Ensuring More Accurate, Generalisable, and Interpretable Machine Learning Models for Bioinformatics

Overview

Machine Learning has become an essential tool in Life Science, letting scientists explore and learn from large and complex biological datasets. To collectively unravel the puzzle of life, we must ensure that machine learning models make the most of the available data and that they are correctly generalizable, robust, and interpretable to provide trustworthy and actionable insights. This advanced course is designed for scientists who already have a foundational understanding of machine learning and seek to enhance their core skills in this domain.

This course focuses on best practices and advanced techniques in Machine Learning, aiming to provide you with the tools needed to develop more accurate, generalizable, and transparent models.

Audience

This course is addressed to life scientists, bioinformaticians, and computational biologists who would like to learn more about general best practices in Machine Learning and get more out of their Machine Learning models: more precise hyper-parameters, more generalizable models, and more interpretable models.

Learning outcomes

At the end of this course, you will be able to:

• Use the hyperopt library to efficiently explore your hyper-parameter space with Bayesian Optimization and tune your models.
• Evaluate the generalizability of your generated models using best practices such as nested cross-validation.
• Explain the role of each feature in your model's prediction, even for so-called "black-box" models
• Examine the results of your models and assess their quality.

Prerequisites

Knowledge / competencies

• Good knowledge of the basics of machine learning, such as K-fold cross-validation, Decision Tree, and evaluation metrics.

• Fluency with the Python programming language, including working knowledge of standard data analysis libraries such as numpy, pandas, matplotlib, and scikit-learn.

• Familiarity with different omics data technologies (highly recommended).

The competencies and knowledge levels required correspond to those taught in courses such as Introduction to Machine Learning in Life Sciences.

Before applying to this course, please self-assess your Python and Machine Learning skills using the quiz here. We recommend a score of at least 6/10.

Technical

Your laptop must have a recent Python version (minimum 3.0) and several Python libraries installed. The needed libraries will be indicated in the course GitHub repo and here in due time.

Schedule - CET time zone

Please take note of the following information:

This schedule is subject to change to allow time for questions and discussions with the course participants.

The group work involves a project designed by the course trainers, where participants collaborate in smaller groups to address the same question.

In addition to the suggested projects, the group work may also incorporate some Bring Your Own Data components. However, this depends on the data's cleanliness, the feasibility of your objective, and the interest of other participants. In any case, the instructors will be happy to discuss your data.

Application

The registration fees for academics are 100 CHF and 500 CHF for for-profit companies.

You will be informed by email of your registration confirmation. Upon reception of the confirmation email, participants will be asked to confirm attendance by paying the fees within 5 days.

While participants are registered on a first come, first served basis, exceptions may be made to ensure diversity and equity, which may increase the time before your registration is confirmed.

Applications close on 29.09.2024 or as soon as the maximum capacity has been reached. The deadline for free-of-charge cancellation is set to 01.10.2024. Cancellation after this date will not be reimbursed. Please note that participation in SIB courses is subject to our general conditions.

Venue and Time

This course will be streamed.

The course will start at 9:00 and end around 17:00. Precise information will be provided to the participants in due time.

Precise information will be provided to the participants in due time.

Additional information

Coordination: Patricia Palagi

We will recommend 0.25 ECTS credits for this course (given a passed exam at the end of the course).

You are welcome to register to the SIB courses mailing list to be informed of all future courses and workshops, as well as all important deadlines using the form here.

Please note that participation in SIB courses is subject to our general conditions.

SIB abides by the ELIXIR Code of Conduct. Participants of SIB courses are also required to abide by the same code.

For more information, please contact training@sib.swiss.