Allissa Dillman and LaFrancis Gibson on the Power of Hackathons

March 31, 2026

Joining Trends from the Trenches is Allissa Dillman, founder and CEO of BioData Sage, and LaFrancis Gibson, manager for health promotion at ORAU, who discuss this year’s Bio-IT World Conference & Expo Hackathon sponsored by the National Institutes of Health (NIH) Common Fund Data Ecosystem (CFDE) Training Center. With host Allison Proffitt, they discuss which tools and datasets from CFDE will be part of the Hackathon, who should participate, learning and collaborating beyond datasets and tools, FAIR principles in an ever-developing AI environment, and a message beyond the teams that are participating.

GUEST BIOs

Allissa Dillman, Founder & CEO, BioData Sage
Allissa Dillman, Ph.D., is the founder and CEO of BioData Sage, which helps integrate strategic data science practices into the biomedical and biological sciences. She holds a doctorate in computational neuroscience from Karolinska Institutet and has over a decade of experience spanning government, academia, and nonprofit work, with a focus on strategic planning, community engagement, and workforce training. She is also co-PI of the NIH Common Fund Data Ecosystem Training Center, leading efforts to develop training resources, podcasts, and community events, such as hackathons, to support biomedical research. Dr. Dillman is an adjunct faculty member at Montgomery College, where she teaches courses in data science and bioinformatics. Previously at NIH, she led workforce development and international training initiatives and organized hackathons, bootcamps, and community-building programs. Her work centers on building inclusive, reproducible, and impactful data science ecosystems through partnerships across major research and public institutions.

LaFrancis Gibson, Manager for Health Promotion, ORAU
LaFrancis Gibson serves as a manager for health promotion at ORAU and has created a unique skill set and understanding of community-based programs and how to manage them to maximize success. With more than 15 years of experience in managing outreach initiatives, training development, and program evaluation, she ensures project success through her expertise in budget control, risk mitigation, and strategic communication with stakeholders from government agencies such as the National Institutes of Health (NIH), National Library of Medicine, and Centers for Disease Control and Prevention. Ms. Gibson is experienced in working with multiple stakeholders and partners across federal agencies, Minority-Serving Institutions, community and faith-based organizations, and national organizations. She currently serves as the contact principal investigator for the NIH Common Fund Data Ecosystem Training Center, providing oversite, coordination, and compliance in delivering bioinformatics training programs that align with NIH mission.

CO-HOST BIO

Allison Proffitt, Executive Editor, Bio-IT World
Allison Proffitt is the executive editor of Bio-IT World and other titles under the Cambridge Innovation Publishing umbrella and the producer of several podcasts including Trends from the Trenches, Breaking Health, Medtech Talk, Bioprocessing Unfiltered, The Scope of Things, and more in the healthcare and life sciences space. She is a science writer with a background in biology and chemistry, research experience in cancer biology, and an expanding repertoire in biotech, AI, and battery chemistry. Previously, she worked at Chemical & Engineering News, Modern Drug Discovery, and Forbes. She has a bachelor’s degree in the communication of science, engineering, and technology from Vanderbilt University and a master’s degree in science and medical writing from Johns Hopkins University.

TRANSCRIPT

Welcome And What We Cover

Welcome to Bio-IT World's Trends from the Trenches Podcast, your insider's look at the science, technology, and executive trends driving the life sciences. I'm Allison Proffitt, Executive Editor at Bio-IT World, and in our episode this month, I am excited to speak with Allissa Dillman and LaFrancis Gibson, both data scientists, experts in training and education, and leaders from the NIH Common Fund Data Ecosystem, CFDE. They are, again, facilitating the Bio-IT World Hackathon at the Bio-IT World Conference and Expo in May. So welcome to you both. Can you introduce yourselves, give us a little bit of the breadth of the work you do? Because I think it's pretty extensive.

Yeah, so I'm Allissa Dillman. A little bit about me. So I'm the training and engagement director for the CFTE Training Center. I also run my own company, BioData Sage, and I mostly focus on training and engagement around bioinformatics and data science. So I run lots of hackathons and workshops and conferences, basically all the different ways that you can kind of learn more technical skills.

