Week 13: AI, SynBio, and Scaling Health Innovation (ARPA-H)

w13h1 w13h1

Bio design living Materials

This week covers designing, programming, and fabricating engineered living materials — such as self-healing concretes, adaptive biofilms, and responsive biomaterials — by integrating genetic circuit design, materials science, and bioprocess engineering

Some thoughts about the lecture / Week

  • This week made me realize that AI in biology is not only about predicting proteins or analyzing sequences, but also about understanding how scientists actually work in the lab. Small details like pipetting angle, bubbles, or viscosity can completely change experimental results.

  • One thing I found interesting was the idea of “sensorized laboratories,” where cameras and computer vision systems record experiments from multiple perspectives. It showed how future AI models may learn directly from real laboratory behavior rather than only from published papers.

  • I also learned that scientific publications usually report only successful methods, while failed experiments and troubleshooting are rarely documented. This creates limitations for training AI systems because models do not learn the real-world mistakes scientists encounter.

  • The lecture highlighted how standardization and automation could improve reproducibility in biology and healthcare innovation, especially during emergencies like COVID-19 testing and vaccine distribution.

  • I liked the idea of “gamifying science” through leaderboards and collaborative optimization. It made science feel more interactive and community-driven, while also encouraging better experimental practices and shared learning.

Final Project:

This week, I was adjusting some of my DNA design for a biosensor. The progress was a bit chaotic in making the proper documentation, but I made it. Next week, it’s going to be the slides of the final presentation and some thoughts I had after presenting my idea on an international field! Thanks for reading!