Boosted regression trees, multivariate adaptive regression splines and their two-step combinations with multiple linear regression or partial least squares to predict blood-brain barrier passage: a case study

The use of some unconventional non-linear modeling techniques, i.e. classification and regression trees and multivariate adaptive regression splines-based methods, was explored to model the blood-brain barrier (BBB) passage of drugs and drug-like molecules. The data set contains BBB passage values for 299 structural and pharmacological diverse drugs, originating from a structured knowledge-based database. Models were built using boosted regression trees (BRT) and multivariate adaptive regression splines (MARS), as well as their respective combinations with stepwise multiple linear regression (MLR) and partial least squares (PLS) regression in two-step approaches. The best models were obtained using combinations of MARS with either stepwise MLR or PLS. It could be concluded that the use of combinations of a linear with a non-linear modeling technique results in some improved properties compared to the individual linear and non-linear models and that, when the use of such a combination is appropriate, combinations using MARS as non-linear technique should be preferred over those with BRT, due to some serious drawbacks of the BRT approaches.     

Determining orthogonal and similar chromatographic systems from the injection of mixtures in liquid chromatography-diode array detection and the interpretation of correlation coefficients color maps

Generic orthogonal chromatographic systems might be helpful tools as potential starting points in the development of methods to separate impurities and the active substance in drugs with unknown impurity profiles. The orthogonality of 38 chromatographic systems was evaluated from weighted-average-linkage dendrograms and color maps, both based on the correlation coefficients between the retention factors on the different systems. On each chromatographic system, 68 drug substances were injected as mixtures of three or four components to increase the throughput. The (overlapping) peaks were identified and resolved with a peak purity algorithm, orthogonal projection approach (OPA). The visualization techniques applied allowed a simple evaluation of orthogonal and (groups of) similar systems.     

GC Methods for the Determination of Methanol and Ethanol in Insulating Mineral Oils as Markers of Cellulose Degradation in Power Transformers

Methanol and ethanol in transformer oils have been recently proposed as new markers of thermal and mechanical degradation of cellulose (the solid insulation in power transformers). In this work, we optimized and compared the performance of two headspace gas chromatographic methods based on flame ionization (HS–GC–FID) and mass spectrometry detection (HS–GC–MS) to determine methanol and ethanol in insulating mineral oil. For methanol and ethanol, the detection limits were 12 and 27 μg kg^?1 (HS–GC–FID) and 1.3 and 3.1 μg kg^?1 (HS–GC–MS). Repeatability was evaluated in transformer oils for both the methods at different concentration levels of analytes and RSD values were found to lie between 1.8 and 16 %. The accuracy of the methods was assessed under a proficiency test (Cigré JWG A2/D1.46). The methods were compared by a F-test and a one-sided paired t test performed on 21 transformer oils in service. Correlations of methanol and ethanol content in sampled oils against their actual time of service are provided. For each sample, the content of traditional markers (furan-2-carbaldehyde and CO₂) was also measured, finding a correlation between light alcohols and CO₂ content. This indicates that methanol and ethanol determination may be helpful in providing further information on the thermal degradation conditions of transformers’ solid insulation. The method developed is currently routinely applied by the laboratories of Sea Marconi Technologies for the assessment of transformers’ conditions.     

Comparative enantioseparations of pharmaceuticals in capillary electrochromatography on polysaccharide-based chiral stationary phases containing selectors with or without chlorinated derivatives

The screening conditions of an existing chiral strategy in CEC were tested for their applicability on four chlorine-containing polysaccharide-based stationary phases. The selectors of these phases are cellulose tris(3-chloro-4-methylphenylcarbamate), amylose tris(5-chloro-2-methylphenylcarbamate), cellulose tris(4-chloro-3-methylphenylcarbamate) and cellulose tris(3,5-dichlorophenylcarbamate). The enantioselectivity of these phases was compared with those of the four phases without chlorine (Chiralpak? AD-RH, Chiralcel? OD-RH, Chiralpak? AS-RH and Chiralcel? OJ-RH) used in the earlier defined strategy. A test set of 48 structurally diverse drug compounds was analyzed using the screening conditions of the strategy. These results led to possibilities to upgrade the current screening strategy so that improved success rates are obtained. The chlorine-containing chiral stationary phases demonstrated an added value to the screening process since they showed enantioresolution for compounds not resolved by the chiral stationary phases not containing chlorine in their structure.     

