Training courses 2017: Training course on Machine Learning for Biologists

IMPORTANT DATES for this Training course:

Deadline for applications: 21 July 2017
Course date: 4-7 September 2017
Chosen participants will be notified by 25 July 2017. A maximum of 24 candidates will be accepted in the course.

Course Description

The aim of the course is to provide a practical introduction to the analysis of “omics” data. Topics will range from data visualization/exploration to univariate/multivariate analysis and machine learning. Practical examples and applications will be illustrated by using R and Python.

Course Milestones:

Data exploration and visualization
Univariate/Multivariate analysis
Introduction to machine learning: classifiers, performance measures, diagnostics
Machine learning tools for the analysis of Gene Expression data
The Data Analysis Plan (DAP) - intro to unbiased pipelines for (binary) classification
Performance measures and diagnostic plots - Accuracy, MCC, Stability: theory and graphics
Differential network analysis – co-expression networks, graph comparison, community detection: theory and examples in R/Python, visualization by the igraph library and use of the ReNette web interface
Basic application of ML to gene prediction

Target audience

A maximum of 24 students will be accepted. Computational biologists, bioinformaticians, biological data analysts.

Learning objectives/outcomes

Acquisition of a working knowledge about running a full classification/profiling pipeline on omics data (e.g., gene expression). After this course participants should be able to:

Develop and implement visualization/exploration tools
Understand advantages and limitations of univariate and multivariate data analysis strategies in the context of omics research
Understand principles and applications of ML
Implement reproducible workflows in data analysis

Course prerequisites

basics of R, Python, *nix (Unix or Linux) shell, basic statistical knowledge, basic knowledge of linear algebra

FEE: The fee will cover attendance to lectures and practicals, coffee breaks, lunches, the welcome aperitivo on Monday September 4th. The participants are expected to pay their own travel and accommodation costs.

Candidates from a bioinformatics lab from one of the ELIXIR-IIB member institutions (see the list at the bottom): 120 euros.
All the other candidates: 160 euros

Application Form

VENUE: Fondazione Edmund Mach, Palazzo della Ricerca e della Conoscenza, Via E. Mach 1, San Michele all’Adige (TN), Italy, (zip code 38010).
General Info: http://www.fmach.it/eng/CRI/general-info
The campus: http://www.fmach.it/eng/About-us/Campus

To get in:

By train: Trenitalia (national railways) stops at Trento or Mezzocorona and then local train to San Michele all’Adige
By car: The venue is 5’ drive from the highway A22 (autostrada del Brennero), exit of “San Michele all’Adige/Mezzocorona”.
By fly: From Verona airport please use the shuttle bus to Verona train station and then the train towards Trento.

SUGGESTED ACCOMMODATIONS: A limited number of rooms will be available at the venue (costs 20 EUR/night). Please contact the local organizer (alessandro.cestaro@fmach.it). Here is some information about hotels close to San Michele all’Adige.

Instructors and helpers

Giuseppe Jurman - Fondazione Bruno Kessler, Trento, IT
Marco Chierici - Fondazione Bruno Kessler, Trento, IT
Davide Albanese - Fondazione E. Mach, Trento, IT
Pietro Franceschi - Fondazione E. Mach, Trento, IT
Marco Moretto - Fondazione E. Mach, Trento, IT
Paolo Sonego - Fondazione E. Mach, Trento, IT
Samantha Riccadonna - Fondazione E. Mach, Trento, IT
Andrea Cattani - Fondazione E. Mach, Trento, IT

Organisers

Alessandro Cestaro (Local Organizer, Fondazione E. Mach, Trento, IT )
Vincenza Colonna (ELIXIR-IIB Training Coordinator Deputy, CNR, IT)

Programme

Course materials are available from the github repository:
https://github.com/ELIXIR-IIB-training/MLB2017

Monday 04 September 2017 - Introduction
14:00-15:30	Course opening	Participants’ self-presentations
15:30-16:15	Plenary lecture	P. Franceschi, S. Riccadonna	Introduction to "omics" data. Principles of data exploration and analysis
16:15-16:45	Coffee break
16.45-19:00	Practical	P. Franceschi, S. Riccadonna	Introduction to "omics" data. Principles of data exploration and analysis
19:30-21:00	Welcome aperitivo at Cantina storica Istituto Agrario San Michele
Tuesday 05 September 2017 - Machine Learning
09:30-09:40	Previously On	Recap of previous lessons by participants
09:40-10:30	Lecture	D. Albanese, P. Franceschi, S. Riccadonna	Univariate and Multivariate analysis
10:30-11:00	Coffee break
11:00-13:00	Practical	D. Albanese, P. Franceschi, S. Riccadonna	Univariate and Multivariate analysis. Practical session with R
13:00-14:30	Lunch
14:30-16:15	Lecture	D. Albanese, P. Franceschi, S. Ricadonna	Machine Learning: introduction and applications to biological data. Classification basics, model selection and prediction
16:15-16:45	Coffee Break
16:45-18:30	Practical	D. Albanese, P. Franceschi, S. Riccadonna	Performance measures and diagnostic plots
Wednesday 06 September 2017
09:30-09:40	Previously On	Recap of previous lessons by participants
09:40-10:30	Lecture	P. Sonego, S. Riccadonna	Analyzing Gene Expression Data
10:30-11:00	Coffee break
11:00-13:00	Practical	P. Sonego, S. Riccadonna	Analyzing Gene Expression Data
13:00-14:30	Lunch
14:30-16:15	Lecture	M. Chierici, G. Jurman	The Data Analysis Plan (DAP) - intro to unbiased pipelines for (binary) classification
16:15-16:45	Coffee Break
16:45-18:30	Practical	M. Chierici, G. Jurman	Implementation of a basic DAP in Python (Scikit-Learn) with feature ranking and classification
Thursday 07 September 2017
09:15-09:25	Previously On	Recap of previous lessons by participants
09:25-10:00	Lecture	M. Moretto, A. Cestaro	Gene prediction methods as an example of ML on genomic data
10:00-10:30	Coffee Break
10:30-12:30	Practical	M. Moretto, A. Cestaro	Training a gene prediciton method
12:30-13:00	Wrap-up and feedback