Computing Toolbox

Key Takeaways

[Music] [Applause] [Music] welcome to another episode of dat to skeptic animal intelligence today is a first I've turned over the interview seat to Becky to interview an author about a particular book for today's episode Becky can you give us the details on that so I got to sit down with madlin and she's co-author for computing skills for biologist a toolbox this book is really great so it was actually Kyle that found this and asked me to it was you you found it and had me look at it and I started going through the tutorials and basically this is like a cookbook of computing skills that not only biologists but I'd say any natural scientist could use so I've started going through it I've been going through the Unix and shell scripting because automating things will save me a lot of time but Madeline and I talked about how the book is useful if you're a professor you can use it for your class or if you're a student like me you can just get it and start working through everything it has a great companion website and then she also had really great advice for students that might be transitioning or considering transitioning into industry and how these Computing and data analysis skills can really help you there well both great topics I look forward to learning more let's jump right into the interview I'm Main vilus and I am a trained biologist I studied the planned micro uh interactions in my previous life and now I'm actually working in finance I did a in liquidity and funding something entirely else and now I'm analyzing climate risk in the financial world so how did that transition happen going from a biologist into industry I think it started with my my reluctance actually to leave Academia I was really a full-blood researcher I came to the US to do a postto at the University of Chicago Academia is hard so involves a lot of failure and keep like lots of work to keep trying and trying solving challenging questions and I was really behind it I really wanted to make that work so it took me a while to accept that maybe Academia is not going to work out so and then I wrote a book to stay a little longer in Academia it took another stint at a small scientific publisher so I did a little bit of data science there marketing analytics I was a science writer and editor there for a while then Co hit and it's really this transition into Finance came with Co everybody locked down and I was actually laid off by the P scientific publisher which obviously sucked for the ego but it was a very short recovery phase and um friend of mine who was already working at the Bank of Montreal at the time so beo short Bank of Montreal was saying like oh you know bimo treats their people well so why don't you come to bimo and I thought this actually really bizarre because again I was a trained biologist I had obviously the experience from the book and as a science writer but had never done anything remotely like associated with the financial world so it's why would they even hire me so well I interviewed and was very relaxed because obviously I was there just to gain the interview experience they wanted to hire me and that was in the liquidity and funding group so and I was like okay well they seem to need quantitative skills and they tell me that they will teach me the rest so why don't I give this a try so that's how I ended up in finance if anybody would have told me that for years prior I would have or or just like you know when I started the job with someone would have predicted four months ahead like oh you will be working in finance I would have declared them totally crazy so it sounds like you had some transferable skills that they really liked to bring you on so what do you think transferred really well from your science background sounds like data analysis is probably one of but what else so as a biologist it's very easy to produce data but it's hard to get to the conclusions so and that's actually now I'm circling like or migrating a little bit back and forth here speaking about the book and the the transition but it's it's really connected obviously that's like my career path producing data is easy analyzing and concluding is hard so and I realized during my time as a life science researcher that I really had to learn computational skills I did this initially really autodidactic it was like a little bit of so I already programmed an r that was kind of like what I what I already had under my belt but I realized I needed a little bit of shell scripting because if you wanted to do anything on a server if you wanted to do something in like parallel Computing like it needed some some shell wrapper so I had no idea right and whenever I went to the library I could get a book with a thousand pages in Unix but that's not what really what I what I wanted what I what I had time for and that's actually what gave the idea to the book so I can get into depth a little bit more let me get back to your questions of the transferable skills right these analytic skills definitely like transferred really well and that's in the end what they hired me for that like the critical thinking skills being able to analyze data but surprisingly also writing as a scientist it's something that we just have to do we have to publish our results so when I only re realized that in Industry the ability to communicate and write technical technical writing as much as like just what you did the meth so to speak the methodology section was really a valuable skill just the ability to know how to site is already something that you know in the academic world obviously everybody knows that but within like the General Industry setting that's not a given so these were all skills that that were very very valuable so what we're going to talk about today is the book Computing skills for biologists a toolbox by Stefano alisina and mlin Willams so can you tell us a little bit about how the book started and your co-author it was really born out of the need that I had myself at the time right I realized I needed computational skills in order to analyze data and I picked a lot of up here and there but like at the time all these like tutorials that are available today like data camp and like I mean you name it right there's so much out there 15 years ago that didn't exist particularly it didn't exist with a focus