Predictive-property-ranked variable reduction in partial least squares modelling with final complexity adapted models: Comparison of properties for ranking

The calibration performance of partial least squares regression for one response (PLS1) can be improved by eliminating uninformative variables. Many variable-reduction methods are based on so-called predictor-variable properties or predictive properties, which are functions of various PLS-model parameters, and which may change during the steps of the variable-reduction process. Recently, a new predictive-property-ranked variable reduction method with final complexity adapted models, denoted as PPRVR-FCAM or simply FCAM, was introduced. It is a backward variable elimination method applied on the predictive-property-ranked variables. The variable number is first reduced, with constant PLS1 model complexity A, until A variables remain, followed by a further decrease in PLS complexity, allowing the final selection of small numbers of variables.     

Development of a capillary electrophoresis method for the separation of flavonolignans in silymarin complex

CE method for the baseline separation of structurally similar flavonolignans silybin A, silybin B, isosilybin A, isosilybin B, silychristin, silydianin, and their precursor taxifolin in silymarin complex has been developed and validated. The optimized background electrolyte was 100 mmol/L boric acid (pH 9.0) containing 5 mmol/L heptakis(2,3,6-tri-O-methyl)-β-CD and 10% (v/v) of methanol. The separation was carried out in an 80.5/72 cm (50 μm id) fused silica capillary at +25 kV with UV detection at 200 nm. Genistein (10 μg/mL) was used as internal standard. The resolution between the diastereomers of silybin and isosilybin was 1.73 and 2.59, respectively. The method was validated for each analyte in a concentration range of 2.5–50 μg/mL. The calibration curves were rectilinear with correlation coefficients ≥0.9972. The method was applied to determine flavonolignans in two dietary supplements containing Silybum marianum extract. The accuracy was evaluated by comparing the results of the CE analyses of the dietary supplements with those of the reference United States Pharmacopeial HPLC method. The unpaired t-test did not show a statistically significant difference between the results of both the proposed CE and the reference method (p > 0.05, n = 3).     

Anthracene silica gel,a new polycyclic-aromatic-bonded stationary phase for HPLC

Summary N(9-methylanthracene)aminopropyl silica gel is introduced as a new reversed-phase packing material. It has specific selectivity for some aromatic compounds. In this paper the anthracene phase is compared with commercial octadecyl-, phenyl- and aminopropyl-bonded silica gels.     

Fast and convenient base-mediated synthesis of 3-substituted quinolines

In a convenient method, 3-substituted quinolines are readily synthesized in a two-step process with initial oxazine formation and subsequent base-mediated cyclization.     

The HITRAN 2008 molecular spectroscopic database

This paper describes the status of the 2008 edition of the HITRAN molecular spectroscopic database. The new edition is the first official public release since the 2004 edition, although a number of crucial updates had been made available online since 2004. The HITRAN compilation consists of several components that serve as input for radiative-transfer calculation codes: individual line parameters for the microwave through visible spectra of molecules in the gas phase; absorption cross-sections for molecules having dense spectral features, i.e. spectra in which the individual lines are not resolved; individual line parameters and absorption cross-sections for bands in the ultraviolet; refractive indices of aerosols, tables and files of general properties associated with the database; and database management software. The line-by-line portion of the database contains spectroscopic parameters for 42 molecules including many of their isotopologues.     

The synthesis of polyamine networks using cationic living poly(1-tert-butylaziridine)

Polyamine networks have been synthesized by end-linking bifunctional living poly-TBA with polyfunctional nucleophilic reagents. The bifunctional living poly-TBA was obtained by polymerization of TBA initiated by low molecular weight bifunctional living poly(tetrahydrofuran). The swelling degrees of the networks in several solvents are reported.