DATA SCIENCE HACKATHON

AUGUST 9-10, 2021

The 2021 Data Science Hackathon will be held virtually online during August 9-10, 2021. This would be the third data science hackathon organized by the UAB Informatics Institute. We invite informaticians, statisticians, bioengineers, data scientists, cancer researchers, and oncologists at all levels to participate in solving cancer-related research, prevention, and healthcare practice problems using Artificial Intelligence (AI) and data science. Registration (here) is FREE but limited (first 120 who form a team successfully).

We are providing all hackathon registrants an optional “pre-hackathon” module in the next eight weeks. The pre-hackathon module includes informational webinars, training boot camps, and virtual networking events weekly to help those uncomfortable with the hackathon “ease” into the hackathon competition. These events will be recorded, and the content will be made available online. In addition, we have prepared slack channels, virtual social network lounges, AirTables, cancer clinical data sets, and U-BRITE computing infrastructure to make teaming up and generating ideas easier than before. 

Should you wish to participate as a “mentor” or a “judge”, you do NOT need to register. Just go to the website and fill out your profile to become available to a team as a “mentor”, or stand by to serve as a “judge” in the end. We will be in touch once you indicate your preference.

To form teams and network with others virtually, please join the Hackathon slack channel. Should you have any questions, feel free to post them on the slack channel. The organizers will be there as well to answer questions.

Timeline of Steps Prior to the Hackathon

Calendar of Events

Scroll down for video recording of the informational sessions below. 

Stage 1 - Registration

Click here to register for the AI to Fight Against Cancer Hackathon. Space is limited to the first 120 to successfully form a team.

After you register, learn more, attend our Informational Sessions to learn more the hackathon.
 
Informational Session 1 – Friday, June 25, 2021 at 10 a.m. – Recording below
Informational Session 2 – Friday, July 16, 2021 at 10 a.m. – Recording below

Recording of Informational and Training Sessions

Cancer Data Science Hackathon Guidelines for Projects, Competition, Evaluation and Roles

We have provided example topics. Teams may select and develop, broaden, or narrow one of the example topics, or teams can also design a new topic and submit it by the posted deadline. All Cancer Data Science Hackathon topics should be rooted in the scientific literature, especially recently published papers, and have the potential to impact the prevention, diagnosis, treatment, or research of cancer. A panel of scientific experts (“judges”) will be asked to evaluate and improve the submitted topics. In the end, a successful hackathon team will be expected to develop answers to a particular cancer hypothesis/problem using data science approaches, by following these evaluation criteria:
  1. Does the project involve a highly interesting hypothesis, phrased as a question statement? All good topics should be firmly grounded in at least one scientific paper on cancer published in the biomedical literature.
  2. Does the project have the potential to advance public biomedical knowledge and/or understanding of cancer? The advancement may be defined broadly, from epidemiology to treatment, including basic, translational, and clinical, and population impacts.
  3. Does the project involve the innovative use of data science tools from GitHub/GitLab? At least one of the team members should be skilled in deploying the software tools and designated as a technical contact in the team’s roster before the Hackathon.
  4. Does the project leverage existing publicly available data sets, particularly those on the U-BRITE platform? If additional data sets are needed, beyond what is already available, the team should work ahead of time (some of these data require non-trivial effort to stage) to communicate with the Hackathon IT Architect ( Zhandos Sembay ), who will assess whether we can bring them to U-BRITE before the Hackathon.
  5. Is the project feasible to perform with 50-80 hours of total team efforts within the 2-day Hackathon period? The team should stay focused on collecting key evidence to address the hypothesis with the understanding that future work may be needed to clear all doubts.
  6. Does the informatics approach adopted demonstrate rigor? Is the sample set too small or biased? What statistical techniques are used to quantify the biases or validity of conclusions made?
We will not evaluate the hackathon project based on these factors:
  1. Experimental validation of findings. In silico validations, using cross-validations, statistics, independent literature search, or unbiased separate data sets, are sufficient.
  2. Comprehensiveness of subject problem examined. Due to time constraints, preliminary conclusions that can help answer the question are adequate.
  3. Publication-readiness. We understand that reaching a point for publishing findings requires additional effort, especially by testing multiple alternative hypotheses.
  1. Provide scientific guidance on the overall project design, data to be used, and choices of analytical tools necessary to complete the project on time.
  2. Facilitate constructive and critical discussions among the teams to improve the overall rigor and reproducibility of the approach.
  3. Coordinate tasks and different roles to be performed by different team members.
  4. Help overcome obstacles as they arise before and during the Hackathon.
  5. Help the team prepare a final presentation for the competition.
  1. Form a team at least one week before the Hackathon and send us a list of team members.
  2. Discuss and pick a topic for the Hackathon.
  3. Prepare for the Hackathon by installing necessary software, attending training sessions, ensuring .readiness, and working with organizers to bring in necessary public data sets or software tools into U-BRITE beforehand..
  4. Divide up roles: data cleaning, programming, informatics approach design, data analysis, technical writing, etc.
  5. Develop your solution during the Hackathon.
  6. Prepare and deliver a presentation about your work (see below). Start preparing after the Hackathon is complete so you can focus on your project during the two days of the Hackathon. Presentations will be recorded.
  1. Attend presentations as assigned by organizers, either live Zoom or watch the recording.
  2. Judges may also be mentors but will need to recuse themselves from evaluating teams they are mentoring.
  3. Evaluate presentations based on scientific rigor, innovation, and presentation
  4. Submit evaluations to organizers.
  1. Each team will have ten minutes (less than ten slides) to report problem addressed, solutions, approach, discussion, and conclusions.
  2. Codes, results, and documentation all need to be deposited in U-BRITE by Tuesday, August 10, at 11:59 pm. Participants agree to make their work open-source and openly accessible to U-BRITE users.
  3. Prepare your presentation after (not during) the Hackathon.
  4. Team presentations will be held via Zoom Friday, August 13, at 10:00 am. We will send out a schedule of presentations.
  5. Judges will rate team projects based on the above criteria.
  6. We will present awards during the closing session on Friday, August 20, at 10:00 am via Zoom.

