Jon Krohn: 00:06
This is Five-Minute Friday on NotebookLM.
00:27
Welcome back to the Super Data Science Podcast. I’m your host, Jon Krohn. Let’s start things off with an Apple Podcast review that came in since our most recent Friday episode. This generous five-star review is from Douglas A. Smith, who’s principal of AI and data science at Ampersand Consulting. In his review, he says that he truly appreciates the time and effort that the whole team puts into creating these podcasts, and he says that I, that Jon, goes above and beyond with questioning and sharing his insights, exceptional job at their selection of knowledgeable guests, like Dr. Julia Silge. That was episode number 817. And yeah, it’s interesting that we have had two Apple podcast reviews specifically mentioning Julia Silge, so that’s awesome. I don’t think we’d had any guests mentioned in the reviews for a couple of years, so that must have been a really special episode.
01:22
And then finally, Doug says, “I recommend this podcast most often when asked what resources I use to stay up to date.” Nice. Thanks to everyone who’s given us recent ratings or feedback on whatever podcasting platform or YouTube. Really appreciate that. Apple Podcast reviews are especially helpful to us because you can provide written feedback and I keep a close eye on those, so if you leave one, I’ll be sure to read it on air like I did in today’s episode just now.
01:54
All right, now onto today’s episode topic, which is on Google’s newly released and frankly sensational product, NotebookLM. We’ve of course got a link for you in the show notes so you can try it out for yourself. All you need is a Google login, which is as easy as having a Gmail account and is for use of NotebookLM is likewise totally free.
02:15
Now, this is a tool that anyone can use. This isn’t just for hands-on practitioners like data scientists or machine learning engineers. The user interface for NotebookLM is simple click and point and intuitive. You provide NotebookLM with resources, so you can upload documents or you can provide links to websites. You can provide up to 50 resources, which is a pretty vast amount, and it’s not obvious to me that there’s a file size limit for the resources you upload. But inspired by Dr. Brett Tully, who was our guest in episode number 533, he messaged me a NotebookLM generated podcast conversation about his doctoral dissertation. And so I similarly uploaded my 12 megabyte 200-page doctoral dissertation into NotebookLM. I did actually want to upload my book, Deep Learning Illustrated, but the manuscript PDF I had immediately to hand is 236 megabytes and it turns out NotebookLM has a limit of 200 megabytes per document.
03:16
Oh, well. So there you go. I’m learning the limits of this tool. But still, you can have up to fifty 200 megabyte documents apparently uploaded into NotebookLM. So all I’ve described so far is that you can easily upload documents into this thing. That doesn’t sound very interesting. We’ve had note-taking tools for forever. What’s cool about NotebookLM is that it allows you to converse with whatever resources you upload. It even suggests starter questions based on the specific resources that you upload to get your chatbot style conversation with all of your resources going. So that’s cool. So you can have that kind of ChatGPT style, or Google would probably prefer I say Gemini style, style of conversation with whatever resources are in there.
04:04
In addition, NotebookLM is also pre-configured to automatically generate helpful documents from all the resources you provide. So things like a table of contents, a timeline of events, a briefing doc, a study guide, a frequently asked set of questions. So clearly, NotebookLM is handy for all kinds of summarization or interactive conversation that might be invaluable for understanding or analyzing vast amounts of information across, again, all of those resources, up to 50 resources at once.
04:36
The most impressive aspect of NotebookLM, however, and the aspect of NotebookLM that has generated all of the buzz since it came out recently, is the podcast episodes that it can create from your resources. Within a few minutes, NotebookLM created an 11-minute podcast episode about my 200-page doctoral dissertation, and 99.9% of it, to my ear, was indistinguishable from a real podcast. On top of that, the podcast conversation between the two hosts is extremely compelling. I think my PhD is super, super dry. But listening to this conversation, the NotebookLM podcast hosts managed to make it seem interesting and practical. These digital hosts also cleverly use analogies that I never had in my thesis itself to make the complex concepts that I cover in my PhD thesis easier for a lay audience to understand.
05:36
The hype around NotebookLM is no joke. And to prove it to you, here’s the entirety of the 11-minute podcast episode that NotebookLM generated based on my PhD note, that NotebookLM didn’t output a transcript. It actually rendered this audio with just a few minutes of processing. And I won’t be offended if you skip to the end of this episode, or you just kind of pick and choose parts of the next 11 minutes to listen to, but based on how incredible the whole thing was, and to give you a sense of the way that these digital hosts set up the episode, go through the content with all those analogies and then summarize the episode, I just decided to leave the whole thing in in its entirety for you to check out if you want to.