Great. La Francis?

Yes, hello, good afternoon. So I am the contact PI for the CFDE training center. So I come with about 14 years of experience in public health. My background centers around management working with stakeholders, community engagement, as well as training, development, and support. And we've been supporting the CFDE training center now for about two years. We're year two of our project and with three more years to go. Awesome.

So y'all are not new to Bio-IT World. You've been running this Hackathon for a couple years. How many years?

Yeah, so the Bio-IT World Hackathon actually started back in 2017. I will say I was not at the first iteration. In 2018, I was part of the sort of like foundational organizing group. At the time I was actually so pregnant, I could not fly. I had to take a train to Boston. But yeah, so I've been sort of part of the Bio-IT Hackathon since 2018.

Awesome. That's way more than a couple years. Sorry.

Well, we had some, we had some time off during COVID. So it hasn't necessarily been like every year.

Why Hackathons Work

So okay, I want to talk about what we'll do there with the CFDE data and tools. But first step back for me, y'all both do a lot with with training and education. What does a hackathon in general hope to accomplish? Is this an education play? Is this like a specific problem solving play? Is this data gathering? Yeah.

Tracking Results After The Event

So the thing I really love about hackathons is every hackathon kind of has these three components, a creation component, an education component, and a collaboration component. And it doesn't really matter kind of like what you're focused on per se, like every hackathon will have all three of these things, right? Because you're trying to build something, whether that's a wearable or an infographic or a bioinformatics pipeline. You're gonna be collaborating with people that you normally don't get to work with, and you're gonna learn from each other, from the organizers about whatever the theme is on. So the nice thing is like every hackathon does kind of have each of these components, and that's part of the reason why we're so excited about using hackathons. So one of the things we want to do is really get people to know more about all of the data and tools and resources that are part of CFTE. You know, it's there's 20 odd different programs that turn out data and tools. And so there's that education component of look at all the amazing things that we've got for you to use that are open and available. You know, and then there's the collaboration component. The really cool thing about Bio-IT as a conference is it really does have quite a wonderful mix of both like researchers, but also folks like in tech and IT, and also in, you know, health IT and all sorts of like those kinds of backgrounds. So like we really get to collaborate with people who use very different tools or at least kind of like, you know, different like frameworks as well for how they deal with clinical data versus, say, you know, biological data. So kind of that collaboration component is there. And then also we're just really excited to see what people build with the tools and the resources and how they integrate them into, you know, their own work and what they pull in from their own work into the hackathon. So we get to kind of see them in real time use these data and resources and see how they use them. Great.

Is there do you track that? I mean, do you look later at publications and say, hey, that started at one of our hackathons?

Yeah. So we we do try and track those sorts of things as much as we can, and we really support trying to create publications during the Hackathon. So we do have, we make sure that every team sort of indicates that they're gonna have a writer for the team. And that's not just for making sure that they have good documentation for whatever they build, but also kind of writing at least kind of the skeleton of a paper. That we do try and really facilitate, you know, that we're we're writing things that we can actually publish, if not like a research paper, then, you know, a white paper or blog post or something that's sort of like here's the things that we've learned and here's what we're doing. But we will also have post-hackathon interviews that are sort of like at later dates after the hackathon to try and see, you know, have you been able to integrate CFD data tools into your work? If not, like, you know, why not? So we're also gonna kind of check in with our participants afterwards to see how they're doing and what they're doing.

Omics Projects And Key Datasets

Okay, very cool. So this year, this year's hackathon, as far as I know, though, you can correct me if it's broader than this, but there are six projects listed on the Bio-IT World site that are leveraging Omics data specifically and CFDE tools. Can you tell me about those tool sets and data sets?

