yeah.

Benjamin James Kuper-Smith: Okay 

yes, at the end of each episode I have three recurring questions that I ask each guest. The first is, what's a book or paper you think more people should read? Famous, not famous, old, new whatever, just something you think people should read.

Laura Luebbert: well, actually I was thinking about that and if you don't mind, do you mind if I instead recommend a book instead of [01:37:00] a journal article?

Benjamin James Kuper-Smith: A book or paper. I mean, it

Laura Luebbert: Oh, book of Andrew. Oh, awesome. Perfect. Yes.

Benjamin James Kuper-Smith: hope you're going to recommend now, what's it, Von den Bienen by Frisch?

Laura Luebbert: I think that's a great book. Yeah.

Benjamin James Kuper-Smith: read it, I don't know. Would you believe it's funny? Yeah, anyway, sorry.

Laura Luebbert: No, I actually, so I've been thinking about question and I wanted to advertise reading books just for fun. So I feel like I've seen it a lot in my colleagues, especially the, when they read a book, it has to fulfill some sort of function. Either it's like an educational book or it's classic literature, which is.

But it doesn't quite suck you in and give you back energy in the way that, your favorite Netflix show does. And so I'm a huge fan of fantasy books. And so, I

Benjamin James Kuper-Smith: Is that fantasy?

Laura Luebbert: a long time ago. Um, No um, but yeah, I just, so one [01:38:00] of the series that I read recently, that is a bit more modern fantasy, not your typical dragons and fairies.

But I thought it was a really nice series was Rivers of London by Ben Aronovich. And it really had this like binge watching effect, you just don't want to put it down in the way that you just want to binge watch a series. And I feel like for some reason people are like, They don't have that for books.

Like they're totally, nobody is like, what am I going to learn from this Netflix series? But for some reason, there's a bit of a stigma against reading books just for fun, or at least that's what I've observed. So I wanted to advertise just a totally fun book. I hope that's okay.

Benjamin James Kuper-Smith: I mean, it's your recommendation. It's not mine. Yeah.

Laura Luebbert: my recommendation. Yeah.

Benjamin James Kuper-Smith: Okay. Yeah, I mean, I'll put that in the description. Everything we talked about in the description. Um, Second question is something you wish you'd learned sooner. It can be [01:39:00] personal, professional, whatever you want. But, if you'd learned that maybe a little bit sooner, then.

Would have helped

Laura Luebbert: Um, I think that in the beginning of my scientific career, I was very worried about doing science and just just generally acting exactly like all the people around me did. And I was very afraid of, speaking up even just, asking questions at the top and things like that, because I was just so worried that I would say something stupid or get something wrong.

And so I think I think I wish I'd known earlier that it pays off to be present and be bold. And it's totally fine. Sometimes you get things wrong. It doesn't always work out. I think at the end of the day you gain so much confidence and you feel so much more comfortable that you are actually able to do better work.

So I wish I had known that earlier, that you can [01:40:00] be Professional and an expert scientist and still also just be yourself probably.

Benjamin James Kuper-Smith: How does one Like is it just trying it and step by step doing more of it? Or how do you go about doing that and learning it like, you know applying it?

Laura Luebbert: Yeah, I think again I think having had a very supportive mentor and that like I had in Lior was very helpful. I would definitely say just try it. I think generally being bold pays off. So, yeah. I think, I wish I would have known that earlier,

Benjamin James Kuper-Smith: okay. And then, so yeah, usually I ask I mean, I guess, yeah so the third question is advised for PhD students or postdoc on that kind of transition period. I don't know, I mean, this is the first time now that I've asked this question for someone who's exactly on that transition

Laura Luebbert: I was, yeah.

Benjamin James Kuper-Smith: So I don't know whether you want to take it exactly like that, or maybe you think you're still too much in that, and you can pick any other kind of academic stage if you want to, [01:41:00] but 

Laura Luebbert: yeah.

Benjamin James Kuper-Smith: you want.

Laura Luebbert: Yeah, I'm literally in the middle of that transition period. So, I would maybe give the advice to not Take on a huge defrauding case in the middle of the transition period. Uh, um,

Benjamin James Kuper-Smith: So it's funny because I also did, I mean, mine isn't that big as yours, but I also did it between, yeah, I mean,

Laura Luebbert: yeah. On the other hand, maybe this is a time

Benjamin James Kuper-Smith: I mean, when I was basically,

Laura Luebbert: the headspace. Yeah, exactly. To take in something else. No, but I, I, the most important advice I would give beginning PhD students is choose your advisors. character very carefully. And I think I really underestimated how much more important that is than their project.

So I mean, at least for me, especially because I was willing to pretty much do any sort of scientific project and get excited about anything. Maybe that's a bit of a [01:42:00] different situation for other people, but from my experience, from my own PhD and what I've also seen in other PhDs is that what is way more important.

then the research that you're actually doing in your PhD is what sort of mentorship are you getting during it and that can really make or break your PhD, I think.

