Deep Learning in the Life Sciences
Course Description
This course introduces foundations and state-of-the-art machine learning challenges
in genomics
and the life sciences. We introduce deep learning approaches to key problems,
comparing and
contrasting their power and limitations. We seek to enable students to evaluate
various solutions
to crucial problems we face in this rapidly developing field and to execute new
enabling solutions
that can have a large impact.
Recommended Books
1. Dive Into Deep Learning, A. Zhang, Z.C. Lipton, M. Li, A. Smola.
2. Deep Learning” by Goodfellow, Bengio, and Courville
Course Learning Outcomes:
After completing this course, a student will be able to:
1. Knowledgeable about deep learning methods in life sciences, especially in tasks
like sequence and structure analysis and evolution, biological networks
2. able to understand the key algorithms for the main tasks
3. able to implement and apply the techniques to real-world datasets
Assessment System
Quizzes 10-15%
Assignments 5-10%
Midterms 30-40%
ESE 40-50%
Lecture Plan
S.NO. Description Quizzes Assignment
1. Overview of the Course/ ML Review
2. Neural Networks (CNN (review) 1
3. Neural Networks (RNN & GNN)
4. Neural Networks (Deconvolutional Networks)
5. Interpretability, Dimensionality Reduction
6. Relevance Propagation, Convolution Arithmetic
7. Maximum entropy methods
8. Discriminative Localisation
9.Interpreting ML Models: visualise Filters, Measure
Gradients, Perturb inputs.
10. Tensor Flow Introduction 3
11.Deep Learning Problems and compute solutions;
Genomic regulatory codes
12.Deep Learning Problems and compute solutions;
Gene regulation – Single Cell RNA-seq
13.Deep Learning Problems and compute solutions;
Genetic Variations & Diseases
14.Graphs & Proteins
• Protein structure prediction
• Functional classifications
15.Biomedical imaging
• Video Processing, structure determination