Reach out if you have questions.

We look forward to an innovative, productive, and fun hackathon!

Thanks to all organizers, mentors, judges, and participants!

Stage 2 - Profile, Project Idea and Resource Sharing

Submit Expertise Profile Information

Once you have registered at eventbrite, fill out this form to provide your profile information to identify your preferred role and to help with team formation.

Expertise profile information deadline is July 31. Participants who submit their profile information the deadline will be eligible for contest awards. Teams must have at least one team member who registered by the deadline.

If you would like to be either a mentor or a judge please indicate this under the Role question.

See other participants below!

Submit a Project Idea

Once you have registered at eventbrite and submitted your participant profile, fill out this form to provide project idea.

See other projects below!

Prepare a Data Set



Stage 3 - Team Formation

Form a Team

Whether you are looking for a team, recruiting members for your team, or already have a team, through Slack (24/7) and the Wonder Lounge (especially during our networking sessions) to meet other participants, socialize, discuss project ideas, and form teams. While individuals may participate, only teams will be eligible for the competition. Don’t wait, as only a maximum of 25 teams will eligible to compete for awards. If you are finding it challenging to find a suitable team or recruit team members, let us know and we can help suggest people who would be a good match.

At least one member of the team or the team mentor must be affiliated with the network of our co-sponsors.

Networking Events

Project Discussion and Networking Session 1
Friday, July 2, 2021
10:00am– 11:00am

Project Discussion and Networking Session 2
Friday, July 9, 2021
1:00pm– 2:00pm

Readiness and Project Final Discussion Session
Friday, August 6, 2021
10:00am- 12:00pm

Stage 4 - Learning, Training, and Preparation

Attend Pre-Hackathon Webinar/Trainings

AI Boot Camp
Friday, July 23, 2021
10:00 am – 12:00 pm

Special Lecture by Jake Chen, PhD: AI in Cancer Research and Oncology Applications: An Overview of Recent Advances
Friday, July 30, 2021
10:00 – 11:00 am

U-BRITE Boot Camp
Friday, July 30, 2021
11:00 am – 1:00 pm

Check Technical Readiness

Follow the checklist to make sure you are technically ready. Also attend our readiness session on 8/6/2021

Readiness and Project Discussion Session
Friday, August 6, 2021
10:00 am – 12:00 pm

Preparation Instructions for AI Against Cancer Data Science Hackathon

  1. Sign up for an Airtable account at https://airtable.com/if you don’t already have one.
  2. We will send out instructions and/or invitations on how to access the Hackathon base in Airtable to view and edit information there.
  3. Click on the link to access the information.
  4. In the Participants table, review and update your information as necessary, including your preferred/available roles.
  5. Open the Projects table and review the projects there. in the Projects table. See if any interest you and think of how to use/revise it to develop a project. Or you can enter a new project idea. See attached guidelines for development of projects.
  6. Look for people you may want to team up with in the Projects and Teams tables. You can join an existing team or form one of your own. Teams should have about 2-4 people and have people from different backgrounds and fields.

You can also look in the Participants Grouped by Title table to help you find potential team

If you are from outside UAB and we have not yet been in contact with us about XIAS so you can access Cheaha/High-Performance Computing (HPC) at UAB, please contact Zhandos Sembay right away to set up an account.

Obtain a Cheaha account if you plan to do data wrangling or software development and have not done so already. See instructions on requesting one. It is not necessary for all team members to have an account. Users from outside UAB will need to be sponsored for outside access (see #2).

Obtain a GitLab account (https://gitlab.rc.uab.edu). Go to the LDAP tab, use your BlazerID login and password. See #2 if you are from outside UAB. Use the GitLab for code sharing.

Use Box for sharing files and documents (not necessarily code to be shared with the wider community. Box is especially useful for transferring or sharing files that are too large send via email. You can store files for your team here (login UAB BlazerID. We will send you a link to view and edit documents for your team.

Note that if you plan to participate as a expert, panelist, or mentor in a field of expertise but not involving technical aspects such as software development, coding, data wrangling (I am not a technical person), then it may be optional for you to complete all the technical preparations. However, you may learn something, improve your ability to understand and collaborate on technical aspects, and perhaps even move toward becoming technically savvy yourself.

