All Events Seminars & Lectures

Introduction to multivariate data analysis

Event Details:

  • Date: - -
  • Venue: Collegio A. Volta, Pavia
  • Register

On December 12, 2019 Debora Paris, Consiglio Nazionale delle Ricerche (CNR) Istituto di chimica biomolecolare (ICB), will give a presentation entitled Introduction to multivariate data analysis at 14:00 in the College Lecture theatre at Collegio A. Volta All students are invited to attend, especially those reading Medicine, Biology, Biotechnology and Pharmaceutical Sciences.

The poster of the lecture can be downloaded here.

The advent of new efficient technologies which allow to rapidly collecting a wide range of measurable features and observations in any biochemical or biophysical field of research has inexorably led to the acquisition of large volume of data. Such collection consist of different sources of variance, including both information of interest and redundant or unnecessary variance,  not strictly connected to the biological variation. Actually, such abundance of data does not entirely imply information.  In this context, the goal of Multivariate Statistical Analysis is to find the best way to reduce dimensionality, so to extract the majority of reliable and statistically significant information from excessive and irrelevant data, and possibly to correlate it as much as possible to the biological issues under investigation. With this purpose, a simple introduction of multivariate analysis methods will be briefly presented: in particular,   the most popular regression algorithms principal component analysis (PCA) [1] and partial least squares projection to latent structures (PLS) [2], together with the “filtered versions” orthogonal projection to latent structures (OPLS, O2PLS) [3].

[1] Eriksson, L., E. Johansson, N. Kettaneh-Wold, J. Trygg, C. Wikstr, S. Wold. 2006. Multi- and megavariate data analysis. Basic principles and applications. Umeå: MKS Umetrics AB.
[2] Wold S, Sjostrom M, Eriksson L: PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems 2001, 58:109-130.
[3] Trygg J, Wold S: Orthogonal projections to latent structures (O-PLS). Journal of Chemometrics 2002, 16:119-128.

Currently a researcher at the CNR-ICB Institute of Biomolecular Chemistry of Pozzuoli, Napoli (Italy), Debora Paris has been working on NMR spectroscopy applied to metabolomics since 2004. She received the Graduation in Physics cum laude from the University of Napoli “Federico II” in the 2006 starting her experiences with the metabolomic study of the apoptotic process induced by ionizing radiation in MG-63 human spheroids by Nuclear Magnetic Resonance spectroscopy. She received the Ph.D in Chemistry from the University of Napoli “Federico II” in 2009, defending the PhD thesis “Development and Application of Advanced NMR Techniques for Biomarkers Identification”.

Main scientific interests:

-Research and development of specific NMR techniques for metabolomic investigations and two-dimensional experiments for fast acquisition (fast-NMR);
-Analysis and assignment of the metabolic profile of the spectral signals of various biological samples: microalgae, cells in vivo, cell extracts, tissue extracts, biofluids (blood, urine, saliva, respiratory condensate);
-Application of multivariate statistical data analysis, data projection and filtering methods and pattern recognition techniques for modeling, spectral classification and integration between data from different platforms.

The First DNA Fingerprint (Wellcome Collection)

Log In

Create an account