Induction of decision trees using genetic programming for modelling ecotoxicity data: adaptive discretization of real-valued endpoints

Recent literature has demonstrated the applicability of genetic programming to induction of decision trees for modelling toxicity endpoints. Compared with other decision tree induction techniques that are based upon recursive partitioning employing greedy searches to choose the best splitting attribute and value at each node that will necessarily miss regions of the search space, the genetic programming based approach can overcome the problem. However, the method still requires the discretization of the often continuous-valued toxicity endpoints prior to the tree induction. A novel extension of this method, YAdapt, is introduced in this work which models the original continuous endpoint by adaptively finding suitable ranges to describe the endpoints during the tree induction process, removing the need for discretization prior to tree induction and allowing the ordinal nature of the endpoint to be taken into account in the models built.     

Identification tree based on fragmentation rules for structure elucidation of organophosphorus esters by electrospray mass spectrometry

Organophosphorus compounds have played important roles as pesticides, chemical warfare agents and extractors of radioactive material. Structural elucidation of phosphonates poses a particular challenge because their initial forms can be hydrolyzed, thus, degradation products may predominate in samples acquired in the field. The analysis of non‐volatile organophosphorus compounds and their degradation products is possible using electrospray tandem mass spectrometry ESI‐MS/MS. Here, we present a generic strategy that allows the unambiguous identification of substituents for two families of organophosphorus compounds: the phosphonates and phosphates. General fragmentation rules were deduced based on the study of decomposition pathways of 55 organophosphorus esters, including examples found in the literature. Multistage MS (MSⁿ) experiments at high resolution in a hybrid mass spectrometer provide accurate mass measurements, whereas collision‐induced dissociation experiments in a triple quadrupole give access to small fragment ions. The creation of a specific nomenclature for each possible structure of organophosphorus compound, depending on the alkyl side chain linked to the oxygen, was achieved by applying these fragmentation rules. This led to the creation of an ‘identification tree’ based upon the unique consecutive decomposition pathways uncovered for each individual compound. Hence, seven structural motifs were created that orient an unequivocal identification using the ‘identification tree’. Despite the similar structures of the ensemble of phosphate and phosphonate esters, distinct identifications based upon characteristic neutral losses and diagnostic fragment ions were possible in all cases. Copyright © 2013 John Wiley & Sons, Ltd.     

Three‐unit semicircles curve: A compact 3D graphical representation of DNA sequences based on classifications of nucleotides

A new three‐dimensional graphical representation of DNA sequences, three‐unit semicircles (TUS)‐curve, which maps a given sequence into a dot sequences embedded in three‐unit semicircles, is proposed based on three biclassifications of nucleotides. TUS‐curve has the merit of compactness and could avoid the degeneracy and loss of information. The geometrical center of the curve, which indicates the distribution of base frequencies of the corresponding DNA sequence, is extracted and applied to analyze the similarity of various species. Phylogenetic tree of 11 species based on their first exons of β‐globin genes showed that the TUS‐curve is a powerful tool to get valuable biological information. © 2011 Wiley Periodicals, Inc. Int J Quantum Chem, 2011     

Identification of triacylglycerol using automated annotation of high resolution multistage mass spectral trees

High complexity of identification for non-target triacylglycerols (TAGs) is a major challenge in lipidomics analysis. To identify non-target TAGs, a powerful tool named accurate MSⁿ spectrometry generating so-called ion trees is used. In this paper, we presented a technique for efficient structural elucidation of TAGs on MSⁿ spectral trees produced by LTQ Orbitrap MSⁿ, which was implemented as an open source software package, or TIT. The TIT software was used to support automatic annotation of non-target TAGs on MSⁿ ion trees from a self-built fragment ion database. This database includes 19108 simulate TAG molecules from a random combination of fatty acids and corresponding 500582 self-built multistage fragment ions (MS ≤ 3). Our software can identify TAGs using a “stage-by-stage elimination” strategy. By utilizing the MS¹ accurate mass and referenced RKMD, the TIT software can discriminate unique elemental composition candidates. The regiospecific isomers of fatty acyl chains will be distinguished using MS² and MS³ fragment spectra. We applied the algorithm to the selection of 45 TAG standards and demonstrated that the molecular ions could be 100% correctly assigned. Therefore, the TIT software could be applied to TAG identification in complex biological samples such as mouse plasma extracts.     