Benjamin James Kuper-Smith: Yeah, I agree. It's uh,

Laura Luebbert: standard advice, but

Benjamin James Kuper-Smith: I think, I mean, like most of this, like something I wish I'd learned It seems like there's about like Usually it's one of ten answers it tends to be. But I think, they're worth repeating. And, I mean, there's so many people I know who or have heard of at least, who start a PhD having basically never met their supervisor before.

Especially sometimes in Germany, you have this kind of like where actually the supervisor is funding you apply to them And they've met them for 10 minutes and then They go there and maybe they've met like some people from the lab But maybe they haven't even done that and [01:43:00] then they go there and everyone Like it's like the entire house is on fire basically because everyone hates each other And just like I think it's worth repeating that yeah, make sure you're spending time with the right people for the next Few years.

Laura Luebbert: And I mean, we say that, but honestly it's easier said than done because yeah, at the end of the day, right? Like how much time do you really get to spend with your supervisor before you start your PhD?

Benjamin James Kuper-Smith: But at

Laura Luebbert: I,

Benjamin James Kuper-Smith: ask about you can, meet the lab, meet the people. I mean, that's usually, I mean, to be honest, I think that's how you usually find out about the character of the supervisor, is by the, just by the supervisor saying, please meet the people in my lab. Right? Because if they don't do that, then that's already a bit of a hmm, interesting.

Laura Luebbert: I agree. Yeah. Yeah. I agree.

Benjamin James Kuper-Smith: Okay.

Laura Luebbert: yeah.

Benjamin James Kuper-Smith: Cool. Uh, Well, thank you very much. Uh, Well, exactly. It'll be a bit

Laura Luebbert: Yeah. Thank you so much. Yeah. We covered a lot of topics, so I hope I wasn't rambling too much, but,

Benjamin James Kuper-Smith: I don't think

Laura Luebbert: This is, That's really cool.

Benjamin James Kuper-Smith: Yeah. [01:44:00] Thanks. Actually I have

Laura Luebbert: Oh

Benjamin James Kuper-Smith: This is basically a cut, but I'll before I stop recording, I thought there's one question I thought I wanted to ask about Catalan and Catalonia, which is a bit specific. So I think I might just include it at the end of this as a fun little afterthought which is basically, so I have a very specific thing I'd like to do and maybe Catalan plays a role in that or maybe not.

And so as someone who has a background. In that maybe you have some sort of interesting insight into it. Which is basically like after, so I want to do this postdoc for a few years. And then I mean, I actually want to do postdoc for a while. So I want to do another postdoc after that. Because I just want to do science and learn stuff.

And not live tech admin That kind of stuff to do. But between the two postdocs one thing I might do which I think is like a little dream project I might do is spend six months to a year Going to a place. I don't speak the language and i'm just like fully immersing myself like learning the language for like several hours a day and but by the end I have this whole like, like really be part of the community and learn the language and everything and so I basically have [01:45:00] this like There's two thoughts I have on this.

One is I want to learn, the classic, like I want to learn a language that's obvious, like that it's very widely usable, like Italian. I mean, Italian is, I guess, mainly in Italy, but Spanish would be an obvious one where if you speak Spanish, then you can speak to an entire continent or, and it's there's a kind of obvious utility in learning these big languages.

However, part of me also thinks that it would be super cool to learn a kind of weird language and weird to define as in you only really use it in one area. I actually got the idea from Basque because I thought like, when I was like

Laura Luebbert: that is a really, yeah. That is a really weird language.

Benjamin James Kuper-Smith: exactly. It's, I mean, I mean, for if people are still listening then Basque is a language isolate.

It's literally not related to any other language on the planet Current or dead. They can't find any connection to anything. And So that might be a step too far because like you have, I mean, I'm assuming they have a lot of Spanish words and it just because a lot of Spanish influences, but like the [01:46:00] language itself, you'd have to learn every single word from scratch.

There's no starting point where, but anyway, so I was wondering like, for example, Catalan might be like a, in between, it's like a unique, small language that Is, I didn't look this up, but I'm assuming it's a romance language. So

Laura Luebbert: It's a Latin language, yeah. It's between Spanish and Italian.

Benjamin James Kuper-Smith: Okay. So yeah, so I both of them, I don't speak either of those languages, but they're intuitive to me because I lots of Latin and English and some French. So it's actually doable anyway, so I'm just curious what you think about this. Like big language versus like small regional language because it seems to me that.