06:24
And if you do, you’ll learn a bit about the genetic basis of anxiety, which was the focus of my PhD dissertation. So I developed and applied machine learning models for analyzing the vast amounts of genetic data that I could find associated with anxiety. And the reason, in case you’re curious why I picked anxiety as an episode topic, is a huge amount of people around the world are adversely impacted by it. And so I thought that this would be a highly impactful PhD project. All right, I’ll stop giving the backstory. Here is the podcast episode.
Digital Host 1: 07:04
Ever feel like anxiety is totally wired into us? No matter how much we try to meditate or do those breathing exercises, it’s always kind of lurking there. Well, today we’re going deep into where that feeling comes from, the science of it. We’re going to explore the genetics behind anxiety, which is super interesting.
Digital Host 2: 07:24
Yeah, it’s wild. We always talk about anxiety, like it’s all in our heads, right?
Digital Host 1: 07:26
Right.
Digital Host 2: 07:27
But it’s actually way more complicated than that. What if I told you we could understand human anxiety better by studying mice?
Digital Host 1: 07:34
Mice? Okay, hold on. I think I know what you’re getting at, but how can studying mice, those little lab ones, actually tell us anything about human anxiety? It just seems so complex.
Digital Host 2: 07:45
Well, that’s the cool part about science. It’s always showing us these connections we’d never expect. So get this. Mice and humans actually share a surprising amount of their genes, their genetic material, and the way they react to stress, like their behavior and even their bodies, it’s shockingly similar to how we react.
Digital Host 1: 08:03
So it’s like looking at a simpler version, a blueprint almost, but still has a lot of the same instructions as ours.
Digital Host 2: 08:09
Exactly. And this PhD thesis we’re diving into today, it’s by Jonathan Krohn. He uses those similarities to his advantage. He uses mice to figure out the genetics of anxiety. And let me tell you, his findings are really going to make you rethink what you thought you knew about anxiety.
Digital Host 1: 08:25
Okay. You’ve definitely got my attention. So break it down for us. What’s Krohn’s deal, what makes his research so special?
Digital Host 2: 08:31
So Krohn, he goes deep into this thing called genetic architecture, which might sound kind of intimidating.
Digital Host 1: 08:37
Yeah, a little bit.
Digital Host 2: 08:38
But it’s actually pretty easy to understand. Think of it like baking a cake. It’s not just one ingredient that makes the cake taste amazing. It’s all of them together, the flour, the sugar, the eggs, all working together to make that final delicious product.
Digital Host 1: 08:54
So you’re saying anxiety is like that cake then, like a bunch of different genetic ingredients all mixed together that create how we experience anxiety?
Digital Host 2: 09:03
Exactly. Now, to make things even more interesting, Krohn uses a special kind of mouse for his research. They’re called heterogeneous stock mice, or HS mice for short.
Digital Host 1: 09:12
HS mice.
Digital Host 2: 09:13
And these aren’t your average lab mice. They’ve got a way bigger range of genetic diversity, which is like having a stronger magnifying glass to see those tiny genetic differences that might be playing a role in anxiety.
Digital Host 1: 09:25
So it’s like he’s working with a way more complete recipe book for anxiety genes.
Digital Host 2: 09:29
You got it. And because he’s using this diverse group of mice, he’s finding some pretty amazing stuff about how anxiety works. He found that the way genes affect anxiety can be totally different in males and females. It’s called gene-by-sex interactions.
Digital Host 1: 09:43
Hold up. Are you serious? So my genes could be expressing anxiety differently than yours just because I’m female. That’s insane.
Digital Host 2: 09:50
It’s a total game changer. It shows just how important it is to think about those sex differences when we’re trying to understand something as complex as anxiety. But it gets even crazier. Krohn’s work also found that our environment can actually mess with how our genes influence anxiety.
Digital Host 1: 10:06
Okay, now my mind is officially blown. Are you saying that where we live or even just our experiences can actually change how our genes work when it comes to anxiety? How is that even possible?
Digital Host 2: 10:17
Well, get this. Krohn found that even something as small as who was handling the mice in the lab, like the specific experimenter, could actually affect how anxious the mice got. It’s like if the experimenter was a little rougher or seemed stressed themselves, it actually interacted with certain genes in the mice to make them more or less anxious.
Digital Host 1: 10:37
That’s wild. It really makes you realize how connected our genes and our environment are. It’s not just nature or nurture. They’re both working together all the time.
