The Personal Genome: AI & Precision Medicine

Data

Official data in SubjectManager for the following academic year: 2026-2027

Topic

The era of "one-size-fits-all" medicine is ending. This course explores how AI and Genomic data are used to tailor medical treatment to the individual characteristics of each patient. Instead of guessing which drug will work, we focus on how machine learning models predict disease risk, select the most effective drugs (Pharmacogenomics), and discover new therapeutic targets that human eyes might miss. You will learn how the "Biotech-AI" partnership is turning medicine into a precise, personalized science.

Lectures

• 1. P4 Medicine: Introduction to Predictive, Preventive, Personalized, and Participatory healthcare. - Zand Afshin
• 2. The DNA Filter (SNPs): Why some drugs are "miracles" for some patients but "poison" for others. - Zand Afshin
• 3. Choosing the Dose: Using genetics to find the perfect drug concentration for Oncology and Psychiatry. - Zand Afshin
• 4. Liquid Biopsies: How AI finds "needles in a haystack"—detecting cancer markers in a simple blood sample. - Zand Afshin
• 5. AlphaFold & Protein Logic: How AI predicts the 3D shape of proteins in seconds to help design better drugs. - Zand Afshin
• 6. AI Image Recognition: Training computers to "see" disease in microscope slides and cell cultures. - Zand Afshin
• 7. The Gut Connection: How your microbiome influences the success of therapy and personalized nutrition. - Nancy Zeineddine
• 8. Solving Rare Mysteries: Using global databases and AI to diagnose diseases that have no name. - Nancy Zeineddine
• 9. Digital Twins: Creating virtual models of a patient to test drug reactions before they take a pill. - Nancy Zeineddine
• 10. The Big Data Puzzle: How we combine DNA, RNA, and protein data into one clear health picture. - Nancy Zeineddine
• 11. Personalized Vaccines: Exploring mRNA technology and AI-designed vaccines for cancer. - Nancy Zeineddine
• 12. Future Brainstorming: Interactive Summary on the ethics of knowing your genetic future. - Nancy Zeineddine

Practices

Seminars

Reading material

Obligatory literature

Literature developed by the Department

Course materials consist of weekly presentation slides covering the lecture topics. These are supplemented by curated "Fact Sheets".Additionally, a "Curriculum Question Bank" consisting of 25 comprehensive questions will be distributed at the start of the semester to guide self-study and provide the basis for both the "Group-Pulse" interactive activity and the final written assessment.

Notes

Recommended literature

Conditions for acceptance of the semester

One assignment at the end of the course

Mid-term exams

It's not required

Making up for missed classes

Not possible

Exam topics/questions

Logic: The "Why" of AI in Medicine

1.            Why is AI better than a human at finding patterns in a dataset of 3 billion DNA letters?

2.            What is the logical difference between "Reactive" medicine and "Predictive" (P4) medicine?

3.            Why do we need AI to help us understand protein folding (AlphaFold) rather than using traditional lab methods?

4.            How does the logic of "Digital Twins" reduce the risk of clinical trials?

5.            Why is a simple blood test (Liquid Biopsy) replacing invasive surgeries for cancer tracking?

II. Application: Real-World Personalization

1.            In Pharmacogenomics, how can a single DNA letter change (SNP) decide if a patient gets a side effect?

2.            Describe a scenario where AI could help a doctor choose the right antidepressant in one day instead of months.

3.            How can AI analyze a patient's gut microbiome to improve their cancer treatment?

4.            Why is AI essential for diagnosing rare "orphan" diseases that local doctors have never seen?

5.            How can personalized mRNA vaccines be "programmed" to attack a specific patient’s tumor?

III. Mechanism: How the Tech Works

1.            Explain how a "Polygenic Risk Score" combines many tiny genetic variations to predict a major disease.

2.            What is the mechanism by which AI "learns" to identify cancerous cells in a medical image?

3.            How does AlphaFold use existing protein data to predict the structures of unknown proteins?

4.            Describe how a "Microfluidic Chip" helps in collecting data for Multi-Omics integration.

5.            How does a "Pharmacogenomic Panel" work in a hospital lab to guide drug prescriptions?

IV. Diagnostics: Finding the "Needle"

1.            What are the "Red Flags" an AI looks for in a genomic sequence to identify a high-risk mutation?

2.            How do we ensure that an AI model is "unbiased" and works for patients of all ethnic backgrounds?

3.            Why is "Data Quality" more important than "Data Quantity" when training a medical AI?

4.            How can AI help differentiate between "noise" and "real disease signals" in a blood sample?

5.            Why is "Real-Time Tracking" through wearable devices becoming a part of precision medicine?

V. Ethics & The Future: Knowing the Code

1.            Is it ethical to know your risk for a disease (like Alzheimer’s) if there is currently no cure?

2.            Who should own your genomic data—you, the hospital, or the company that analyzed it?

3.            How do we protect "Genomic Privacy" in an age where hacking is a constant threat?

4.            Should parents be allowed to see the full "Precision Health" report of their unborn children?

5.            How can we make sure that AI-driven medicine is available to everyone, not just the "genetically wealthy"?

Examiners

Instructor / tutor of practices and seminars