Sure, yeah. So we have projects, a couple of projects actually focused around Motor Pack, which is actually an exercise-based data set where they have actually looked at molecular signatures across lots of different data types in rats. And then they're also currently building out that data set in humans as well. So there's just a huge wealth of information because it's multiple tissues, multiple ohmix data types, a bunch of clinical measures, and sort of like sedentary rats and then rats who've been doing exercise. There's a lot of really neat things to do with that data set, like just in and of itself, but it's also really cool to plug into other kinds of data sets. So for instance, data sets on disease models such as diabetes, hypertension, cancer, right? And kind of seeing, you know, where there's an overlap in these exercise models with disease models. But you can also, you know, interlay them with things like drug, you know, side effects, drugs repurposing, drug response and things like that, and see, you know, are there you know certain drugs that could maybe mimic certain molecular signatures of exercise and stuff like that? So that's that's actually kind of encompassing a couple of different projects. But there's also projects that are really interested in looking at visible neural networks for cancer drug responses, and they're gonna be using a couple of different data sets, everything from like protein-protein interactions to perturbation data. There's a couple of data sets that are actually focused on like drugs and drug side effects as well, that we'll be using, like drug central, for instance, in that work. And then also things that are more sort of on the like biology side of things where we're looking at like post-translational modifications, kind of like looking for them, where they're all at within the data sets, but then also seeing if we can map them to, like I said, some of these other data sets around disease and like molecular mechanisms. So those are kind of the overall overarching effects. A good chunk of the teams will also be playing with things like AI and ML in their models and kind of seeing, you know, what they what they can pull out.

How Teams Mix Skills

So when somebody wants to be a participant in this, do they pick their favorite data type? Do they come at it from the tool angle? Like how do they how do they align their participation?

Cloud Options For Fast Compute

Yeah. So I mean, we'll meet people who have both sort of, you know, biological and clinical expertise, drug expertise, but also technical expertise, right? These are pretty complex data sets. And the metadata around them is kind of complicated too. So having folks that really have a strong technical background in just handling like computationally really complex data, you know, and they're gonna be kind of all working side by side, right? So, like no one person needs to have all of those skill sets. We'll kind of make sure that each team has a good balance of both sort of the subject matter expertise and the technical expertise. So you can kind of participate from any of those angles, right? If you're more on the clinical side, there's definitely projects that have a clinical focus and would really appreciate that clinical expertise, versus if you're like, I have no idea what omics data is, but I can build you some really cool things on the cloud and I know how to use AI and ML. And that's what I'm gonna contribute. So like there's space for all of those different types of skills and backgrounds.

So is it all on the cloud? Is that where the compute is? I don't think that we're hosting a cluster at Bio-IT World this year.

Yeah, no. So we've got we're gonna have a couple of different computational tooling options. So CFDE itself has something called a QUIC. So it's the we have our cloud workspace. We'll be offering if folks want to work on the QIC, so that'll be freely available. We also have another option called Cloud Bank, which we will provide, which has like the three commercial cloud vendors, so Azure, AWS, and GCP will be available. So we've got like a couple of different ways that people can sort of work together in the same computational workspace. They're not required. They can work locally if they want to. But like I said, it's it's a little bit harder when you're kind of dealing with these large data sets. And also, like, right, it's only two days. So we want those computations to go pretty quick. So we will have lots of different computational cloud-based tools available.

So these tools and data sets, are these things that folks can access outside of the hack, or is there some sort of proprietary access here?

No, that's a great question. So the QIC is available so that that CFTE cloud-based platform is available and open outside of the hackathon. You can actually check it out now if you want to. And then the data sets are all available. I won't say like perfectly open just because some of them are human data sets and so they are restricted access. But you can apply for access at any time outside of the hackathon as well.

So if you wanted to follow up, if you were working on something that you thought, hmm, we we want to carry on this train of thought, you could still get access to it afterwards.

Correct. Yep. And we we really highly encourage and offer as much support as we can if there are teams that do want to continue on with their project.

I would like to just also add on to that when we talk about the coming fund data ecosystem and the coming fund data, those data sets that we're utilizing at the hackathon, we're approaching it from the FAIR principle around making sure that the data is findable, accessible, interoperable, and as well as reusable. So we want to make sure it's open access that people can use beyond the life of the hackathon. Yeah.

