Deep Learning for Genomics
Welcome to the Special Course in Machine Learning in Spring 2025! This course is designed to provide practical insights into the potential and limitations of deep learning in genomics. This is done by reading, presenting and discussing impactful papers of the domain. On top of that, we will also provide some practical tasks.
This course will focus on:
- Understanding Deep Learning: The basics of how deep learning works and understanding different neural network architectures.
- Popular Genomic Assays: Understanding the fundamentals of RNA-seq, ChIP-seq, and DNase-seq, and how these datasets can be integrated with deep learning.
- Genomic Applications: Using deep learning for gene expression analysis, splicing predictions, and modelling chromatin activity.
- Limitations: Addressing the challenges and boundaries of deep learning in genomics.
- Interpretability: Techniques for understanding and explaining model predictions in a biological context.
Structure
There will be two types of seminars: paper seminars, which will take the form of discussions centered around pre-selected papers, and coding seminars, which will demonstrate the discussed models in action.
Prerequisites
We expect you to be comfortable with:
- Python
- Basics of machine learning
We do not expect the students to have a strong background in genetics/genomics.
Organization
Seminars are scheduled on Tuesday 16.15 - 18.00 in 1019 (most current info is on Slack), Delta (Narva mnt 18).
Zoom: link
We use Slack for communications: link.
NB!: the first seminar will take place on February 11th, 2025!
Grading
To pass the course, students need to
- Attend > 50% (6) seminars + submit the paper questions before the seminars
- Submit 3 homeworks
Contacts
- Dzvinka Yarish (dzvenymyra-marta.yarish@ut.ee)
- Kaur Alasoo (kaur.alasoo@ut.ee)