Digital Host 2: 10:46
Exactly. And this is where Krohn’s work takes a really cool turn. He starts looking at how genes work together, kind of like departments in a company all working on a big project. He uses something called gene co-expression networks to map out these interactions and figure out which groups of genes are firing together in response to anxiety.
Digital Host 1: 11:06
So instead of just looking at individual genes, he’s mapping out this whole communication network between them. That’s next level.
Digital Host 2: 11:12
It is, and it’s leading to some seriously cool discoveries. But to really get the impact of Krohn’s work, we got to talk about this secret weapon he developed, a statistical technique called SPIV.
Digital Host 1: 11:22
Okay, you’ve definitely piqued my interest. What in the world is SPIV and what makes it so groundbreaking for understanding anxiety? You totally got me hanging. SPIV, this statistical tool Krohn made, it’s like a secret weapon for understanding anxiety. [inaudible 00:11:40] give me the details.
Digital Host 2: 11:41
All right, all right. So imagine trying to solve a crime, right? It’s not enough to just find clues that might be related to it. You need that smoking gun, the one piece of evidence that directly connects the suspect, right? Well, that’s kind of what SPIV does, but for genetics.
Digital Host 1: 11:54
So instead of just finding genes that pop up around anxiety, SPIV actually helps find the ones that are the masterminds behind it.
Digital Host 2: 12:01
Exactly. It cuts through all the noise and finds those genes that actually cause changes in anxiety. Other methods, they have a harder time doing that. But SPIV, it’s like this high-powered magnifying glass, helps researchers zero in on those crucial genetic triggers.
Digital Host 1: 12:17
That’s seriously impressive. We can finally see those genetic codes way more clearly now. So what did Krohn find using SPIV? Any smoking gun genes for anxiety?
Digital Host 2: 12:27
Well, he did find a few that look super promising, genes that seem to be directly involved in how those mice behave when they’re anxious. But got to be real. This is all still pretty new research, and we got to do more to make sure these findings hold true for humans too. But even still, the potential here is huge.
Digital Host 1: 12:43
It’s like we’re on the edge of understanding anxiety on this whole other level, down to its molecules and stuff, right?
Digital Host 2: 12:48
Absolutely. And that opens up so many doors, especially when it comes to developing new treatments for anxiety.
Digital Host 1: 12:56
Okay, now this is what I find really interesting. What could those treatments look like? Are we talking gene therapy, new meds, something else entirely?
Digital Host 2: 13:05
Imagine we could identify people who are genetically more likely to have anxiety, and we can do it early on. Then we could intervene early. We could do preventative stuff, personalized therapy, maybe even gene therapies designed for their specific genes. Instead of just dealing with the symptoms, we could tackle the root causes of anxiety.
Digital Host 1: 13:24
Wow, that would be revolutionary, like a whole new era of mental health care where we use genetics to create personalized treatments. It’s really mind-blowing. But I got to ask, what about the ethics of all of this? This is some powerful information and we got to be careful with it, right?
Digital Host 2: 13:40
You’re totally right. Knowing someone might be genetically predisposed to anxiety could be a huge help. But yeah, there are big questions about privacy, about discrimination, even about how this knowledge might be used or misused. It’s a big deal.
Digital Host 1: 13:55
So it’s not just about the scientific discoveries, but about making sure we’re using those discoveries in the right way.
Digital Host 2: 14:01
Hundred percent. We need to be having these conversations now, especially as this research keeps developing. We can’t ignore those potential consequences. But at the same time, we can’t forget about how much this could help people, how many lives it could improve. It’s a delicate balance.
Digital Host 1: 14:16
Absolutely. And speaking of helping people, Krohn’s work also has some really interesting insights into things we can do right now to manage anxiety, even without fancy genetic tests or anything like that.
Digital Host 2: 14:27
Oh yeah, for sure. Remember those gene environment interactions we talked about?
Digital Host 1: 14:30
Yeah.
Digital Host 2: 14:31
Well, they’re not just some cool scientific thing. They remind us that we’ve got more control than we think.
Digital Host 1: 14:36
Okay. Now you’ve got me really curious. What are some of those empowering takeaways from Krohn’s work? What can we actually do?
Digital Host 2: 14:43
Well, his research really highlights how big of a deal our environment is. It can actually change how our genes express themselves. Things like managing stress, exercising, sleeping well, even our relationships, those can all affect how vulnerable we are to anxiety. And get this, it’s regardless of our genes.