Detailed patient data for UAB patients will be available to participants from UAB for whom that data sharing is permitted. Unfortunately, participants from institutions outside UAB will not be able to access this data.

The datasets will not include overt identifiers (name, medical record number, etc) but may include zip codes and dates related to clinical values. This form of dataset is considered to be a HIPAA Limited Dateset and requires some restrictions on access and use. As such, per IRB and institutional policy, release of of datasets requires HIPAA training, IRB Human Subjects Protection training and an Acknowledgement of Restrictions on the Use of Limited Datasets.

Prior to the Hackathon any participant from UAB seeking access to Clinical datasets from UAB Medicine must complete the following (if you have not previously done so or if your training is out of date). If you are from outside UAB, we will work with you to gain access to necessary resources (see #3).  If IRB Human Subject Protection training and HIPAA Training have not been completed those are required as well, but those are typically in place as a matter of normal practice for those involved in research. See below for instructions on how to access and receive training for these resources:

  1. See the CCTS i2b2 web page, has more information about i2b2 and how to Complete HIPAA training in the UAB Learning Management System. The link can be found here.
  2. Complete IRB Human Subjects Protection training. You can complete this through the UAB Learning Management System or CITI. See information and options for IRB initial training or IRB continuing training. If you are new to UAB, you can use training from your previous institution or other approved organizations. We recommend taking the CITI course, which has modules that are common to many research institutions.

Complete the “Acknowledgement of Restrictions on the Use of Limited Data Sets” in the UAB Learning Management System. This link will take you to the course in the campus-side training system. Email the certificate to i2b2support@uabmc.edu.

We would like to get everyone thinking about and preparing their projects for the upcoming hackathon as early as possible. See guidelines on this page describing characteristics of good projects. See the Projects table in the Hackathon Airtable base for projects and project ideas. The table includes detailed information about the project and individual(s) who submitted the project. Researchers have also indicated the suitable of the project and assessments of readiness of the project. Feel free to take any of these and revise as you wish (e.g., narrow the scope, focus on one aspect or use case).

 

If you have your own project in mind, you can enter it directly in the Projects table on the Cancer Hackathon Airtable base, or you can submit it through our project submission form on this web page or via email. Please include detailed information about the project, such as what your research question is and how you would answer the question. Also include any relevant publications and datasets (from UAB and/or publicly available).

Please sign up for teams in the Teams table in Airtable. Teams will ideally have about 2-4 members and have a mix of backgrounds and areas of expertise. You may wish to pair up with people you know, people with common interests, and/or people with complementary backgrounds.

Each team will work on one project, but multiple teams may work separately on the same project – they may use different approaches.

Teams may have members from multiple organizations. At least one member of each team must be affiliated with a CTSA site.

Sign up for your project and team and/or invite others to join your team (preferred). If you do not organize or join a team, we, will do our best to match people with projects and teams.

Stage 5 - Attend Hackathon

Hackathon Day 1
Monday, August 9, 2021
9:00 am – 5:00 pm

Hackathon Day 2
Tuesday, August 10, 2021
9:00 am – 5:00 pm

Stage 6 - Project Showcase, Competition, and Awards

Project Showcase and Competition

Project Showcase

Friday, August 13, 2021

10:00 am – 12:00 pm


Join us as hackathon participants present their work. After the showcase, judges will watch presentations and evaluate projects and presentations.

Award Presentations

Closing Ceremony and Presentation of Awards

Friday, August 20, 2021

10:00 am – 11:00 am


We will celebrate all that we have accomplished together. We will review and reflect on the Hackathon and its impact and acknowledge everyone’s contributions. Finally, we will present awards to the winning teams.

Sponsors

Organizing Commitee

Chairs

Amy Wang, M.D., MBI

Associate Professor
Departments of Medicine and Family and Community Medicine
UAB School of Medicine

Co-Chairs

Jake Y. Chen, Ph.D.

Professor
Departments of Genetics, Computer Science and Biomedical Engineering
Chief Bioinformatics Officer
Associate Director
UAB Informatics Institute

Matt Might, Ph.D.

Professor
Department of Medicine
Director
Hugh Kaul Precision Medicine Institute

Committee Members

Andrew Crouse, Ph.D.

Director of Research and Operations
Hugh Kaul Precision Medicine Institute

Lucio Miele, M.D., Ph.D.

Professor and Department Head
Department of Genetics
Assistant Dean for Translational Science
LSU School of Medicine

Zhongming Zhao, PhD, MS

Chair Professor for Precision Health
Director, Center for Precision Health
School of Biomedical Informatics
University of Texas Health Science Center at Houston

Nafisa Ajala

Informatics Program Specialist
UAB Informatics Institute

Matthew Wyatt, MSHI

Director of Clinical Research Informatics
UAB Informatics Institute

Aik Choon Tan, Ph.D.

Vice Chair
Department of Biostatistics and Bioinformatics
Moffitt Cancer Center

Heather Watts

Communication Specialist
UAB Informatics Institute

Amanda Carballo

Administrator
UAB Informatics Institute