PARAFAC analysis of fluorescence datasets
PARAllel FACtor analysis (PARAFAC) is used to interpret natural organic matter (NOM) fluorescence when measured using excitation emission matrix (EEM) spectroscopy. PARAFAC resolves overlapping fluorescence signals into independent spectra representing different chemical components. When fluorescence signals conform to Beers Law, this process can lead to the identification and quantification of independently varying fluorophores. However, many practical and analytical hurdles stand between EEM datasets and their chemical interpretation. This course will provide instruction on best-practice application of PARAFAC to fluorescence datasets using the drEEM toolbox for MATLAB.
This course is offered through the PhD program in Water Environment Technology at the Department of Architecture and Civil Engineering, Chalmers University of Technology. Participants who complete the course including submitting a final report can obtain official credit (2.5 ETS points, Pass/Fail grade).
Topics to be covered
(1) PARAFAC model and assumptions; (2) pre-processing for
non-trilinearity (scatter) and concentration variability; (3) identifying and
treating outliers; (4) model development and visualization; (5) model
validation: split-half analysis, core consistency, residual analysis; (6) sensitivity
analyses of larger datasets; (7) comparing models using the OpenFlour database.
An introductory MATLAB course.
Basic theory of fluorescence spectroscopy and some familiarity with fluorescence literature.
Obtaining course credit
To obtain ETS points for your participation, you will
(1) Read key literature and participate in the tutorial sessions.
(2) Submit a report documenting the results of applaying PARAFAC to a real dataset. The report deadline is approx. 4 weeks after the course ends.
The previous course was held on 5th-7th
June, 2018. The current target is to hold the next course in 2020.
The course is offered within a N5T collaboration with the Technical University of Denmark (DTU).