Home About us Contact
|
Home About us Contact | ||
|
|
||
PLS Regression (pl + regression)
Selected AbstractsSpatio-temporal patterns of fish assemblages in a large regulated alluvial riverFRESHWATER BIOLOGY, Issue 7 2009RENAUD RIFFLART Summary 1. The River Durance, the last alpine tributary of the River Rhône, is a large, braided alluvial hydrosystem. Following large-scale regulation, flow downstream of the Serre-Ponçon dam has been maintained at 1/40th of previous annual mean discharge. To assess the effects of historical disturbances, fish assemblages and habitat use were analysed during five summers in a representative reach of the middle Durance. 2. Habitat availability and use were assessed with a multi-scale approach including the variables water depth, current velocity, roughness height of substratum, amount of woody debris and lateral/longitudinal location. Eighteen fish species were sampled by electrofishing in 289 habitat sample units. 3. Partial least square (PLS) regression showed that taxa were mainly distributed according to relationships between their total length and water depth/velocity variables. Fish assemblage composition was also related to roughness height as well as distance from the bank or to the nearest large woody debris. However, PLS regression revealed no significant differences in habitat selection between two periods of varying hydromorphological stability. 4. Fish distribution patterns and density were related to proximity to the bank and cover, indicating that local scale variables need to be considered in conservation and restoration programmes. [source] The PLS multivariate regression model: testing the significance of successive PLS componentsJOURNAL OF CHEMOMETRICS, Issue 6 2001Aziz Lazraq Abstract In this paper we write the PLS multivariate regression model in terms of a redundancy index and obtain some properties of the successive PLS components. We study their significance in the model and build tests of hypotheses to this effect. A stopping rule is given to obtain the right number of PLS components, and a numerical measure is defined to assess the overall quality of the PLS regression. Finally, an algorithm is given explicitly as it was coded in S-Plus and is applied in some chosen examples. Copyright © 2001 John Wiley & Sons, Ltd. [source] Toward robust QSPR models: Synergistic utilization of robust regression and variable eliminationJOURNAL OF COMPUTATIONAL CHEMISTRY, Issue 6 2008Rainer Grohmann Abstract Widely used regression approaches in modeling quantitative structure,property relationships, such as PLS regression, are highly susceptible to outlying observations that will impair the prognostic value of a model. Our aim is to compile homogeneous datasets as the basis for regression modeling by removing outlying compounds and applying variable selection. We investigate different approaches to create robust, outlier-resistant regression models in the field of prediction of drug molecules' permeability. The objective is to join the strength of outlier detection and variable elimination increasing the predictive power of prognostic regression models. In conclusion, outlier detection is employed to identify multiple, homogeneous data subsets for regression modeling. © 2007 Wiley Periodicals, Inc. J Comput Chem 2008 [source] Establishing quantitative in-line analysis of multiple solid-state transformations during dehydrationJOURNAL OF PHARMACEUTICAL SCIENCES, Issue 11 2008Karin Kogermann Abstract The aim of the study was to conduct quantitative solid phase analysis of piroxicam (PRX) and carbamazepine (CBZ) during isothermal dehydration in situ, and additionally exploit the constructed quantitative models to analyze the solid-state forms in-line during fluidized bed drying. Vibrational spectroscopy (near-infrared (NIR), Raman) was employed for monitoring the dehydration and the quantitative model was based on partial least squares (PLS) regression. PLS quantification was confirmed experimentally using isothermal thermogravimetric analysis (TGA) and X-ray powder diffractometry (XRPD). To appraise the quality of quantitative models several model parameters were evaluated. The hot-stage spectroscopy quantification results were found to be in reasonable agreement with TGA and XRPD results. Quantification of PRX forms showed complementary results with both spectroscopic techniques. The solid-state forms observed during CBZ dihydrate dehydration were quantified with Raman spectroscopy, but NIR spectroscopy failed to differentiate between the anhydrous solid-state forms of CBZ. In addition to in situ dehydration quantification, Raman spectroscopy in combination with PLS regression enabled in-line analysis of the solid-state transformations of CBZ during dehydration in a fluidized bed dryer. © 2008 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 97:4983,4999, 2008 [source] ESTIMATION OF