CHEM 169 / 269

Applied AI & Machine Learning for Biochemistry Winter 2026 Head Routesetter: Adrián Jinich Format: In-person, studio-style ("climbing gym" model)

Course Overview

This course introduces modern AI and machine-learning tools as they are actually used in contemporary biochemistry and molecular biology. Rather than focusing on abstract theory or rote implementation, the course emphasizes hands-on problem solving, tool fluency, and learning by doing.

You will work through a series of short, self-contained exercises ("routes") organized into thematic groups ("walls"), similar to a climbing gym. Each route targets a specific skill or concept, and you are encouraged to choose your own path through the material while meeting core expectations. No prior machine learning experience is required. Prior programming experience is helpful but not required.

Learning Goals

By the end of the course, you should be able to:

• Work comfortably in Python for data analysis
• Load, clean, and manipulate biological datasets
• Use common ML and data-science libraries
• Read and reason about AI/ML workflows used in modern research
• Build and explain small, working computational analyses
• Reflect productively on your own learning process

Course Structure: The Climbing Gym Model

Instead of traditional lectures and weekly problem sets, this course is organized around:

• Walls: thematic areas (e.g., Python basics, data manipulation, representations)
• Routes: individual exercises you complete at your own pace
• Climbs: your attempts at routes, documented via notebooks and short reflections

Some routes are expected for everyone. Others will be stretch routes for students who want to explore further. Routes are released iteratively during the quarter, at every session.

What a Route Submission Includes

Most routes require submitting:

1. A runnable Colab/Jupyter notebook
   • Must run top-to-bottom
   • Partial solutions are acceptable
2. A short logbook reflection (plain text)
   • What was confusing?
   • What helped?
   • What did you try?

Effort, honesty, and reflection matter more than perfection. This the key feedback that will help us shape the course as it evolves. You are the key users, so your feedback is the heart of the whole course.

Grading Policy (This is Active but Evolving)

Important: This grading structure is active but may be refined in early weeks. Any changes will be announced clearly and will never penalize students retroactively.

Grades are based on engagement, progress, and evidence of learning, not speed or prior experience.

Grading Components

1. Route Completion & Attempts (~40%) Submitting route notebooks and logbooks Credit for good-faith attempts You are not expected to perfectly solve every route

2. Logbook / Reflection (~15%) Short written reflections submitted with routes Used to understand learning progress and improve the course Treated as a first-class component of the grade

3. Comps (In-class Challenges, i.e. the Midterm) (15%) Time-boxed, hands-on route challenges during class Open-note, tool-friendly Focus on reasoning and problem-solving, not memorization

4. Final Assessment (15%) The final assessment for this course will focus on integrating skills developed through the routes, including:

   • working with data in Python,
   • reasoning through computational problems,
   • using computational tools appropriately,
   • and clearly explaining your approach.

   The exact format of the final assessment is still being finalized and may take one of several forms, such as:

   • a multi-route "comp" (capstone climb),
   • a short project that extends an existing route,
   • or a guided, multi-stage assignment.

   Regardless of format:

   • expectations will be clearly communicated well in advance,
   • grading criteria will be transparent,
   • and the assessment will reward process, reasoning, and effort, not just final results.

   You will not be expected to independently invent a large research project or master entirely new tools at the last minute.

5. Attendance & Engagement (15%) This is a studio-style, in-person course where learning happens through active work during class. Regular attendance and good-faith engagement therefore contribute approximately 15% of the final grade.

   Students who attend class consistently and participate in good faith will receive full credit in this category. Occasional absences for reasonable reasons are expected. Repeated or sustained absence may reduce credit, as it limits participation in in-class routes, discussions, and comps.

What This Course Is (and Is Not) This course is:

• Hands-on
• Iterative
• Designed for learning by doing

This course is not:

• A traditional lecture course
• A memorization-based exam course
• A "cram before the midterm" course

Tools & Platforms

You will use:

• Google Colab / Jupyter notebooks
• Python data-science libraries
• Google Docs (temporary route drafts)
• Canvas (official course communication)

A dedicated course "gym portal" is under active development.

A Final Note

If you:

• show up,
• attempt routes honestly,
• submit notebooks and reflections,
• and ask for help when stuck,

you can do very well in this course, regardless of prior experience.