on biological sciences what is the kind of tool set that you need as a as a life science researcher in order to be successful and Stefano was actually the first one at the University of Chicago like that I encounter during my career scientific career that had created a course and I approached him at the time to say like you need to publish this as a book this is so good I had so many hours of trying to just fing figuring it out under my belt it was like this is like so structured is so to the point and like this this needs to be brought to a broader audience and apparently I was very convincing and we started writing the book together Stephano is now the chair of ecology and evolution at the University of Chicago and initially I had thought like oh you know this will be very easy because we basically just use his already existing lecture material and that's the book will be easy but he said no no no no we're going to do this properly we'll write it from scratch that sounds about right I think some of my best projects I've started out going this will be easy and then it wasn't so I noticed because I have a copy of the book here and it's dedicated to all the biologists who think they can't code and I might throw myself in that camp and I kind of want to learn it so what inspired that dedication that's really again from my own experience right because you kind of have the feeling oh you shouldn't you should just know more in order to analyze your data you so you're not riant on someone else for me coding initially learning how to code was hard because again I tried it on my own I didn't have like any kind of structured instructions it was always kind of like a side thing that I had to do but I felt like oh I also don't really have the time actually to dedicate uh to learning to code and I guess I don't know maybe also the experience uh from a female perspective that you always think like right the impostor syndrome is like in your own way and like it's it's just hard and everybody else can do it better but this is exactly what the book tries to address giving you a foundation like an entry point to a topic and just give you enough knowledge so you can explore and learn a lot then you can advance from there right if you've never written a single line of python you do not want to go like through this 1,000 page book this is really just a primer in the in Computing skills for biologists so you can get started and I think like if you work through all the materials and you're able to solve study questions that we provide I think this is a very good foundation that makes sense to me and I feel like I'm at the beginning of that Journey myself in 2023 just 10 vulnerabilities accounted for over half of the incidents responded to by Arctic Wolf incident response wouldn't you like to know how to take them off the table and make life more difficult for cyber criminals that's just one of the essential insights you'll find in the Arctic Wolf lab's 2024 threats report authored by their elite team of security researchers data scientists and security development engineers and backed by the data gain from trillions of weekly observations within thousands of unique environments this report offers expert analysis into attack types root causes top vulnerabilities ttps and more discover the attack vectors behind nearly half of all successful cyber crimes why Ransom demands climb 20% from 2023 and find out why 2024 will be an especially volatile year for cyber security I got my copy you should get yours by going to Arctic wolf.com datas skeptic that's Arctic wolf.com sdas skeptic so in the book it opens with philosophies and goals of computing and biology and they seem like these are just great for Science and I'll just list them really quick automation reproducibility openness Simplicity correctness and then science as software development and we don't have to go through every single one of those but these seem like really good goals for any science but at the same time I can see Computing is integral to actually making these things happen so how did you and Stefano come up with these this is I think really driven by Stefanos experience he had previously before he started on his academic career he had worked much more intensely really as a software developer he as a principal investigator he had noticed lots of people can start to learn but then writing code that is robust that is correct that was like less of kind of like a focus in the beginning and I I mean there is a struggle there if you're learning to code you cannot do everything at once and of course at the beginning your code will be awful but that's why really the book as we said starts with a foundation but we want to get you to a point that you know what is a debugger that is not something that usually if you're self-taught that's something that you kind of want to get into there's a chapter on writing good code how to name your variables commenting it's very like maybe straightforward and obvious advice but as a beginner often we we noticed that people are struggling and I I did it because I was the beginner right and he had seen this a lot within his courses that he taught that makes sense if you're self-taught that it might be a little bit harder to learn some of those best practices without like a mentor so in the book there's a chapter on R and as a biology graduate student we get a lot of R hammered into us but there's so many other cool tools and I was wondering if you have a few favorites that the book covers that you find especially useful I think this has really shifted r as the language that is the most fluent for me as well and at the time I had very little python exposure but if you're leaving the biological sciences then obviously python is the much more ubiquitous language and now with like lots of machine learning well there's implementations and and R it's usually if you want to be at the Forefront python is your language but then one of the things that I really learned while writing the book for instance was Version Control and Version Control you might initially not quite see why that is even necessary coming from this autodidactic perspective but once you get into and you force yourself to spend the few minutes of like properly documenting and tracking your code it makes such a huge difference in terms of reproducibility