Course syllabus adopted 2025-02-03 by Head of Programme (or corresponding).
Overview
- Swedish nameArtificiell intelligens för molekyler
- CodeDAT675
- Credits7.5 Credits
- OwnerMPDSC
- Education cycleSecond-cycle
- Main field of studySoftware Engineering
- DepartmentCOMPUTER SCIENCE AND ENGINEERING
- GradingTH - Pass with distinction (5), Pass with credit (4), Pass (3), Fail
Course round 1
- Teaching language English
- Application code 87140
- Maximum participants50 (at least 10% of the seats are reserved for exchange students)
- Open for exchange studentsYes
Credit distribution
Module | Sp1 | Sp2 | Sp3 | Sp4 | Summer | Not Sp | Examination dates |
---|---|---|---|---|---|---|---|
0125 Written and oral assignments 3 c Grading: TH | 3 c | ||||||
0225 Project 4.5 c Grading: TH | 4.5 c |
In programmes
- MPAEM - Materials Engineering, Year 1 (elective)
- MPCAS - Complex Adaptive Systems, Year 1 (elective)
- MPDSC - Data Science and AI, Year 1 (elective)
Examiner
- Rocio Mercado
- Assistant Professor, Data Science and AI, Computer Science and Engineering
Eligibility
General entry requirements for Master's level (second cycle)Applicants enrolled in a programme at Chalmers where the course is included in the study programme are exempted from fulfilling the requirements above.
Specific entry requirements
English 6 (or by other approved means with the equivalent proficiency level)Applicants enrolled in a programme at Chalmers where the course is included in the study programme are exempted from fulfilling the requirements above.
Course specific prerequisites
To enroll in the course, students should have completed and passed the following:- DAT565 Introduction to Data Science and AI or equivalent
- At least one other formal programming course
Aim
This course provides a comprehensive introduction to the transformative role of AI in molecular sciences. Structured into four modules, it covers molecular representations, property prediction, and generative AI techniques, culminating in a hands-on project tackling real-world challenges.Emphasizing practical skills, the course integrates programming, machine learning frameworks, and data handling best practices. Designed for students with a solid programming foundation, it bridges diverse scientific backgrounds to explore the intersection of AI and molecular science.
Learning outcomes (after completion of the course the student should be able to)
Knowledge and Understanding
On successful completion of the course, students will be able to:
1. Articulate the role of molecular data in addressing global challenges such as healthcare innovation, sustainability, and biotechnology, and explain its distinct importance compared to other areas of data science.
2. Describe the foundations of cheminformatics and the evolution of AI methods for molecular applications, highlighting historical milestones and technological advancements.
3. Explain the principles of molecular data representation and analysis, including key challenges, pitfalls, and best practices when working with this specialized data.
4. Demonstrate knowledge of state-of-the-art AI methods for molecular property prediction and molecular generation, including their strengths, limitations, and potential for industry applications.
5. Critically evaluate AI models in the context of molecular science, considering data quality, biases, and the broader implications of model design and performance.
Skills and Abilities
On successful completion of the course, students will be able to:
1. Preprocess, analyze, and manage molecular datasets, ensuring data quality and addressing common challenges such as data imbalance and representation biases.
2. Design and implement machine learning pipelines tailored to molecular property prediction and generative tasks, considering real-world constraints such as missing data, data sparsity, computational resources, and model complexity.
3. Apply advanced techniques in optimization, statistics, and algorithm development to molecular AI tasks and analyze the results in a meaningful and reproducible manner.
4. Use modern programming tools and libraries (e.g., PyTorch, RDKit, scikit-learn, HuggingFace) to develop scalable and efficient molecular AI workflows, while adapting to emerging technologies in the field.
5. Communicate findings, insights, and implications of molecular AI research effectively to interdisciplinary audiences, fostering collaboration across fields.
Judgement and Approach
On successful completion of the course, students will be able to:
1. Critically assess molecular AI models for reliability, reproducibility, scalability, and applicability to real-world problems across industries.
2. Reflect on the ethical, societal, and environmental implications of AI-driven solutions in molecular sciences, contributing to the responsible development and application of these technologies.
3. Evaluate the strategic potential of integrating molecular AI in solving future global challenges, identifying opportunities for innovation and cross-disciplinary collaboration.
4. Foster a mindset of continuous learning and critical inquiry, recognizing the evolving nature of AI technologies and their applications in molecular sciences.
Content
This course explores the transformative role of artificial intelligence (AI) in molecular sciences, offering students a comprehensive introduction to the principles, methods, and applications of AI for molecules.The course is structured into four modules, beginning with an introduction to the field and machine representations of molecules (Module 1). Building on this foundation, the course then delves into molecular property prediction using machine learning (Module 2), focusing on best practices for data preparation and constructing predictive models. In Module 3, students explore generative AI techniques for designing novel molecules, equipping them with skills in molecular engineering. Finally, the course culminates in a hands-on project (Module 4), where students apply their knowledge to tackle a real-world challenge in molecular engineering.
The course emphasizes practical skills development, integrating programming, cheminformatics, machine learning frameworks, and good data handling practices. It is designed for students from diverse scientific backgrounds who already have a solid foundation in programming and would like to learn more about methods at the intersection of AI and molecular science.
Organisation
The course consists of 16 lectures (two 90-min lectures per week), 2 lab sessions per week, and a final project. The following components of the course are compulsory:- 2 assignments
- Project proposal
- Final project report
- Final presentation
Literature
The course uses open-source resources (free e-book, blogs, and scientific articles) as part of the course literature.Examination including compulsory elements
The grading scale (5, 4, 3) comprises:- 2 assignments (5 points each)
- Project proposal (5 points)
- Final project report (5 points)
- Final presentation (5 points)
- 5: 24-25 points (out of 25)
- 4: 21-23 points
- 3: 17-20 points
- U: <17 points
The course examiner may assess individual students in other ways than what is stated above if there are special reasons for doing so, for example if a student has a decision from Chalmers about disability study support.