So actually I had a question about FAIR. So we AI is not actually particularly new, but neither is FAIR. And I was wondering when we first started covering FAIR, gosh, like 20 years ago maybe, the way that we are thinking about FAIR principles changing in an age with so many AI algorithms accessible to us? Is it becoming, I don't know, different in terms of how what we consider findable, what we used to think of as findable, and now what we think of as findable? Or is are FAIR principles a persistent underlying factor of the data that we need to keep in mind always, even as we move forward in with AI algorithms?

Subscribe And Share Topic Ideas

That's a great question. And and I would say it's both. On one hand, you you need to sort of keep the FAIR principles at heart because you know, we're going to have things that require machine readability, interoperability. And you know, even with AI, those data structures need to be present. But I will say, you know, one thing that we've seen sort of a huge explosion in with AI and actually quite a few of the different programs and you know platforms that are part of CFDE and now use things like AI chatbots to help you find things using like human-described languages, right? Like as a biologist, I can sort of say, like, hey, I'm interested in this biological question. Tell me what data sets you have, tell me what I need to be thinking about, tell me help me design and experiment with that data. So these AI chatbots are sort of like helping it be more findable and reusable because you can have like that discussion, that open-ended discussion with them. But you still kind of need, you know, the the under underlying like fundamental structures so that way you like, you know, for instance, the I AI can go and like search all of the data very easily because it has all of the you know linked metadata and all that.

Researching How Teams Use GenAI

Are you enjoying the conversation? We'd love to hear from you. Please subscribe to the podcast and give us a rating. It helps other people find and join the conversation. If you've got speaker or topic ideas, we'd love to hear those too. You can send them in a podcast review.

So when we, Alyssa, when you and I had spoken earlier, you talked about some secondary questions that you wanted to explore at the hack that are maybe more fundamental about how people learn and how they collaborate beyond just these tools and these data sets. Can we can you tell me a little bit about those questions and how you want to approach that and what data you're hoping to gather there?

Yeah, we're actually collaborating with a research group that has been sort of exploring hackathons from a research lens for quite some time. So we were really excited to partner with them to really try to better understand how heterogeneous hackathon teams incorporate Gen AI into collaboration. So we're really interested to sort of first of all see how different folks with different backgrounds utilize AI. But then like, you know, how how, like if you've got six different people, for instance, and like one is a bio background, you know, one is a technical, like a maybe an IT background, one is a clinical background, and they're all using AI. Like how how are they using AI? How does that look different? And then how do they come together at the end and kind of put all of those different pieces together? Do they use AI to kind of cover the parts that they are, you know, less sure about, or do they use AI to help communicate with each other because they have different backgrounds? Like we're really curious to kind of see like how that works and how functional diverse functionally diverse individuals kind of verify, correct, and integrate AI inputs like as a team. Interesting.

Verify, correct, and what was the other one?

Integrate.

Integrate the inputs. Okay. That's a fantastic question. How are you going to how do you how do you gather that? You're gonna hover over the teams and pay attention to who's doing what?

Yeah, that's a great question. So part of the reason why we are offering things like the cloud bank tools and the quick is that we will be able to see like their workspaces. We will be able to see, you know, what kinds of conversations they're having with AI, how they're using those tools. So that's like one way we'll be able to kind of observe is you know, through them sharing their chat links and and like I said, using these shared spaces that we'll we'll create. But we're also going to interview folks like while they're at the hackathon. And then, like I said, we're also going to touch base with them. I can't remember if it's three months or six months after the hackathon to kind of keep having these conversations. So it'll be kind of a mix of observation, both like a technical observation of like the work, but also, you know, observing them as they are working and then interviewing them.

That seems like such a gosh, a question that we need to consider way beyond life sciences, even just how does AI change the way people work in general? And it's a really interesting application. Will you be hopefully publishing those findings? Do you expect them to impact the way groups work in general? Would you think it might change how you how you organize next year's hack?

