The Hybrid Hackathons of the Future — now with Librarians!

Authors

Hannah Gunderman, Data, Gaming, and Popular Culture Librarian, Carnegie Mellon University Libraries

Ben Busby, Director, Solution Science and Principal Scientist, DNAnexus

Introduction

With the world still reckoning with the impacts of the COVID-19 pandemic, one thing that has remained constant is the need to change how people collaborate and communicate ideas, often shifting to remote and virtual formats.  The COVID-19 pandemic accelerated the rate at which hackathons are hosted in a virtual format. Remote hackathons have the potential to mirror the personally and professionally transforming experiences conveyed by in-person events to those who can not travel due to financial, physical, or environmental constraints. Remote hackathons allow the intellectual wealth of scientists in these countries to be applied to the important topics and goals of the hackathon, while supporting their health and safety through virtual participation. We hope that hackathons will retain a hybrid model to maximize the scientific contributions of both in-person and remote participants. 

Why are hackathons important?

Hackathons allow for concentrated, focused effort on a task or goal by bringing together scientific experts in a particular discipline, such as structural variants, or united by a common goal, such as ending neurofibromatosis.  Some hackathons solve thorny problems, make life easier for practitioners of specific disciplines, or push the boundaries of what a particular scientific field can do.  That said, hackathons not only produce content (usually software), but ideally also actively facilitate education and networking. Those who participate often have professionally transformative experiences that can lead to a wider scientific network, job opportunities, and increased confidence in their coding and research skills. 

Hackathons largely follow the model of “disruptive innovation” by serving as a prototyping layer across scientific organizations, producing new ideas and technologies that the community can then assess for value in their larger goals and initiatives. The prototypes that emerge either push the envelope of what is possible with biomedical informatics, or make day-to-day bioinformatics easier.  While the code isn’t necessarily persistent, these proof-of-concepts are intended for the community to build upon. Hackathons foster an environment with “buzz,” an economic geography concept referring to the serendipitous sharing of creative ideas that happens when people engage in face-to-face interactions. The last year has taught us that these benefits from hackathons are also afforded through hybrid or fully-remote formats, providing hope for a positive future of hybrid hackathons in scientific advancement and discovery. 

How do hackathons benefit the participants?

Not only do hackathons have an undeniable benefit to the broader scientific community, but, they also can provide transformative and impactful experiences for the participants themselves. These experiences largely revolve around the areas of confidence-building, educational development, and camaraderie. 

As described earlier, the “buzz” created in hackathon environments helps advance the sharing of creative and innovative ideas. Through this exchange of ideas, participants can advance their journeys in computational problem-solving and modern software development techniques. In the bioinformatics space, there are many beginner data scientists who are still learning foundational skills in computation and scientific collaboration. Hackathons, whether remote or in-person, offer a concentrated space for beginner data scientists to advance their skills in both of these areas alongside more established bioinformatics researchers. Not only does this afford educational benefits to these participants, but it can also increase their confidence as scientists who can contribute to important research endeavours. 

Finally, hackathons also create the opportunity for participants to forge close personal friendships and bonds, which can lead to long-term collaborations and network-building. 

Participants often find themselves in intensely challenging and time-limited environments as they race to accomplish the goals of the hackathon, and going through these transformative experiences together can lead to strong friendships and connections that span beyond the bounds of the hackathon itself. This is not limited to in-person hackathons, however: video-conferencing software such as Zoom and collaborative tools such as Slack allow participants to interact with each other and build both interpersonal and professional connections. 

A Retrospective Look Into CMU-DNAnexus Virtual “Genomic Data to the Clinic” Hackathon

The CMU-DNAnexus Virtual “Genomic Data to the Clinic” Hackathon (June 1st – June 4th 2021) was focused on bringing complex genomic data into the clinic.  Specifically, we focused on integrating Expressed Variants, Polygenic Risk Scores, Structural Variants and T-Cell Receptors into an Electronic Medical Record readable format using OMOP and worked on a clinically presentable interface. Remote support was offered by librarians from Carnegie Mellon University Libraries who have specialties in data management, bioinformatics, and information sciences. This support included collating important resources found by hackathon participants (such as tools, software, literature, etc.) into a single spreadsheet for easy access, reviewing the hackathon manuscript for syntax and readability, and preparing the manuscript for submission to BioHackrXiv. Communication platforms such as Zoom and Slack can offer ways to stay in touch and facilitate collaboration during a remote hackathon, but information can still get lost in translation in environments where we can’t see each other face-to-face. Librarians are trained in the information sciences and well-positioned to assist in keeping information organized and accessible during a remote or hybrid hackathon. 

Participants not only effectively used online collaboration tools to create innovative workflows and deliverables supporting the goals of the hackathon, but also used tools such as Slack to develop interpersonal friendships. Much of the same dynamic energy and “buzz” felt during an in-person hackathon was also felt in this virtual space and the experience has already led to some promising future collaborations and scientific endeavors, including an accepted proposal for a presentation at the 2021 annual meeting of the American Society of Human Genetics that will share the scientific findings from this hackathon. 

Upcoming hybrid hackatons

Although the pandemic is experiencing a long tail, we can still begin to envision what our post-pandemic future may look like, taking the lessons we have learned from navigating our remote environment for the past several months. One of the lessons we can bring into a post-pandemic future is that hackathons with a virtual option can help us create more equitable and diverse intellectual spaces for tackling the most pressing issues we face in bioinformatics. Moving forward, hackathons should take a hybrid model and allow for both in-person and remote participation, while allowing more team leads the sequestration they need to fully focus their energies on these efforts instead of juggling both work and the hackathon.

