Hands-On Introduction to Machine Learning and Deep Learning for Biology

The aim of this course is to introduce the participants to machine learning techniques that can be applied to address biological questions. After the course, participants will be able to apply the techniques learned to their data sets, solving classification and regression problems using machine learning.
Content:
1. A gentle introduction to Python 3, including: syntax, data structures, conditionals, loops, functions, object-oriented programming , and libraries (Numpy, Pandas, SciPy, Matplotlib) (3 hours, optional)
2. Supervised learning, including: algorithms (decision trees, ensemble learning, k-nearest neighbors, neural networks), cross validation, grid search, libraries (Sklearn) (7 hours)
3. Unsupervised learning, including: clustering algorithms (k-means, GMM) and dimensionality reduction (Principal Component Analysis, Independent Component Analysis, Randomized Projections, manifold learning) (7 hours)
4. An introduction to deep learning, including: models, libraries (Pytorch, Keras) (3 hours)
All lessons will interleave theoretical and practical activities. Session 1 is optional, to introduce students without prior programming experience in Python. Sessions 2-4 will each analyze a case study using phenotypic or genomic data