Objectives

• This course aims at teaching students how to perform such an analysis, emphasizing both, theoretical and practical

Outcomes

Students will have the opportunity to work both independently during tutorials and in teams during the research project. Students will learn the theoretical basics of ML and aquire extensive hands-on experience on:-Python scripting language and packages e.g. pytorch, scikit learn.-Data-specific cleaning and preprocessing pipelines for the vast amounts of biomedical data needed for ML analysis-Learn to apply unsupervised and supervised learning methods to analyse biomedical data and build predictive models.- Adequate evaluation of statistical learning models depending on data size, quality, and distribution.- Students will be developing their own statistical learning pipeline to analyse and evaluate biomedical data sets and determine the most determining factors. Finally the students will be interpreting the obtained results, and scientifically discussing the results.

Course Contents

Statistical learning is gaining more and more importance in the field of bioinformatics and biomedicine. In particular, analyzing biomedical molecular data from Omics (genomic, transcriptomic, epigenomic, proteomic, etc) opens novel opportunities for early diagnosis and prognosis of complex diseases and fostering the development of novel treatment approaches. However, many challenges arise from this kind of analysis. For a robust machine learning (ML) analysis, we need to carefully integrate and preprocess the data and choose the appropriate models for the statistical learning. Finally, the quality and performance of such models need careful evaluation and the identification of important factors and potential biomarkers.

Description

Data will be available soon!