Module: IN2393

Credit: 6 ECTS

Lecturer: Julien Gagneur, Matthias Heinig, Johannes Linder, Annalisa Marsico

Lecture: Tuesdays, 14:00 - 15:30, starting on 14th April 2026, Lecture hall 2 "Interims I", Boltzmannstr. 5 in 85748 Garching

Exercise: Thursdays, 12:00 - 14:00, starting on 16th April 2026, Lecture hall 5901.EG.051, Hans-Piloty-Str.1 in 85748 Garching

 

Note: Due to the MVG strike, the lecture on Tuesday 14th April 2026 is postponed to Thursday 16th April 12:00-14:00. Lecture hall 5901.EG.051, Hans-Piloty-Str.1 in 85748 Garching.
 

Lecture Language: English

Prerequisite (recommended):

• One introductory lecture on machine learning (e.g IN2064; MA4802)
• Strong interest in biological and biomedical research questions
• Basics in python programming

Who can attend

Generally, the module is geared toward students from bioinformatics, computer science, as well as other students with quantitative training (physics, applied maths) and an interest to dive into molecular biology. Students from biology or medicine are welcome guaranteed they have some background in machine learning (see above) and no inhibition with basic programming.

The module is an elective module in the catalogue of:

• MSc Bioinformatics
• MSc Informatics
• MSc Information Systems
• MSc Informatics: Games Engineering
• MSc Data Engineering and Analytics
• MSc Physics

Intended Learning Outcomes:

Gene expression refers to how cells read the information encoded in genomes. At the end of the module, students are able to:

• Explain the major steps of gene expression, from DNA accessibility to protein abundance.
• Explain genome-wide experimental assays used to measure different stages of gene expression.
• Explain the principle and applications of massively parallel reporter assays.
• Apply deep learning methods for sequence-based prediction tasks in regulatory genomics.
• Apply and interpret model interpretation techniques to analyze learned sequence features.
• Apply convolutional neural networks and transformer-based models to genomic sequence data.
• Evaluate the performance of deep learning models for genome-wide assays.
• Interpret model predictions in a biological context and discuss limitations of the applied approaches.

Content:

Gene expression refers to how cells read the information encoded in genomes. This lecture introduces biological and computational concepts to study gene expression. It consists of two parts:

(1) 6 lectures introduce biological mechanisms, experimental assays, and computational models for regulatory genomics. The six lectures are supported with modeling exercises in python.

(2) A 7-8 week hands-on project

The lectures are organized around steps of gene expression:

• Introduction to gene regulation and sequence-based computational models of gene regulation
• Transcriptional regulation
• Chromatin-mediated regulation
• RNA splicing
• RNA modification and degradation
• Translation

 Over these lectures, computational methods are introduced including:

• Fitting procedures of deep neural network
• Convolutional Neural Networks
• Transformers
• Embeddings for sequence data
• Multi-task learning and transfer learning
• End-to-end learning
• Analytical and visualisation techniques for model interpretation

Beschreibung der Studien-/Prüfungsleistungen:

Students are evaluated individually based on a project work supervised by members of the teaching team.

The project is structured into several phases, including initiation and problem definition, distribution of roles within the group, definition of evaluation criteria and baseline approaches, iterative development and implementation of modeling ideas, evaluation of results, and identification of limitations. The project concludes with an oral presentation and a written report.

Assessment is based on:

• Final presentation (10 minutes): evaluation of clarity of presentation and slides, correct application and explanation of relevant concepts and methods (see learning outcomes), and achieved results.
• Written project report (maximum 20 pages): evaluation of conciseness, language quality, and correct application and explanation of the employed methods and concepts.

The final presentation includes questions related to the presentation and the written report. These questions assess the student’s understanding of the applied methods, data, and results, and ensure that the use of permitted tools, including AI-assisted tools, has not replaced the student’s own understanding.