Login for PhD students/staff at UCPH      Login for others
International School of Chemometrics 2025 - Metabolomics - 1.5 ECTS
Provider: Faculty of Science

Activity no.: 5328-25-04-37There are 16 available seats 
Enrollment deadline: 05/04/2025
Date and time21.05.2025, at: 09:00 - 22.05.2025, at: 16:00
Regular seats40
ECTS credits1.50
Contact personHenriette Hansen    E-mail address: henha@food.ku.dk
Enrolment Handling/Course OrganiserRasmus Bro    E-mail address: rb@food.ku.dk
Written languageEnglish
Teaching languageEnglish
Semester/BlockSpring
Scheme groupNot included in the scheme group
Exam formCourse participation
Course workload
Course workload categoryHours
Lectures60.00
Preparation / Self-Study20.00
Class Instruction60.00
Theoretical exercises79.00
E-learning56.00
Sum275.00
Aim and content
Four week school designed to introduce different key aspects of Data Science and Machine Learning in different branches of science (chemistry, food & feed, physics, environmental, political economics, etc). The course has the following modules:

1) PROGRAMMING - Introduction to Programming for Multivariate data analysis in Matlab, Python and R
This online seminar is based on online pre-recorded videos that are thought to be an introduction to the main aspects of dealing with Matlab, R and Python focused on Multivariate Data Analysis.

2) BASIC - Basic Introduction to Chemometrics and Linear Algebra
This seminar includes two parts:
EXPLORE (PCA) and LINAL (Linear Algebra).

3) INTERMEDIATE:
This seminar includes three parts:
• DoE, Design of Experiments
• VARSEL, variable selection methods;
• MCR, Multivariate Curve Resolution.

4) DL - Non-Linear Modeling / Deep Learning
This seminar includes two parts: methods for non-linear modeling and the different architectures of Artificial Neural Networks (basic structures, shallow neural networks and deep neural networks).

5) Classification
The course will deal with the main linear classification methods like Discriminant Analysis, Partial Least Squares Discriminant Analysis (PLS-DA) and SIMCA. We will see both theoretical aspects and practical applications. It will also deal with Support Vectors Machine and Random Forests.

6) Optimization
This seminar will give an overview of different optimization methods that are extremely useful for optimizing hyperparameters in models. Methods like Particle swarm optimization (PSO) or Gauss-Newton will be taught and different examples discussed.

7) Metabolomics
This seminar will deal with the chemometric approaches for integrating (“fusing”) data from different sources. First of all, the various configurations which may occur when dealing with multiple data matrices will be presented and discussed, and a hierarchy/systematization of the possible data fusion approaches will be introduced. The main multi-block strategies for data exploration and predictive modeling will then be discussed and compared. Further classification of models depending on whether the globally common, locally common and distinct information is considered or not will also be introduced. The theoretical and algorithmic description of the methods will be accompanied by worked examples of real data sets.

8) GLUE (How not to make Chemometrics) and WORKSHOP
We will take a very close look at all the most common mistakes that even experienced people will do when doing multivariate analysis. We will cover exploration, calibration, interpretation, visualization and many other subjects. This is done with a focus on the most common problems as well as sound alternatives to address them.

Formal requirements
None required. We start from the basics and go all the way to advanced.

Learning outcome
Knowledge
Upon completing the course, students will:
- Understand the foundational principles of data science methods, specifically in chemometrics and multivariate analysis.
- Gain theoretical knowledge in statistical and machine learning techniques such as PCA, multivariate regression, variable selection methods, and non-linear modeling.
- Comprehend advanced data analysis methods including multiway data analysis, ANOVA Simultaneous Component Analysis (ASCA), and data fusion strategies.
- Learn the main experimental designs used in Design of Experiments (DoE), their applications, and their limitations.
- Understand the implications of improper analysis in chemometrics and methods to avoid common mistakes in multivariate data analysis.

Skills
Students will develop the ability to:
- Apply data analysis techniques (PCA, MCR, etc.) to real-world data sets in their own research fields.
- Code basic algorithms in Matlab, Python, or R for multivariate data analysis and create analytical data pipelines.
- Perform experimental design using DoE principles to optimize processes and interpret data effectively.
- Integrate data from multiple sources (Data Fusion) and interpret the resulting fused datasets to address complex research questions.
- Conduct a critical analysis of data, identifying and troubleshooting potential errors in data interpretation and visualization.

Competences
By the end of the course, students will be able to:
- Analyze and interpret diverse data types independently, drawing meaningful conclusions from complex datasets.
- Solve domain-specific data problems in a structured and reproducible manner.
- Collaborate effectively with researchers from diverse scientific backgrounds, communicating data science concepts clearly.
- Assess and select appropriate data analysis methods based on the research context and the nature of the data.

Target group
All PhD students who aim to use data science within chemical and related areas.

Teaching and learning methods
The seminars of the School of Chemometrics will consist of a mix of presentations from world leading researchers mixed with practical exercises in data analytic software that provides the student with practical experience on how to apply the tools learned in the course. The exercises are done under the supervision of the teachers.

The initial week on programming offers teaching in three different languages and all the teaching in this part is based on e-learning. The student can choose between either programming in MATLAB, Python or R in this first week.

The rest of the school (three weeks) is physical on-site training.

Lecturers
• Assoc. Prof. Davide Ballabio, University of Milano-Bicocca will teach several days on classification and general data science
• Assoc. Prof. Agnieszka Smolinska, Maastricht University will teach design of experiment
• Prof. José Amigo Rubio, University of Basque Country will teach courses on programming, basic chemometrics, MCR and general data science. They are also the main responsible for the day-to-day activities throughout the course.
• There are several other guest lectures but these are the ones for which we have applied for funding

Remarks
Course fee: 6000 DKK for the full course. Fee covers additional invited teachers for the course, social events, facilities for poster.

PhD students enrolled at a Danish PhD school that is a member of the open market for PhD courses: free of charge.

Master's students from Danish universities: free of charge.

Search
Click the search button to search Courses.


Course calendar
See which courses you can attend and when
JanFebMarApr
MayJunJulAug
SepOctNovDec



Publication of new courses
All planned PhD courses at the PhD School are visible in the course catalogue. Courses are published regularly.