Course code:
371H2
Course name:
Chemometrics

Academic year:

2024/2025.

Attendance requirements:

There are no requirements.

ECTS:

10

Study level:

doctoral academic studies

Study programs:

Chemistry: 1. year, winter semester, elective (E71H2) course

Chemistry: 1. year, winter semester, elective (E72H2) course

Chemistry: 2. year, winter semester, elective (E73H2) course

Chemistry: 2. year, winter semester, elective (E74H2) course

Teacher:

Filip Lj. Andriæ, Ph.D.
associate professor, Faculty of Chemistry, Studentski trg 12-16, Beograd

Assistants:

Hours of instruction:

Weekly: five hours of lectures + three hours of study research work

Goals:

The course aims to familiarize students with the basics of advanced data processing while meeting the contemporary standards of the courses offered at the University of Leuven (Belgium), Umea University (Sweden) and the University of Bergen (Norway). Taking into account the vast amount of data generated by various instrumental techniques (spectroscopy and spectrophotometry, chromatography, mass spectrometry, nuclear magnetic resonance, electrochemical experiments, miniaturization and hyphenation of multiple instrumental techniques), as well as the need to optimize experimental conditions in various aspects of fundamental and applied chemistry, environmental chemistry and biochemistry, the aim of the course is to introduce the students through practical examples, in a simple and understandable way to application of modern techniques for data processing and related software in the analysis of food and natural products, environmental chemistry, medicinal chemistry, archaeometry, chemistry of artworks, phytochemistry, biochemistry, biotechnology, quality control etc.

Outcome:

Upon completion of the course, the student should be able to: recognize and understand the importance of concepts of pattern recognition, explorative data analysis, basics of modelling and classification, fundamentals of experimental design and optimization; to properly use and select methods for data processing and adequately interpret the obtained results, to use standard computer programs for data processing; to use scientific and professional literature in the field of chemometrics in accordance with the specific needs of doctoral studies, related job or research interests.

Teaching methods:

Lectures.

Extracurricular activities:

Coursebooks:

Main coursebooks:

  • Richard G. Brereton, Chemometrics: Data Analysis for the Laboratory and Chemical Plant, Wiley, 2003.
  • K. Varmuza, P. Filmoser: Introduction to Multivariate Statistical Analysis in Chemometrics, CRC Press, Taylor and Francis Gorup, 2009, Boca Raton

Supplementary coursebooks:

  • Material and handouts provided during the lectures.

Additional material:

  Course activities and grading method

Lectures:

10 points (5 hours a week)

Syllabus:

1. BASIC ELEMENTS OF STATISTICS
Statistics of repeated measurement statistics - measures of central tendency and dispersion, measurement errors. Basic parametric and non-parametric significance hypothesis testing. Univariate linear and curvilinear regression.

2. EXPERIMENTAL DESIGN AND OPTIMIZATION
Basics of experimental design, factorial, central, Box-Benken design, and mixture design. Simplex optimization. Response surface method in optimization of experimental conditions. Multicriteria optimization.

3. EXPLORATIVE DATA ANALYSIS AND PATTERN RECOGNITION
Principal Component Analysis (PCA), hierarchical cluster analysis (HCA), k-means cluster analysis, concepts of internal and external cluster validation.

4. MULTIVARIATE ANALYSIS OF VARIANCE (MANOVA)

5. LINEAR REGRESSION AND CALIBRATION
Multiple linear regression (MLR), principal component regression (PCR), partial least squares regression (PLS), cross-validation techniques, variable selection and optimal model complexity, statistical parameters of model quality.

6. LINEAR CLASSIFICATION TECHNIQUES
Linear discriminant analysis (LDA), partial least squares-discriminant analysis (PLS-DA), SIMCA, statistical performance of classification models.

7. NONLINEAR AND ALGORITHMS INSPIRED BY NATURE
Artificial neural networks (ANN), regression and classification trees (RT), random forests, k-nearest neighbors (kNN).

All data processing methods are demonstrated on practical examples related to the chromatographic behavior of biologically active and environmentally important compounds, food analysis and quality control of food products, analysis of pollutants in the environment and food, quantitative structure activity and property relationships (QSAR and QSPR), mineral composition and spectroscopic data of archaeological specimens, soil samples, optimization of enzymatic activities polymerization, and degradation conditions etc.

Semester papers:

30 points

Written exam:

60 points

Study research work:

0 points (3 hours a week)