Digital Host 1: 15:01
So we might not be able to change our DNA, but we can change how it acts just by making different choices in our lives.
Digital Host 2: 15:07
You got it. It’s not about being stuck with whatever genes we’re given. It’s about recognizing we have the power to steer the ship.
Digital Host 1: 15:15
I love that analogy. It’s a message of hope. We’re not just puppets to our genes. But let’s dive into those lifestyle factors a bit more. What kind of impact are we talking about here? Anything surprising? So we’re talking about actually changing how our genes affect our anxiety just by making different choices, right? I’m hooked. Tell me more. How much of a difference can these lifestyle changes really make?
Digital Host 2: 15:36
Well, Krohn’s work really shows how important it is to tackle anxiety from all angles, not just one. He found that if you do things to manage stress, like meditation or mindfulness, it can actually change how some of those stress response genes actually work.
Digital Host 1: 15:51
Wait, so you’re telling me that just by meditating, we could be changing our brains to be less freaked out by stress? That’s mind-blowing.
Digital Host 2: 15:59
It is pretty amazing. Our brains are way more adaptable than we give them credit for. But it’s not just about meditation either. Krohn found that physical activity can make a big difference too.
Digital Host 1: 16:09
Okay, yeah, exercise is good for you, I get it, but what’s it got to do with genes?
Digital Host 2: 16:14
So working out, it doesn’t just make us feel good in the moment. It can actually change how our genes express themselves in a way that makes us more resilient to anxiety over time. Pretty cool, right?
Digital Host 1: 16:26
So hitting the gym, it’s not just good for our bodies, it’s good for our genes. I’m totally down with that.
Digital Host 2: 16:32
Right. And it’s not just exercise either. Even how well we sleep can affect our genes. Krohn’s research found that if we don’t get enough sleep, if we’re constantly sleep-deprived, it can actually mess with our gene expression in a way that makes us way more vulnerable to anxiety.
Digital Host 1: 16:46
That makes sense, actually. We all know how cranky and on edge we feel after a bad night’s sleep. I can only imagine what that’s doing to us on a genetic level. Yikes.
Digital Host 2: 16:55
Yeah, it’s a big deal. It’s a good reminder that we really got to prioritize getting enough sleep. And one more thing, Krohn’s work highlights something we don’t always think about enough, our connections with others. Having strong relationships and feeling like you belong, like you’ve got a good support system, that can actually have a positive impact on our genes too. It’s like a shield against anxiety.
Digital Host 1: 17:17
So hanging out with loved ones, it’s not just fun. It’s like genetic self-care. I love that.
Digital Host 2: 17:21
Right. And all of this just goes to show that our DNA doesn’t have the final say. We have a real impact on how our genes express themselves, and it starts with the choices we make every day.
Digital Host 1: 17:31
It’s empowering, isn’t it, knowing we have that power. So as we wrap up this awesome deep dive into Jonathan Krohn’s research, what’s the one thing you hope listeners take away from it all?
Digital Host 2: 17:41
I think the biggest takeaway is that anxiety is complicated. It’s got all these layers and it’s shaped by both our genes and our experiences. Our genes are definitely part of it, but they’re not the whole story. We can make choices that affect how we experience anxiety. Like we can choose how we react to things, how we take care of ourselves, and even who we surround ourselves with. It all makes a difference.
Digital Host 1: 18:03
Well said. And for everyone listening, if you’re interested in learning even more about the link between anxiety and genetics, check out our show notes for links to Jonathan Krohn’s work and some other cool resources. And remember, your genes might be the blueprint, but you have the power to create your own masterpiece. Until next time.
Jon Krohn: 18:21
All right, this is Jon Krohn again, or Jonathan Krohn again, as the NotebookLM podcast hosts called me. I guess we’re past time for me to start worrying about my podcast host job security. That really was stunning. I guess another sign that AGI is near.
18:45
All right. That’s it for today’s episode. If you enjoyed today’s episode or know someone who might, consider sharing this episode with them, leave a review of the show on your favorite podcasting platform, tag me in a LinkedIn or Twitter post with your thoughts. I’ll respond to those. I’d love to hear your thoughts on the NotebookLM generated content in today’s episode. Did it blow your mind too? Were there aspects that could improve? Yeah, and if you aren’t already, be sure to subscribe to the show. Most importantly, however, we hope you’ll just keep on listening. Until next time, keep on rocking it out there. And I’m looking forward to another round of the Super Data Science Podcast with you very soon.
