AI Tools for Teaching and Learning

CSC 196A: Computational Biology

Grok AI, Code Generator, Automatic Data Collection

In CSC 196A: Computational Biology, I introduced a new instructional technique aimed at enhancing student engagement and promoting AI-driven data analysis skills. This change involved assigning a short project in which students used Grok AI to generate an R script for downloading data from public bioinformatics repositories using provided accession numbers. They then analyzed the downloaded data and summarized their findings in a report. This assignment was designed to integrate computational thinking with practical data science workflows relevant to the field.

The activity took place during the ninth week of in-person instruction and was recorded during the final 10 minutes of a 75-minute lecture. Most students enrolled in the course were in their third or fourth year of undergraduate study, making them well-positioned to take on this challenge. The response to the project was positive—students found it engaging and enjoyable. Many appreciated the opportunity to apply AI tools in a real-world context and felt that the hands-on nature of the assignment made the concepts more accessible.

Direction:

This change reflects my ongoing effort to align course activities with the broader objective of AI literacy in bioinformatics. As I continue to refine this approach, I’m particularly interested in understanding whether such assignments effectively support students’ learning goals related to AI implementation and data analysis in scientific contexts.

AI-generated simulated clinical data for bioinformatics tools has greatly assisted my work. Without sensitive or hard-to-access patient data, I can test algorithms and refine bioinformatics pipelines using artificial but realistic data. This method lets me test these tools in a controlled environment, which is useful for genomics and precision medicine. I am aware of the ethical issues surrounding AI content generation, especially with clinical data. Privacy is a major issue. Simulations don't match real patients, but if the AI model was trained on sensitive data, patterns or features from real patients may appear in the synthetic output. This could cause re-identification risks if
synthetic data is too similar to real data. Another issue is data integrity. If not validated, AI-generated data may contain biases or unrealistic variations that skew results. A few ethical best practices help me address these concerns. I disclose how the data is generated and that it is synthetic and unrelated to real people. To make synthetic data reliable for testing, I rigorously validate it to match clinical data statistical properties. Finally, I follow strict data usage and privacy guidelines, especially when handling real-world data for model training to avoid identifiable information in synthetic outputs.

Implementing this AI-integrated data collection assignment significantly enriched both my teaching and student engagement in CSC 196A. The activity prompted me to shift from a traditional lecture format toward a more inquiry-based and student-centered approach. It allowed me to focus on guiding students through the practical challenges of using artificial intelligence for bioinformatics tasks, rather than simply covering theoretical content. Evidence from student submissions showed clear improvement in data interpretation skills. Reports reflected thoughtful analysis and correct integration of both the AI-generated code and the downloaded datasets. Several students went beyond the assignment requirements by customizing their scripts or exploring additional datasets, indicating genuine interest and engagement. This practice reaffirmed the value of incorporating applied, AI-driven activities into STEM education to foster both technical competence and enthusiasm for the subject matter.

My experience with AI has been both enriching and transformative, offering new ways to streamline research and enhance teaching. AI-generated simulated clinical data has proven invaluable in testing bioinformatics tools, allowing me to refine algorithms and
pipelines without relying on sensitive patient information. This controlled testing environment ensures ethical compliance while fostering innovation in fields like genomics and precision medicine. Beyond research, AI has greatly supported my teaching
efforts. Using tools like ChatGPT to create materials, such as a K-Nearest-Neighbor clustering slide, has streamlined content generation and improved clarity in instruction. This efficiency has enabled me to focus on delivering engaging and impactful lessons. Encouraging students to use AI ethically—for brainstorming, drafting, or initial ideation—has further highlighted its potential to enhance learning outcomes

During my own "Observe and Analyze Session," one key takeaway was recognizing how even a brief, well-structured activity can dramatically shift student engagement and focus. I noticed how students became more animated and collaborative when given the AI-guided assignment. What stood out to me was the value of creating space for students to struggle productively with real data using authentic tools. I realized that I tend to over-explain concepts at the start of class; however, in this session, the learning truly happened when students explored on their own using Grok AI.

Evidence of Instructional Change

Evidence of Instructional Change
Example 1	Example 2	Example 3
Assignment posted during semester asking students to collect data using generative AI	Generative AI helps CS instructor to quickly generate code to plot figures that can be used to illustrate clustering algorithm	Code generated by AI

Practice and Reflect Teaching and Learning with AI:

The AI Tools for Teaching & Learning Workshop from the Chancellor's Office has significantly influenced how I approach course design and instructional delivery. The workshop expanded my understanding of AI not just as a technical tool, but as a pedagogical asset that can support personalized learning, streamline content delivery, and foster higher-order thinking.

One of the main opportunities I discovered was the potential for AI to reduce routine instructional overhead—such as generating code templates or summarizing content—allowing both instructors and students to focus more on analysis and critical thinking. However, a key challenge lies in ensuring that students use AI tools ethically and meaningfully, without over-relying on them as shortcuts. This requires continuous guidance and clear scaffolding around AI tool usage.

Course Syllabus

Syllabus_CSC196A.pdf

Faculty Bio

Faculty Bio
Faculty Picture	Faculty Professional Background
	Dr. Bang Tran (he/him) is a faculty member in the Computer Science Department. He received his B.S.C from the Asian Institute of Technology, Thailand in 2010, and M.S. from Thammasat University, Thailand, and Tokyo Institute of Technology, Japan, and his Ph.D. from the University of Nevada, Reno. His research foci include single-cell data deconvolution and systems-level analysis. Read more: https://webpages.csus.edu/s.tran/

Funding Source
	This work was supported by: US Department of Education P031S230256 and P031C210012 *Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or the US Department of Education.