OVERVIEW

This course aims to introduce the basic computational paradigms of data science and technology, with specific focus on the three pillars of Artificial Intelligence for the data world, i.e. numerical simulation, inverse problems theory and machine learning. Then the course will describe some applications in biochemistry, involving pattern recognition methods for image processing in Scanning Tunnelling Microscopy, the mathematical modelling of tracer kinetics in nuclear medicine and the use of Molecular Interaction Maps in oncology.

AIMS AND CONTENT

LEARNING OUTCOMES

This course aims to introduce the basic computational paradigms of data science and technology, with specific focus on the three pillars of Artificial Intelligence for the data world, i.e. numerical simulation, inverse problems theory and machine learning. Then the course will describe some applications in biochemistry, involving pattern recognition methods for image processing in Scanning Tunnelling Microscopy, the mathematical modelling of tracer kinetics in nuclear medicine and the use of Molecular Interaction Maps in oncology.

AIMS AND LEARNING OUTCOMES

The general objective of the course is to provide students with a first overview of the main issues related to modern data science and its cultural background. The course has also two more specific objectives. The first one is to illustrate some computational tools representing the methodological basis for any artificial intelligence approach to data analysis problems. The second one is to describe three applications concerned with the use of data science methods in chemistry and biochemistry: the problem of the automatic recognition and classification of atomic species in Scanning Tunnelling Microscopy; the modelling of glucose metabolism by means of nuclear medicine data; the simulation of the chemical reaction network at the basis of a specific cellular transition in oncogenesis.

PREREQUISITES

Students attending the course should know in advance the basics of

• Linear Algebra (vectors, matrices and their norms; linear systems; inversion of a matrix; eigenvalues)

TEACHING METHODS

lectures and computational laboratory activity

SYLLABUS/CONTENT

The course is characterized by the following three parts:

Computational tools: harmonizing competences (8 hours)

• Basics of numerical analysis (2 hours)
• Basics of Bayesian theory (3 hours)
• Basics of regularization theory (3 hours)

Artificial Intelligence: the many aspects of data modeling (10 hrs)

• Numerical Simulation (2 hours)
• Inverse Problems (4 hours)
• Machine Learning (4 hours)

Applications to chemical and biochemical data (6 hrs)

• STM imaging (2 hrs)
• Tracer kinetics (2 hrs)
• Chemical Reaction Networks (2 hrs)

RECOMMENDED READING/BIBLIOGRAPHY

no bibliography

TEACHERS AND EXAM BOARD

Exam Board

MICHELE PIANA (President)

SARA SOMMARIVA

LESSONS

LESSONS START

not known yet

Class schedule

The timetable for this course is available here: Portale EasyAcademy

EXAMS

EXAM DESCRIPTION

oral

ASSESSMENT METHODS

questions about the course syllabus