right you can go back and say like okay I published this paper but this obviously like this was maybe a year ago and there has been so much code written in between right this has changed this has the code has been modified but you can go back and say like okay what was my version that I exactly submitted at the time maybe with that you might also have a folder that has all your different code vers versions if you are that structured and organized right and you might be able to go back and say like oh what did I what did I submit but then you have this we've all seen this right you have this final final version 3.6 and that's exactly what you can avoid right by using such a tool as Version Control so that is definitely that has most influenced my working my coding at the time in retrospect the chapter that was really very important for my further career path is SQL I had much SQL experience at the time I now use this daily whether my job now right or also like my previous team in the financial world what other chapter I think definitely the writing could code resonated with me and also actually the latch was something that I had maybe not I had some exposure and it was basically my tool to make my resume stand out and make it really pretty but here's the irony obviously ltic helped a lot in writing the book and I not I know not everybody might be writing a book and maybe like that's not the what I'm saying why you need to learn latch but definitely it helps if you're just like a comprehensive document such as your thesis even your master sees this any kind of publication definitely benefits from latake and often it's like this where you just need to invest a few hours to get started it's not about like right away writing your Macros you can really have a little bit of effort and have a big return right away I have found that other biology students that I've worked with if they're not already in computational biology they don't necessarily even know what latex is so what is it so it's a markup language if you and your audience are familiar with HTML so you have the content and the formatting so to say separated from each other right you say like this is my heading but then you have these additional commands around it that say like this is the title and that's why it's bold right that is kind of like the the premise of the separation of content and formatting that's maybe in the most simplest terms it's a markup language and really that understanding then helped for me I mean using ltic like I you can use that in your R studio right you can then use if you have a little bit of ltic knowledge that might be helpful to write out pulling reports of your analyses out of our studio right if you can knit it so there's specific packages there that you can utilize and just having a little bit of foundational understanding there with lch helps you with these other tools so that's also why we shows to have so many different tools knowing a little bit of all of each like having a foundation allows you really to complement having this toolbox right that's the subtitle of the book it's a toolbox so and you can pick and choose and really be prepared to using the best matching tool to your task because that's what I found myself actually always doing because I knew R I try to do everything in R if you have this Hammer everything you want to slap like a nail but that's not actually often the case right you need something more specific like where I said like python might be like advantageous versus r or knowing actual rudimentary SQL is much more helpful and trying to put everything in a gigantic depler chain there's a couple tools that I read about that I'd like to chat quickly about because I know there's students out there like I think half the audience or more will be like yes I know markup language that's great and then we're reaching some ecologists that are out in the woods and getting dirty and they're going to be like what's a markup language or what's this or what's that so one thing I have found useful and I'm I'm still at the beginning of the book is going through shell scripting M and getting Unix set up and I don't know how many as a animal behavior researcher how many hours I've wasted moving video files over one at a time because they're such large files like processes that you could automate instead of you just sitting there so could you tell us a little bit about the shell scripting uh section of the book and why you chose to put it in there why is it useful as you describe it's really if you want to process things in parallel at a time being able to work on a server where maybe your interface isn't a graphical one that's when it's really helpful or being able to schedule scripts right and that's again where these tools really integrate with each other so you can use for instance a shell script to schedule your python or or your R script orever whatever it one should be so I mostly used it at the time for working yeah on a remote server so I knew how to log in how can I create a folder on this remote machine if there's not a graphical interface that's really all that I needed but you can get like the the chapter will show you that you can actually do very powerful things how can I search for a file by name under certain criteria how can I search for text uh within all my flat text files and how to get started is really a little bit as like it's you're based on what operating system you're on so if someone working already with the in the Linux system like is very familiar there's probably not much explaining to do if you're on a Windows computer Windows really tries to hide all the kind of things under the hood so then but there's the command line tool still available to you or if you're installing tools like get bash that gives you a rudimentary terminal then you can also get a little bit into these more command line tools and where you actually write words that are the commands instead of clicking and dropping something which where usually like as a Windows user particular you might be more used to if you're on mecho X that's also usually very accessible to you you have a terminal available so and then right it's also nice because you're independent of like a graphical interface even if you're saying your Version Control you don't need any kind of graphical inter like