BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260721T231234Z UID:13c34af6-25cf-447e-a0f3-fe5866b33301 DTSTART:20260921T090000Z DTEND:20260921T170000Z DESCRIPTION:## Overview\n\nIn bioinformatics and computational biology\, re searchers routinely work with high-dimensional datasets where the number o f features (genes\, proteins\, variants\, metabolites) far exceeds the num ber of samples. This "large p\, small n" scenario presents unique challeng es for building useful models and extracting biological insights from the data. Feature selection methods help identify the most informative variabl es\, helping build models with better generalisability\, better interpreta bility\, and revealing biomarkers.\n\nThis 1-day course provides a hands-o n introduction to feature selection methods specifically tailored for bioi nformatics applications. Participants will learn both classical and state- of-the-art approaches with a focus on their practical application to real biological datasets.\n\n\nThe course will also cover when and why to use d ifferent approaches\, how to validate selected features\, and common pitfa lls in high-dimensional biological data analysis.\n \n## Audience\n\nThis course is designed for PhD students\, postdoctoral and other researchers in the life sciences from both academia and industry who work with high-di mensional biological data and want to improve their predictive modelling a nd biomarker discovery workflows.\n\n## Learning Objectives\n \nAt the en d of the course\, participants will be able to:\n\n* **Explain** the funda mental challenges of feature selection in high-dimensional biological data (curse of dimensionality\, spurious correlations\, multiple testing)\n* * *Distinguish** between filter\, wrapper\, and embedded feature selection m ethods and select appropriate approaches for different biological question s\n* **Implement** and apply basic and advanced feature selection methods using scikit-learn and complementary Python packages (VarianceThreshold\, L1 penalty\, Recursive Feature Elimination\, Boruta\, knockoffs framework) \n* **Recognise** and avoid common pitfalls in feature selection pipelines \, such as data leakage and multicollinearity\n* **Implement** stability s election to assess feature selection robustness.\n\n \n## Prerequisites\n  \n**Knowledge/competencies:**\n \n\nThe course is intended for people a lready familiar with Python programming (including NumPy and Pandas) and f amiliar with different omics data. Participants should be comfortable with fundamental machine learning concepts: supervised learning (classificatio n/regression)\, train/test splits\, cross-validation\, overfitting\, and r egularization (L1/L2 penalties). Basic statistical knowledge (hypothesis t esting\, p-values\, multiple testing correction\, FDR) is expected. \n\nT his course is part of the Machine Learning learning path. To get the most out of this course\, you should meet the learning outcomes of the [First S teps with Python in Life Sciences](https://www.sib.swiss/training/course/F SWPY) and [Introduction to Machine Learning with Python](https://www.sib.s wiss/training/course/INMLP) courses.\n \nUpon completion of this course\, you may wish to attend [Ensuring More Accurate\, Generalisable\, and Inte rpretable Machine Learning Models for Bioinformatics](https://www.sib.swis s/training/course/INTML)\, [Diving into Deep Learning - Theory and Applica tions with PyTorch](https://www.sib.swiss/training/course/DEEPP) and [Fede rated Learning in Bioinformatics courses](https://www.sib.swiss/training/c ourse/FEDBX).\n \n\n**Technical:**\n\n \nYour laptop must have a recent Python version (minimum 3.10) and several Python libraries installed. The needed libraries will be indicated in the course GitHub repo in due time.\ n\n## Schedule – CE(S)T time zone\n\n \n| Time | Module |\n|------|---- ----|\n| 09:00-09:15 | Welcome &\; Foundations: Why feature selection m atters in bioinformatics |\n| 09:15-09:45 | Filters: variance threshold\, correlation\, mutual information |\n| 09:45-10:30 | Wrapper methods: Recur sive Feature Elimination (RFE) and Sequential Feature Selection (SFS) |\n| 10:45-12:00 | Embedded methods: L1 regularisation\, tree-based importance |\n| 13:00-13:45 | Boruta : all-relevant feature selection |\n| 13:45-14: 30 | Knock-offs: tangling with the feature dependency problem |\n| 14:45-1 5:45 | Practical considerations: multicollinearity\, data leakage\, stabil ity selection\, SHAP |\n| 16:00-17:00 | Small project  |\n\n## Applicatio n\n\nThe registration fees for academics are **100 CHF** and **500 CHF** f or for-profit companies.\n\nWhile participants are registered on a first c ome\, first served basis\, exceptions may be made to ensure diversity and equity\, which may increase the time before your registration is confirmed .\n\nApplications will close on **01/09/2026** or as soon as the places wi ll be filled up. Cancellation after **07/09/2026** will not be reimbursed. Please note that participation in SIB courses is subject to our [general conditions](https://www.sib.swiss/legal-documents).\n\nYou will be informe d by email of your registration confirmation. Upon reception of the confir mation email\, participants will be asked to confirm attendance by paying the fees within **5 working days**.\n\n## Venue and Time\n\nThis course wi ll be streamed.\n\nThe course will start at 9:00 CET and end around 17:00 CET.\n\nPrecise information will be provided to the registered participant s in due time.\n\n## Additional information\n\nCoordination: Patricia Pala gi\, SIB Training group.\n\nA **Certificate of Attendance** will be sent p rovided you were present at the course\, whereas a **Certificate of Achiev ement** recommending 0.25 ECTS will be sent provided you passed the exam.\ n\nYou are welcome to register to the SIB courses mailing list to be infor med of all future courses and workshops\, as well as all important deadlin es using the form [here](https://lists.sib.swiss/mailman/listinfo/courses) .\n\nSIB abides by the [ELIXIR Code of Conduct](https://elixir-europe.org/ events/code-of-conduct). Participants of SIB courses are also required to abide by the same code.\n\nFor more information\, please contact [training @sib.swiss](mailto://training@sib.swiss). SUMMARY:Feature Selection for Bioinformatics with Python URL;VALUE=URI:https://www.sib.swiss/training/course/20260921_FEATS END:VEVENT END:VCALENDAR