Further, leveraging the support of librarians in the hackathon space can lead to a more organized, cohesive, and collaborative experience for participants. This is particularly true for fully remote or hybrid hackathons, where clear communication channels are crucial for all participants. Librarians can help facilitate collaboration and coordination between remote and in-person participants, and help collate resources (such as tools, software, and literature) found during the course of the hackathon.  

We are excited to see what the future of hybrid hackathons holds for our field at large, and the scientific discoveries that will result from these events.  Below are some upcoming hackathons you can follow or get involved in!

Everything is bigger in Texas: Pan-Structural Variation hackathon in the Cloud! 

October 10-13, 2021, hosted by the Baylor College of Medicine

BioHackathon Europe

November 8-12, 2021, hosted by ELIXIR Europe

CMU-DNAnexus Hybrid “Genomic Data to the Clinic” Hackathon

March 9-11, 2022, hosted by CMU Libraries (stay tuned for more details!)

We also recommend keeping an eye on future events and initiatives hosted by the DEMON network, an international network for applying data science and AI to dementia!

Keep an eye on this link for more information about these and other events: https://biohackathons.github.io 

Meet Us in Atlanta for AACR

We will be joining thousands of oncology researchers and clinicians this week at the American Association for Cancer Research Annual Meeting, March 29 – April 3 at the Georgia World Congress Center. Stop by our booth #3051 to learn how DNAnexus Apollo enables advanced exploration of cancer datasets for therapeutic target discovery. Our scientific experts will be onsite all week for demos and to answer all your bioinformatics questions.

Last year at AACR, in partnership with St. Jude Children’s Research Hospital and Microsoft, St. Jude Cloud was launched, the largest public repository of pediatric genomic cancer data. Offering a suite of unique analysis tools and visualizations, researchers can efficiently mine, analyze, and visualize next-generation sequencing data for pediatric cancer and other life threatening diseases to accelerate scientific discoveries.

This year, St. Jude Children’s Research Hospital is celebrating the one year anniversary of the St. Jude Cloud and since its inception, 10,000 WGS data have been analyzed in continuation of the search to find a cure for pediatric cancer and other life-threatening diseases. If you want to know more about the St. Jude Cloud, visit their booth #4136. We are excited to celebrate the one year anniversary with St. Jude Children’s Research Hospital and St. Jude Cloud team!

Make sure to stop by either DNAnexus booth or the St. Jude Cloud booth, to celebrate with us at AACR! Check out our conference activities below. Can’t make it to any of our events? Request a meeting.

Demo: Explore Millions of Clinico-Genomic Data Points in Seconds with DNAnexus Apollo

  • Monday, April 1st – 1:00pm-2:00pm
  • Tuesday, April 2nd – 10:30am-11:30am

Cancer genomics is advancing precision medicine, leading to better diagnoses and treatment strategies. The Cancer Genome Atlas (TCGA), extensively characterized over 20,000 primary cancer and matched normal samples, creating a deep resource of over 25 petabytes of genomic, epigenomic, transcriptomic, and proteomic data. This landmark program has already lead to improvements in researchers’ ability to diagnose, treat and prevent cancer, yet mining this type of data for novel findings remains a huge endeavor.

DNAnexus Apollo enables at-scale navigation and exploration on combined genotypic and phenotypic data, whether it is TCGA data, your own protected data, or a combination. What actionable information will you find in your cancer datasets?

Customer Poster

3671 / 29 – Visualize 10,000 Whole Genomes from Pediatric Cancer Patients on St. Jude Cloud 

  • April 2nd, 1 – 5pm
  • Presenter: Clay McLeod, St Jude Children’s Research Hospital

Discussing the State of Biomedical Data & Looking Towards the Future at PMWC

DNAnexus once again attended the Precision Medicine World Conference (PMWC) in January in Santa Clara, California, to learn about new innovative technologies, thriving initiatives, and clinical case studies that enable the translation of precision medicine into direct improvements in health care.

George Asimenos, Chief Technology Officer, hosted a panel discussing the challenges in applying AI on biomedical data and the future of using AI in regulated contexts. Brady Davis, VP Strategy and Marketing, also moderated a discussion on the future of DNA sequencing interpretation, touching on how visualizations are driving the understanding of human genomes & diseases.

You can view the recorded panel discussions below:

Challenges In Applying AI On Biomedical Data

Challenges in AI Panel

Moderator: George Asimenos, PhD, CTO, DNAnexus

Panelists:

  • University of Washington, Paul G. Allen School of Computer Science & Engineering – Su-In Lee, PhD, Associate Professor
  • UCSF, Helen Diller Family Comprehensive Cancer Center – Laura J. van’t Veer, PhD, Leader, Breast Oncology Program, and Director
  • Regeneron Genetics Center – Jeffrey Reid, PhD, Head of Genome Informatics

Abstract:

AI methods promise to revolutionize the way we approach medicine, by building and training models that make accurate predictions. But the application of AI in actual biomedical contexts comes with distinct challenges. This session features speakers in the forefront of real-life AI applications, who will discuss challenges such as how to reach an interpretable/explainable AI model and how to apply AI in a regulated context.

Data Curation, Integration and Visualization

Data Curation Panel

Moderator: Brady Davis, VP Strategy and Marketing, DNAnexus

Panelists:

  • Stanford University, Department of Genetics – J. Michael Cherry, PhD, Professor
  • AncestryDNA – Eurie Hong, PhD, Senior Director of Genomics
  • UC Santa Cruz Genomics Institute – Benedict Paten, PhD, Assistant Professor

Abstract:

Understanding the relationship between the millions of functional DNA elements and their protein regulators, and how they work in conjunction to manifest diverse phenotypes, is a key to advancing our understanding of the human genome and diseases. As the cost of DNA sequencing continue to drop, the interpretation of the ever increasing amount of data generated represents a considerable challenge.