software in order to manipulate your Version Control so we actually that's the case that we that's the the way we we demonstrate Version Control on the terminal so again it's kind of the you need a little bit just this little bit of shell scripting in order to you really get a lot of use out of these other tools and not be lock locked into a particular software suite that makes sense and I went through the process of getting get bash because I'm on a Windows machine and was able to copy all the directories we've talked a bit about what students might need to get started so you mentioned all three operating systems do they need a particularly powerful computer they they don't really I myself I'm probably on a Mac that is I don't know 13 years old and it still serves me serves me well I mean obviously I want to say like if you're doing powerful Computing then obviously it's not enough but I'm saying I would think if you're picking up this book then you're not likely already need a lot of processing power so the book is really written actually really with the field scientists with the ecolog field ecologist in mind for people who are not in a maybe in a computational biology program because they will have their dedicated curriculum this is really to the breath of like either lab scientists or behavioral animal ecologist right that's really the target audience you don't need a super powerful computer to go through the book but your own science is your business do they need to purchase any software no that's a great Point as well everything in the book that we cover is open source and freely avail available this comes with public domain licenses so you don't need to purchase anything I mean the book is relatively cheap as it comes for text to textbooks I want to say and that was one of the choices that we had we don't have any color Color Prints any color figures in the book itself to the price affordable for grad students Master students but then it's the the really funny aspect that we have a chapter on visualization that doesn't have any graphics so it still sense like uh if you pick up the chapter work through it that we have materials on the we on the website and also you generate the graphics while you go along right but that always like is kind of like a funny side note I don't know if you recognize that the visualization chapter doesn't have much doesn't have much in terms of figures I think the book in total has like five figures as a student I definitely appreciate any efforts to to make things a little bit cheaper okay so I think the only other thing that students would need is a text editor that they like on their computer and then that covers everything to get started with the book absolutely I would recommend if you're starting to code having a dedicated where I want to go here the syntax highlighting is really helpful you don't actually you can write python code in notepad you don't need any specific software there either but definitely there's dedicated software many free of charge that you can depending on your operating system and there people you will find they will passionately argue for one over the other I will not go into this discussion I really don't want to have this in the way of actually getting started I find this like often even if you choose the mo the suboptimal tool and you know you don't want to learn them right away that's totally okay you can switch at a later time so don't be blocked by like picking the most optimal editor interface so that's really beside the point so Okay so we've talked about everything you need to get started and you've mentioned that a website so what can either professors or students that are picking up this book what kinds of things can they find on the website that support the book so the website is helpful if you pick up the book and just to set up your environment you will find some instructions both within the book but it relies also on the website there's lots of instructions on how to set up to get started for the particular chapters and I want to point this out actually I don't know I haven't really mentioned that you can read most of the chapters independently you do not think of that you need to go through the book cover to cover you can start you can actually stand alone just work through the SQL chapter other chapters rely a little bit on each other I mentioned right the visualization chapter for instance is based on the r language right if you don't have some basic R understanding that this will be a rough one and the writing good code shows example in Python but I think you can even read read through the uh writing good code chapter without having much python exposure Version Control is entirely independent uh the shell scripting and is entirely independent latch is independent right so you can really also jump back and forth based on your priorities and the website enables that again to show you here's what you particularly need just to to work through that you're not dependent on the website but it will have some useful information we also Target a little bit of mostly right it's a tech it's a textbook it's definitely meant also for the autodidact who like wants some structured to help them along like on their Learning Journey but it's also a book that we really try to bring to the attention to instructors who might base their course based off on the course right again that's the function of a textbook and that's like I revealed this earlier that's really how the book came to existence that because it was based on classes that Stefano taught at the University of Chicago this still amazes me he taught the entire book in 12 weeks so each chapter so I think the python was two weeks but those pure student poor students I read you get you like you get your challenge when you sign up at the University of Chicago but they worked through the entire book in 12 weeks which I still find amazing well that's a pretty good Testament that it can be done I also want to mention that we really do monitor the inbox there so if anybody wants to get in touch if they have questions I there's multiple positions where you can reach out to the author like contact forms and if you fill one and because you get stuck somewhere we will get back to you sometimes it might take a day or two but we're very dedicated so and the other one