Proteins form the molecular basis of all life, and yet, we lack reliable means to predict what they do, how they do it, where and when. Indeed, although their primary constituents - amino acids - are well-characterized, the size and diversity of proteins make physical simulations of their behavior extremely challenging. To overcome these challenges, Machine learning approaches have become increasingly popular since the 90’s for predicting how proteins fold, recognize one another and function. While initially only moderately successful, deep learning has enabled radical progress across tasks, culminating with the celebrated AlphaFold algorithm for protein structure prediction in 2021. This workshop is a joint introduction to machine learning and structural biology, with an emphasis on practical tools and current research topics. Prior knowledge of either domain is desirable but not necessary. In the first part of the semester, we will cover key concepts including protein structures & folding, supervised learning and deep learning. Then, we will discuss the biological and algorithm concepts underlying AlphaFold. The second part of the semester will be dedicated to group research projects related to the topic. Evaluation will be based on the project reports and presentations. The course will be in English.

Computational Genomics

Compulsory Course (Bioinformatics undergraduates); Course jointly taught with Prof. David Burstein and Prof. Irit Gat-Viks., Tel Aviv University, School of Computer Science, 2024

Summary

Biological and medical research has undergone a revolution following the Human Genome Project: First, a single genome was sequenced, and today tens of thousands of human genomes are sequenced every year. At the same time a huge number of genomes of other species (animals, plants, bacteria, viruses) are sequenced. State-of-the-art experimental technologies are evolving and producing information that enables revealing revolutionary insights that will impact our lives, our health and the world around us. Making use of these technologies and analyzing their experimental results requires advanced computational methods. To a large extent the bottleneck of the analysis has shifted from the production of the data to its analysis.

Introduction to Machine Learning

Compulsory Course (Computer Science undergraduates), Tel Aviv University, School of Computer Science, 2024

Summary

This course provides a basic introduction to machine learning (namely the field of computer science studying algorithms that learn from examples). This field underlies modern applications of AI, including machine vision, natural language processing, autonomous driving and others. The course will provide the theoretical foundations for understanding learning algorithms, describe different algorithms, and provide practical experience. Some of the topics studies are:

Supervised learning: PAC learning, VC dimension, perceptron, SVM, stochastic gradient descent, boosting, deep learning and decision trees
Unsupervised learning: principal component analysis, clustering, EM algorithm

Computational Structural Bioinformatics

Elective Course (Bioinformatics / Life Science undergraduates); Course jointly taught with Prof. Nir Ben Tal., Tel Aviv University, School of Computer Science, 2025

Summary

Despite being made up of only ~20 building blocks, proteins have an astounding diversity of shapes and molecular functions. This course is an introduction to Computational Structural Biology - computational methods for studying the link between the sequence, structure and function of biomolecules. After providing background on protein structures, we will present classical and recent algorithms for tackling major structural biology tasks, and demonstrate their applications. The objectives of the course are to:

Study the scientific and algorithmic principles of key tools such as AlphaFold, learn how and when to use them and how to interpret their results.
Deepen our knowledge of protein structures & function.
Study some real-life applications of dynamic programming, deep learning & computer vision algorithms for solving biological problems.
Apply the learned tools during a project on a chosen system of interest.

Advanced topics in Machine Learning for Computational Biology

Elective Course (CS/Bioinformatics M.Sc. open to undergraduates), Tel Aviv University, School of Computer Science, 2025

Summary

Data-driven science is a ubiquitous paradigm in modern biology: heaps of data are collected on a biological system of interest, from which one must discover qualitative scientific insights or build accurate quantitative models. This course is an overview of advanced machine learning algorithms commonly used in modern computational biology research. The goals are three-fold: i) Learn the underlying mathematical principles behind these algorithms ii) Learn how and when to use them for scientific purposes iii) Understand their limitations. The algorithms will be illustrated on various biological systems (brain recordings, single cell data, protein sequences, molecules, etc.). No prior knowledge of biology is required. Course will be given in English. Evaluation will be based on oral presentation of research articles and home assignments. Tentative syllabus below, is subject to changes.

Jérôme Tubiana

Posts by Collection

portfolio

publications

talks

teaching

Summary

Summary

Summary

Summary

Summary