Yeah, no, that's a great question. And we are definitely aiming to publish our findings. And, you know, a big chunk of why we're kind of looking at this is so that next year we can be even more intentional about like what kinds of tools, how we integrate the tools, what kinds of frameworks we have, and sort of what guidance we give the teams when they work together, like in a hackathon perspective for sure. But also, you know, because we're the training center, we really want to start integrating bits of Gen eye AI into our trainings as well. And so, like, this will kind of be our first snapshot of like in real time, how do people use things? Ooh, like that actually is a really interesting way that we keep seeing people use AI and we really like to make that more intentional. Let's build in some of those sorts of walkthroughs in our trainings and things like that. So, we're hoping to kind of utilize these observations in a couple of different ways.

Training Center Mission And Approach

So, I like those ambitious goals for how both the hackathon will be run and how you will kind of assess other questions at the same time. And I love that that will be that that's being explored and that you'll share that with the community. Maybe La Francis, tell me about the training center's goals overall. Is this is this kind of dual role of both sharing data and tools in the one hand and also thinking big picture about how people learn? Is that part of the training center's mission? What does that look like?

Yeah, good question. So as far as the training center mission, so let me just kind of phrase it as this. We've positioned ourselves to serve as a hub for developing and implementing training programs that bridge knowledge gaps, as well as also improve access to CFDE resources and foster collaboration amongst CFDE users and other researchers, individuals who may not even be aware of CFDE or maybe new to the space. And so, in order to accomplish what we've set out to do as the training center hub, we look at opportunities such as the hackathon on how can we go into the communities and provide hands-on experiences, training, and bring in a collaborative approach because we realize learning is multifaceted. It happens when you're amongst your peers, when you're in spaces where you're learning from each other. And so, that's where the education component comes, where it's not just Alyssa and I showing up, you know, teaching and training, but it's more so learning from each other. And so creating these opportunities, such as the hackathon, is one of the many ways in which the training center is really walking in this this path of building training and and reaching our our target audiences.

Do you have a content area that you focus on? And this is you know an NIH effort. Is this focused on a specific content area or is it really broad?

So when we talk about, so this this actual the training center is funded through the NIH Common Fund. And I say NIH Common Fund, that pretty much consists of a diverse set of databases, data sets, tools, and resources that range from various topic areas across genomics to phenotypes and all the in-between within the biomedical space. Okay.

What do you, when you are looking at registrations from Bio-IT or from the Bio-IT World Hackathon, who are you looking for? Are you looking for students? Are you looking for bench scientists? Are you looking for executives? Are you looking for everybody?

So the quick answer would be everybody, right? Right, right. But you know, so yeah, definitely students, data sciences, you know, we hope to see some software developers in the mix. Anyone in a life science based professionals, you name it.

Could these be vendors as well? Are vendors welcome to participate in this and and learn from each other?

Of course.

Great. All right, so is registration still open? Let's say yes, or else this will have been a teaser with no fruit at the end of it.

Registration is still open.

Open, come join us.

Great.

Yes. Until March 31st.

Reuse Data Instead Of Reinventing

Oh, okay. So it's it's there is a there is a deadline here that we need to be aware of. For the Bio-IT World community that is, you know, Bio-IT World is several days past when the hackathon's over. And maybe not everybody can be part of the hackathon. Is there a message or a challenge that you would like to issue beyond the teams that are actually participating? Is there something you want kind of a big picture challenge to go out to our community?

That's a good question. I think, I think for me the challenge is really to sort of look at what's already been built and you know figure out how to utilize those things in new ways, right? Like I think I especially as people like my original background is bioinformatics, and we have a tendency to sort of be like, all right, I'm gonna completely generate a new data set, I'm gonna completely generate a new pipeline, I'm gonna completely generate like a new thing, and then it's novel and like I get to publish it. But there's so much data out there, there's so many resources out there that we don't necessarily need to expend resources reinventing the wheel over and over again. Rather, you know, there are ways to reuse the data in so many new, unique and creative ways. You know, so what we really want to know is like what's what's the barriers to doing that? Like, what can we do to facilitate data reuse in new ways, creative ways, or even just for learning purposes, right? A really great way, for instance, if you're you know more of on-the-bench biologist side of things and you're really interested in kind of beefing up your technical skills, a great way to do that is to take a data set, you know, that has a published paper connected to it or some sort of outcomes connected to it and say, can I reproduce this? Can I, you know, make a plot that looks similar to this, given the methods and how they're described, right? Because at least I kind of know what my output should be. So when I'm kind of playing with this data, I'm like, okay, I will expect hopefully to get at the end of this data exploration, like this plot that looks like this. So even if you're just using the data to kind of like learn technical skills, it's it's you know, a so many data sets and resources are available to kind of learn new technical skills. So just kind of like, you know, making that connection with use, reuse, and and just learning.