there's a box to report errors so far there's one and I'm still proud of that like like multiple years later so we did a very good job like proof reading that's that's actually very impressive I think I would struggle to do that with a book we've talked a bit about the different tools you go over in the book how you got started and why you wanted these skills and I wanted to Circle back and just talk a little bit about what happens to scientists re data analysts if you don't have these upand cominging skills well I would argue that it's hard to be a data analyst if you don't have some computational knowledge again the book is a great foundation for continuing on your way in the data related fields you will notice if you head over to my LinkedIn profile right I never called myself a data scientist I think like that's maybe also because I'm married to a like a machine learning researcher who is like doing the real stuff so I'm very careful of like what I label myself here but I found nevertheless having a good analytical toolbox has been so helpful my titles have never actually or rarely been data analyst but my jobs have been very analytical heavy and that's all also like that how you can also see the book the book is really a preparation for you to have these transferable skills and then going into industry also being open-minded in terms of what jobs can you do with that it's not necessarily to look for the title that is data analyst but there are so many jobs where 90% of your day-to-day work is data analytics but the job function will be called something entirely else so right now I'm a climate specialist previous I was a liquidity and funding manager so that like doesn't say much about analytics but that's in the end I really relied on these foundational skills that I learned as a researcher and then with writing the book and and later certainly I expanded on that the scientific training you had in terms of being able to ask really good questions and figure out if things are true were complemented by these computational skills so while you might not be a hardcore pure data analyst you get to do these really cool hybrid positions it sounds like I would say I'm a hardcore data analyst oh sorry but I'm I'm drawing the distinction of being a data scientist and I mean now we're getting in a discussion to really differentiate the two but I am also right I'm not a programmer where I'm saying like again I'm not writing I'm not a software designer right or software engineer so that's where I'm drawing the distinction I do a lot of data analytics that in that respect yes hardcore even though it's not my job title I think that helps me understand a lot better so is there anything regarding the book that you want to discuss that I didn't bring up or mention there is uh programs and I happy to do some advertisement here like there's the cheeky scientist Association which is like a like a group really writing additional skills how to transition from Academia to Industry and one of the things that they really at like try to drill in is the value of informational interview what is an informational interview you reach out to someone and usually it means you will reach out to 10 people of who you find their job title or whatever you have some contact you might even be called on on LinkedIn you email 10 people and maybe one of them will reply and actually be willing to talk to you and that can really be a 20 minute conversation of what are you doing what are the kind of skills that you need how did you get there and that provides so much information about really what you need and I found I was so worried about my skill set my technical skill set and that in the end is not usually what makes or breaks your success in industry it's really can you work in a team the soft skills how much are you collaborating and I I kept thinking like of course I can collaborate I collaborate with my my colleagues in the lab all the time we share share a PCR machine of course I'm collaborating and I wrote papers with people but it is really a different Beast altogether because in Academia you're often really the maker of your own success I mean to some extent there's also a lot of luck and who you know but you are defining your research story to a large extent and just whether you schedule the PCR machine is not really collaboration yet because in a company there's a lot of redundancy actually and that is on purpose you want multiple people to be doing in a way the same thing because you don't want to have a key dependency you know this one person leaves the company the knowledge is gone right it's really then what collaboration is and can you deal with them can you deal with someone was maybe not easy to deal with that's exactly like what is meant by collaboration and teamwork I underestimated that at the time what that would take and how important that is I spent eight years in in Industry position before coming back to do my PhD and found some of the things if you're in Academia that would be troubling doesn't happen in industry and the kind of collaboration you have to make is very different just like what you were saying it's sort of collaboration time 10 you really rely on each other to make it successful and that is often in Academia right at least let's let's say I I cannot speak for all of Academia but in the field and the in the labs that I worked in you obviously interact with people and you might you will have call authors but you know one person was responsible for that experiment and the other person was responsible for that other experiment and that's exactly not what it is in Industry we have far fewer people involved with each dependency in a PhD dissertation versus if you're in Industry where I found teams were much larger and if you don't have a small piece done on time you have a waterfall effect so it's just the type of collaboration is very different so what have you what did you do in your career to develop those soft skills I think it was actually learning by doing so I mentioned that I worked for a scientific publisher for a while and I did not fit well with the company culture I did a lot of mistakes there I was good at what I did but people really didn't like me and I was like what is going on and I only like it later I realized well I didn't focus on building relationships with