Others

Il corso avrà l'obiettivo di fornire dettagli su come istanziare in maniera automatizzata clusters per l'orchestrazione di container che implementano a loro volta un batch system a servizio. Verrà illustrato principalmente il caso di un sistema di batch basato su HTCondor, dando pero' esempi anche di piattaforme Spark per il training di Machine Learning.

Il corso fornirà gli strumenti di alto livello per l'utilizzo di infastrutture cloud finalizzato al processamento di dataset. In tal senso verranno mostrati strumenti per la gestione delle applicazioni (container), per la gestione del software (cvmfs) e dei dati (e.g. xrootd). Sarà inoltre mostrato concretamente l'esempio del workflow di data anlaisi dell’esperimento AMS, come pilota nell'utilizzo dei batch system a servizio.

Il corso si terrà presso la  Sezione INFN di Perugia, con inizio alle ore 14.00 del 25 Novembre e termine alle ore 13.00 circa del 28 Novembre 2019.

In recent years, the vision of Open Science has emerged as a new paradigm for transparent, data-driven science capable of accelerating competitiveness and innovation.  EGI Notebooks is a services of the EGI e-infrastructure, providing a user-friendly and highly flexible Jupyter-based web environment for the development and execution of data analysis and visualisation ‘notebooks’. Notebooks can contain programming codes in various languages, HTML scripts, dynamic visualization, equations as well as images and explanatory text to provide guidance and context for the analysis. Through notebooks users can easily share concepts, ideas and working applications, capturing the full analytical methodology, connections to data and descriptive text to interpret those data. Binder can turn Jupyter notebooks reproducible and reusable by anyone, anywhere. Zenodo is an open access repository for research publications, scientific data and other 'research objects'. Jupyter, Binder and Zenodo are pillars of Open Science.

Advances in biological and medical technologies drive continuous generation of large amounts of biomedical Big Data. European Nucleotide Archive stores 260 million sequences comprising 339 trillion nucleotides. This will double in less than 3 years if the current rate of growth is sustained! Given the exponential progress in sequencing technology the increase will only get steeper, entailing an intensified demand for experts in NGS data analysis. Big Data requires applying new solution to leverage its potential. Machine Learning (ML) is the answer to the increased complexity of research problems in science, industry and in everyday life. It is our conviction that knowledge of the ML techniques is a crucial skill every data scientist should acquire throughout their training.

For above reasons, #NGSchool2019: Machine Learning for Biomedicine will be focused on Machine Learning (ML) and its application in Bioinformatics & NGS Data Analysis as well as personalised medicine. We will cover the following subjects:

• Introduction to Linux, programming (R and Python) and statistics

• Tools for efficient and reproducible research

• Modern  and  libraries/packages for biomedical data science

• Deep learning in long read sequencing data analysis

• Statistical and probabilistic analysis of biomedical data

• Integration of genomics data using ML for understanding gene regulation in its three dimensional context

• Quality control and typical mistakes of a beginner ML user