Overview
This course provides an introduction to the rapidly growing field of computational connectomics, focusing on the study of the brain’s complex network of connections using computational, machine learning, and AI techniques. Students will explore the foundational concepts of connectomics, examining both structural and functional connectomes to understand how different brain regions interact to support cognitive functions and behavior. Throughout the course, students will apply various computational tools and machine learning algorithms to analyze connectome data, identify patterns in brain networks, and predict neurological outcomes. The curriculum covers data acquisition, preprocessing, and the integration of multi-modal datasets, while also addressing the ethical and practical challenges of connectome research.
Class meetings: WeFr 1:25 – 2:40 pm in Biological Sciences 130
Instructor: Tananun Songdechakraiwut t.song AT duke.edu
Canvas: Instructor will regularly share important announcements on Canvas. Therefore, it’s crucial to check the Canvas page frequently throughout the term.
Prerequisites:
- COMPSCI 201
- Proficiency in Python, R, or Java
Office Hours
Office hours will begin on Friday, January 17, and continue through Wednesday, April 23. Please be advised that office hours will not be held during university holidays, such as spring break. Individual meetings may be arranged upon request.
| Date & Time | Location & Room |
|---|---|
| We 5 - 6pm | LSRC D112 |
| Fr 5 - 6pm | LSRC D112 |
Grading
The following weights are used:
- 70% for team project
- 20% for journal club presentation
- 10% for attendance
Project
For project, students will apply computational techniques to explore and understand network behavior in the human brain. They will then craft a compelling narrative to effectively communicate these findings in a clear and engaging manner. The focus of the project should be on extracting meaningful interpretations from the data, rather than developing complex algorithms or implementing advanced tools.
Projects can be completed individually or in pairs, and teams are formed by the students themselves.
Throughout the semester, there will be five deliverables:
- Proposal
- First Milestone Report + Lightning Presentation
- Second Milestone Report + Video
- Team Project Presentation
- Final Report + Reproducible Code
In the week leading up to each deliverable (excluding the final report and reproducible code), students will need to meet with the instructor to discuss their progress. If the project is completed in pairs, both team members should attend these meetings, as attendance and participation will be factored into their team project grade.
All five deliverables should be submitted via Gradescope. For each deliverable submitted as a pair, only one submission is needed, and both team members should be tagged.
Journal Club Presentation
Students will select an article and share it with the class at least one week before their scheduled presentation. On the presentation day, they will give a slide presentation that covers the background, methods, results, and conclusions of the paper. They should also provide a critique, explore any unanswered questions or future directions, and lead a discussion with the class. Students may work individually or in pairs.
A list of research articles will be provided on Canvas. Students may choose from this list or select their own articles of interest.
Attendance
Students are expected to attend at least 80% of classes, allowing up to 20% to be missed. We will use the Attendance tool on Canvas to track attendance.
Collaboration and Ethics
Team assignments are collaborative efforts, and submissions should reflect the collective understanding of all team members. All code and written responses must be original. Please write the code yourself and be able to explain or reproduce all results you submit. Please acknowledge any external collaborators and sources you consulted in your submission. If you have any questions about collaboration, the instructor is available to provide clarification.
Academic Integrity
Upon enrollment, each student commits to maintaining the high academic standards of Duke University. Academic misconduct refers to behavior that adversely affects the institution’s integrity. This encompasses activities such as cheating, fabrication, plagiarism, unauthorized collaboration, and assisting others in these actions. Engaging in misconduct may lead to disciplinary measures, including but not limited to failure on the assignment/exam/course, written reprimand, disciplinary probation, suspension, or expulsion.