And I will add, you know, with that, when we talk about all these large data sets, data is only impactful if the people can use it, right? And so this is where we hope to bridge that gap at the hackathon and and allowing people to really immerse themselves in this space for two days and work across the different data sets and tools and across each other.

Yeah, and you said that there are there are human data sets that have some restrictions on use, but there are other data sets that don't. So this could be really open to somebody who is new to the space or just wants to play around with some tools. There are data sets that are more accessible really broadly, correct?

Yeah, so I'll highlight the motor pack data set again, just because it is, you know, because it's FRAP data, so it is completely open. Everything from the raw data all the way through to the like the analysis outputs and the summary level data, all of the code used to generate those outputs, all of it is free, openly available from raw data to analysis. So you have both been doing this for a while.

Do you ever get bored or do you feel like, oh gosh, these folks are coming up with the same ideas every time?

Prizes Judging And Free Entry

So I honestly have kind of lost track of how many hackathons I run. This may be around my 70th at this point. No, and and part of it is yes, there they're definitely, especially with some themes in in hackathons, there there does tend to be certain ideas that kind of come up over and over again. But it never gets boring because it's different people and they approach it with very different ideas and tools and, you know, just just novel ways of looking at things, right? So if you have a team that's all gonna explore like how do you high use machine learning with say the motor pack data set, right? Like that's kind of a very broad question, first of all. Yeah. And ask that question in a bunch of different ways. And depending on, you know, the team that ends up forming, you could kind of end up focusing a lot more on like the clinical side of things, or a lot more on sort of like the basic biology side of things, or a lot more on, you know, sort of like the technical side of things, like I just want to build a really cool pipeline. So like the sort of how people solve it is always different. And that's what I really love about hackathons, and in my humble opinion, sort of the power of a hackathon. Like I feel like I could give the same question to five groups of people, and they will all find their own unique and creative way of approaching that problem. That's awesome. I do just want to add, you know, not only should you join us for this hackathon because you'll learn lots of cool things and meet lots of cool people and play with lots of cool data and resources and tools. Also, there is monetary prizes attached. There's real money here. Yes, there's real money here. There's multiple prizes up to $5,000. So come and play with us and our data for two days, and you will potentially walk away with some real hard cash.

So, how do you what's the assessment that decides who gets the cash?

So we will have sort of multiple things that we're looking for, you know, everything from sort of novelty to how well can we actually follow your code on GitHub to your presentation because there's a final presentation as well. So there'll be a couple of different aspects that we'll be kind of like assessing. Great.

And I don't know if we mentioned this, but it is free, free, free. You do not have to register for the main conference to participate in the hackathon. Yep. So we just come spend two days with us and we may even provide you lunch.

Wrap Up And What’s Next

Awesome. Okay, well, thank you both so much. I enjoy talking to you. I'm very much looking forward to seeing you in Boston and and seeing the what comes out of these these two days.

Thank you so much for having us. And I always love talking about hackathons, so this was my favorite.

Thank you. Appreciate the time.

Great. Thanks so much. Thank you for listening to Bio-IT World's Trends from the Trenches Podcast. We hope to see you again soon.

Stay Connected

Meet the Host

Eleanor Howe

Eleanor has been working at the cutting edge of bioinformatics for over 20 years. As founder of Diamond Age, she led the company’s evolution from a small, project-based service provider to a full-fledged consultancy that works closely with clients to tackle their most difficult research challenges. Trained as a computational biologist, Eleanor has deep expertise in transcriptional profiling as well as drug discovery and development. She earned her doctorate degree in bioinformatics from Oxford University and spent years in biomedical research at The Institute for Genomic Research, Dana-Farber Cancer Institute, and The Broad Institute.

Learn more