people and that's the big difference that I do now the first three months and every job is not about showing how great I am and what are all the skills that I bring this is tempting right you want to hit the round the ground running and you want to really show what you've got nope the first three months are about building relationship with the people that you're working with and that's really getting to know people not immediately getting to the agenda and now I'm German I have a little bit of a cultural background also that like I don't usually talk about the weather when I end enter a room so but I had to learn this so and now my really my focus is that I would go around introducing myself and have a chat and try to learn what is their communication style do they want to communicate by email there's generational differences right some people you really need to F to pick up the phone in order to get to get any kind of decision or like an a Buy in right and there's like especially if you're working in international context and obviously like let's face it most industry like you will have a lot of cultural differences like in a diverse team which is good right but then obviously like learning how important is hierarchy for them how important is punctuality right there's lots of different priorities for people and figuring this out that's really where I'm saying this is my first three months and that is something definitely that that's for me the big difference in terms of soft skills and doing a job well and after I've mastered the relationships after I've formed the relationship then I'm actually much more successful employing my my technical skills and being successful in bringing the collective goal forward that makes a lot of sense um especially in big team environments where I've been in if you don't know your co-workers and what motivates them and and what they're up to it's a lot harder so that makes sense the last wrap-up question is what's next for you in terms of your career are you going to keep with the with what you're up to and then also I've noticed on your LinkedIn and just looking that um your passion about science communication obviously you're here and you've written this book so in terms of career and some of your Communications work what's next for for you so I just transitioned within beo so I wanted to stay with the company because again uh good work life balance just feeling appreciated and cared for right is also maybe the other advice that I would give someone to transition in the in the beginning it often feels like I just want a job and I did this mistake as well I just joined like because someone offered me something and I accept it so because I like you know it's rough and particularly in today's job market like it feels like there's it's so hard to get one that you just accept as soon as you're successful but I learned over time right that's not all like if you want to be somewhat happy if you don't want to hate your job like you need to find a company that has a good cultural match and that was the case for me at bimo like because curiosity is rewarded I feel appreciated for the work that I do so that's why I really stayed with the company uh now and was looking for a job maybe to bring me a little bit back into the life science realm and that's exactly what I managed to do so I joined the climate risk team so climate risk analytics so I investigate how does transition risk so policy changes in relation to climate like affect our customers our risk that the bank carries or physical risk how much exposure do we have to an area like because there's a hurricane going on somewhere and this is again very analytical like really relies on the technical skill set as well but brings me a little bit back to like sustainability questions and the life science so definitely like I my heart is still there but there's a lot of things obviously that you're trying to uh navigate with like having a family and making a living which is like when you initially start and you're young maybe they're not is not quite I mean obviously you want to earn money but at a later time there's a lot of more consideration to have there so so and from that aspect I really liked I did a lot of things in my career I tried freelancing so I worked actually as a web developer for a while and that's why the companion side to the book is my doing that was certainly fun and that again also gives a lot of skills that now come back handy as a VAP developer I learned how to design a dashboard use of space and be having it look tidy that was all aspects that I really was focused in and doing ux and UI design and now that comes in again right if I'm developing a dashboard for business decisions then again having this look well very well organized what are the key metrics that I need to present how do I present this like it all comes together for me I have many done so many career Transitions and it still all is woven together with this analytical background even if my job titles are reading like a bizarre Hutch pod it's always the analytical thread that goes through all of that so in terms of right now I'm in a great position what will come next who knows do you have any side projects related to outreach stem communication anything like that right now I I would say the stem communication that I'm currently doing are with my kids to tell them about the world the like Science Education that comes there which unfortunately doesn't leave me much additional time to do a lot of additional Outreach well I appreciate you coming on today to do a little bit with us are there any places that guests could follow you online like um any social media obviously the website for the book will share so I would say my LinkedIn is probably the only thing that I keep up to date my own personal website is terribly outdated that was really when I I set this up when I searched a job four years ago and I haven't really done anything on it so this will not actually be well well representative it's really the LinkedIn and just Reach Out shoot me an email so and there you will find online through my website definitely but also if you write to the books authors on the on the books website you will definitely reach me excellent connect on L on LinkedIn that's maybe the easiest sounds good I'll definitely do that after this