Dear all,

We would like to share an upcoming online course focused on Machine Learning for Multi-Omics Integration, taking place on 21–23 September.

The rapid growth of genomics, transcriptomics, proteomics, metabolomics, and single-cell datasets is creating new opportunities and challenges for biological data analysis. Machine learning methods are increasingly being used to integrate these diverse data types, identify complex patterns, and generate biologically meaningful insights.

This course is aimed at researchers working with omics data and combines lectures with hands-on practical sessions covering both established and emerging approaches for multi-omics integration.

Topics covered include:

• Machine learning concepts for biological data integration
• Feature selection and supervised integration approaches using LASSO, PLS, LDA, mixOmics, and DIABLO
• Unsupervised integration with MOFA1 and MOFA2
• Deep learning approaches, including autoencoders
• Single-cell multi-omics integration methods, including UMAP, Seurat CCA, Dynamic Time Warping (DTW), Weighted Nearest Neighbours (WNN), and batch correction approaches

Participants will gain practical experience with commonly used computational tools and workflows for multi-omics analysis.

More information about the programme and registration is available here: https://www.physalia-courses.org/courses-workshops/multiomics/

Best wishes,

Carlo Pecoraro, Ph.D.

Director, Physalia-courses