On the one hand you have this obvious utility if you learn this language like spanish but on the other hand that's I don't know. It feels like that misses something about learning a language Just learning something because it's useful And I would imagine if I were to learn something like catalan, it would be much more Obviously not useful outside of the area but like in the area be way more useful and it would be super cool yeah, i'm curious just what you think about [01:47:00] that

Laura Luebbert: So, so obviously I'm going to be very biased because I'm very proud of Catalonia. But actually, I just made a book recommendation based on just sometimes doing things just for

Benjamin James Kuper-Smith: Yeah,

Laura Luebbert: everything has to have a utility. So, It's if you're going to read all of these books that are going to teach you something with the same ferocity that you're going to read the fun books, then great, even better.

You, you get double benefit, but the reality is that most people don't. So if you're gonna try to learn Spanish, but you're going to do it half heartedly because you will like. Not that's not really like it's like just from how you talk about it You seem so excited about learning Catalan and Spanish is yeah be useful.

You know, if that's If that's how you really feel Then you know Learn Catalan

Benjamin James Kuper-Smith: I mean to be just for context like I have no connection to Canada at all, right? It's just an example basically and I think you know It's basically also like i've lived in like northern europe most of my life [01:48:00] If I spend half a year and like going to a different culture might as well be by the beach in the south

Laura Luebbert: It Catalonia is Stunning. It is.

Benjamin James Kuper-Smith: so yeah, I mean that's the thing like I have again like I said there's no I have I like watching football.

So I guess I've seen Barcelona, but that's the extent to my connection to Catalonia. But yeah, I'm just curious because yeah, I guess. Okay. So your advice is just Catalan and Catalonia is awesome to it,

Laura Luebbert: Well, yeah, obviously in my very biased opinion Catalonia is awesome. I would love to spend half a year there. I think it's an amazing place to be. Catalan is a really fun language. I think, qua difficulty, it will be very similar to learning Italian or Spanish. 

Benjamin James Kuper-Smith: Yeah. So it's also like a language. I mean, again, like I had Latin for six years in school English and a bit of French. So I think I could learn it very well in Catalan. In a year or half a year, if I ever, if I, again that's like the thing I do in that time. Right. Learning the [01:49:00] language.

So like where, whereas, Basque would be like probably not as fluid after, after any given time.

Laura Luebbert: And Catalan people are very appreciative of. More people trying to

Benjamin James Kuper-Smith: Oh, that's true. They'll love that. Yeah. Yeah. That's like a political statement almost. If I learn Catalan, I'm like making a statement. I'm not even aware of what the consequences are.

Laura Luebbert: Yeah, I mean, yeah obviously if you're looking for utility, if you're looking for a long term impact on your life, Spanish will be much more useful. You never know, Barcelona is a great city. 

Benjamin James Kuper-Smith: Yeah, I actually found out that even did you know in which did you know that they also speak Catalan in a non Spanish or French speaking country

Laura Luebbert: no, I didn't where,

Benjamin James Kuper-Smith: The tiny area in Sardinia, so Italy

Laura Luebbert: Oh yeah. I mean, I'm not surprised. Yeah.

Benjamin James Kuper-Smith: There's just like one small coastal area where they speak apparently some [01:50:00] Catalan

Laura Luebbert: my Italian friends and if they speak Italian and I speak Catalan we do understand each other and that's not necessarily true with my Spanish friends. So, yeah, it is oddly, I don't know if this is actually based on if you analyze the language, you get the same result but just, I don't know.

Based on anecdotal data, it seems to be closer to Italian than Spanish.

Benjamin James Kuper-Smith: Oh, that's interesting. Yeah, I guess geographically I wouldn't have guessed that.

Laura Luebbert: No, geographically, it doesn't make any sense at all. So yeah, I don't know if this is true for real or my Spanish friends were just, not trying hard enough.

Benjamin James Kuper-Smith: Yeah, they were like, I'm not gonna even try.

Laura Luebbert: Yeah. But yeah, in my very biased opinion, course, I think you should learn Catalan.

Benjamin James Kuper-Smith: Yeah, I think something vaguely Italian or Spanish would make the most sense in terms of me actually learning it very quickly. [01:51:00] And the language I mean, it's weird with Spanish. I feel like I can understand a lot of it, especially considering I've never learned a word of Spanish.

It's really shocking to me how, at least on first sight, easy it is. Literally two days ago, there was a guy here who needed directions and his English was terrible. So it's which language do you speak? He's like Spanish. I'm like, yeah, let's speak Spanish. And then basically he went through like Italian or Portuguese.

No. And then we just, I just spoke some French and he spoke some Spanish and I managed to give him directions. It worked out

Laura Luebbert: Oh, it works out. No, I mean, within the Latin languages there's a lot of overlap, but you make it work.

Benjamin James Kuper-Smith: Yeah.

Laura Luebbert: so, but that's really cool. We can make the next podcast in Catalan.

Benjamin James Kuper-Smith: Okay. That's it. Yeah. Yeah, exactly. Yeah, that's okay. That's gonna take a while. Okay. Anyway, thanks.