HEDONIC RESPONSES FROM DESCRIPTIVE SKIN SENSORY DATA BY CHI-SQUARE MINIMIZATIONJOURNAL OF SENSORY STUDIES, Issue 1 2006I.F. ALMEIDA ABSTRACT Six topical formulations were evaluated by a trained panel according to a descriptive analysis methodology and by a group of consumers who rated the products on a hedonic scale. We present a new approach that describes the categorical appreciation of appearance, texture and skinfeel of the formulations by the consumers as a function of related sensory attributes assessed by the trained panel. For each hedonic attribute, a latent random variable depending on the sensory attributes is constructed and made discrete (in a nonlinear fashion) according to the distribution of consumer-hedonic scores in such a way as to minimize a corresponding chi-square criterion. Standard partial least squares (PLS) regression, bootstrapping and cross-validation techniques describing the overall liking of the hedonic attributes as a function of associated sensory attributes were also applied. Results from both methods were compared, and it was concluded that chi-square minimization can work as a complementary method to the PLS regression. [source] Determination of the secondary structure of proteins in different environments by FTIR-ATR spectroscopy and PLS regressionBIOPOLYMERS, Issue 11 2008Yeqiu Wang Abstract The secondary structures of proteins (,-helical, ,-sheet, ,-turn, and random coil) in the solid state and when bound to polymer beads, containing immobilized phenyl and butyl ligands such as those as commonly employed in hydrophobic interaction chromatography, have been investigated using FTIR-ATR spectroscopy and partial least squares (PLS) methods. Proteins with known structural features were used as models, including 12 proteins in the solid state and 7 proteins adsorbed onto the hydrophobic surfaces. A strong PLS correlation was achieved between predictions derived from the experimental data for 4 proteins adsorbed onto the phenyl-modified beads and reference data obtained from the X-ray crystallographic structures with r2 values of 0.9974, 0.9864, 0.9924, and 0.9743 for ,-helical, ,-sheet, ,-turn, and random coiled structures, respectively. On the other hand, proteins adsorbed onto the butyl sorbent underwent greater secondary structural changes compared to the phenyl sorbent as evidenced from the poorer PLS r2 values (r2 are 0.9658, 0.9106, 0.9571, and 0.9340). The results thus indicate that the secondary structures for these proteins were more affected by the butyl sorbent, whereas the secondary structure remains relatively unchanged for the proteins adsorbed onto the phenyl sorbent. This study has important ramifications for understanding the nature of protein secondary structural changes following adsorption onto hydrophobic sorbent surfaces. This knowledge could also enable the development of useful protocols for enhancing the chromatographic purification of proteins in their native bioactive states. © 2008 Wiley Periodicals, Inc. Biopolymers 89: 895,905, 2008. This article was originally published online as an accepted preprint. The "Published Online" date corresponds to the preprint version. You can request a copy of the preprint by emailing the Biopolymers editorial office at biopolymers@wiley.com [source] Fluorescence-based soft-sensor for monitoring ,-lactoglobulin and ,-lactalbumin solubility during thermal aggregationBIOTECHNOLOGY & BIOENGINEERING, Issue 3 2008Rand Elshereef Abstract A soft-sensor for monitoring solubility of native-like ,-lactalbumin (,-LA) and ,-lactoglobulin (,-LG) and their aggregation behavior following heat treatment of mixtures under different treatment conditions was developed using fluorescence spectroscopy data regressed with a multivariate Partial Least Squares (PLS) regression algorithm. PLS regression was used to correlate the concentrations of ,-LA and ,-LG to the fluorescence spectra obtained for their mixtures. Data for the calibration and validation of the soft sensor was derived from fluorescence spectra. The process of thermal induced aggregation of ,-LG and ,-LA protein in mixtures, which involves the disappearance of native-like proteins, was studied under various treatment conditions including different temperatures, pH, total initial protein concentration and proportions of ,-LA and ,-LG. It was demonstrated that the multivariate regression models used could effectively deconvolute multi-wavelength fluorescence spectra collected under a variety of process conditions and provide a fairly accurate quantification of respective native-like proteins despite the significant overlapping between their emission profiles. It was also demonstrated that a PLS model can be used as a black-box prediction tool for estimating protein aggregation when combined with simple mass balances. Bioeng. 2008;99: 567,577. © 2007 Wiley Periodicals, Inc. [source] | |||||||||||||