Graphical approach to correlations in high‐spin systems

We present a graphical technique for generating and indexing spin monomials of high‐spin systems. The procedure consists of developing a graph with at most n line segments from each node in a given row to the one immediately lower, where n is the multiplicity of single‐particle spin function. The paths lead to monomials with definite M values at each node. This technique has been used to generate and diagonalize the model spin Hamiltonian. ©1999 John Wiley & Sons, Inc. Int J Quant Chem 73: 389–393, 1999     

On the Randić Index of Acyclic Conjugated Molecules

The Randić index of an organic molecule whose molecular graph is G is the sum of the weights (d(u)d(v))^−1/2 of all edges uv of G, where d(u) and d(v) are the degrees of the vertices u and v in G. We give a sharp lower bound on the Randić index of conjugated trees (trees with a perfect matching) in terms of the number of vertices. A sharp lower bound on the Randić index of trees with a given size of matching is also given Mei Lu: Partially supported by NNSFC (No. 60172005) Lian-zhu Zhang: Partially supported by NNSFC (No. 10271105) Feng Tian: Partially supported by NNSFC (No. 10431020)     

Covariate model-based fault tree analysis for risk assessment in chemical process industries: A case study of a chlorine manufacturing facility

Fault tree analysis (FTA) is a promising quantitative technique for risk analysis in chemical process industries (CPIs). In FTA, a certain sequence of basic events (causes) leads to one specific Top event (critical event of interest). However, the conventional fault tree analysis has the limitations of staticity and uncertainty. The staticity in conventional FTA arises due to its inability to accommodate time-dependent characteristics of the process system. Whereas uncertainty primarily lies in the failure probability data of basic events. This paper proposes an innovative methodology that uses a time-dependent covariate model to update the failure probability values of major contributing basic events in FTA. A novel subclass of the family of phase-type distributions is used to model the covariates corresponding to the basic events. The newly developed methodology is applied for a case study in a chlorine manufacturing facility to estimate the chlorine release probability. The blockage in the pipeline was identified as the significant reason for chlorine release from expert opinion and sensitivity analysis. The results of the proposed model of FTA are compared with that of conventional FTA.     

HPLC in the investigation of taxonomic problems. Classification of poplar genotypes

Summary The RP-HPLC technique has been applied to the analysis of the flavonoid content of leaf extracts from various genotypes ofPopulus nigra andPopulus deltoides and from several of their hybrids. Two elution procedures different either as regards gradient program or the type of organic modifier used were checked to optimize the reliability of the results obtained. Before making any identification, the chromatographic profiles proved so typical of each species as to allow of their use as fingerprints in this type of taxonomic problem. Individual identifications were carried out by spiking real samples with known amounts of standard substances. The presence of flavonoid compounds previously identified was confirmed and a few new identifications were proposed.     

Surface modification of date palm activated carbonaceous materials for heavy metal removal and CO2 adsorption

In this study, high surface area activated carbon (AC) was prepared from a local palm tree (Phoenix Dactylifera) using a variety of metal carbonates activators and finally achieved an excellent S_BET of 2700 m²/g when Cs₂CO₃ was used as an activating agent at a temperature of 600 °C. Surface modification of AC was carried out using various nitrogen transporting agents, resulting in N-doped ACs with nitrogen content varying from 4.0 to 11.4 %, depending on the functionalizing agents and activators used. The bimodal (presence of micro- as well as meso-porosity) ACs with such excellent surface properties were studied for their CO₂ uptake capacity at two different temperatures (0 and 25 °C) by isotherms recorded at pressure 1 bar and showed a remarkable uptake ability of 3.52 mmol/g (at 25 °C) and 5.6 mmol/g (at 0 °C), respectively. Also, batch experiments with variable pH, contact time, adsorbate concentrations, adsorbent dose, and temperatures were evaluated to understand the mechanism of sorption phenomena of Cr(VI) and Pb(II) achieving > 99.9 % removal capacity by the prepared ACs. Depending on the heavy metal ions being investigated, it was revealed that the pH of the solution and the amount of adsorbent had a direct impact on the total adsorption ability. Nitrogen atoms doped into the carbon frameworks were found to enhance the adsorption in the case of Pb(II) while the removal of Cr(VI) appeared to be unaffected. Maximum adsorption for Cr(VI) was observed at pH 2 and was determined to follow Freundlich isotherm while that of Pb(II) was observed at pH 7 and follows Langmuir isotherm. Best adsorption was found at an adsorbate concentration of 10 ppm and an adsorbent dose of 10 g/L. Kinetic modeling parameters showed the applicability of pseudo-second-order model perfectly.     

The boson gas on a Cayley tree

We analyze the free boson gas on a Cayley tree using two alternative methods. The spectrum of the lattice Laplacian on a finite tree is obtained using a direct iterative method for solving the associated characteristic equation and also using a random walk representation for the corresponding fermion lattice gas. The existence of the thermodynamic limit for the pressure of the boson lattice gas is proven and it is shown that the model exhibits boson condensation into the ground state. The random walk representation is also used to derive an expression for the Bethe approximation to the infinite-volume spectrum. This spectrum turns out to be continuous instead of a dense point spectrum, but there is still boson condensation in this approximation.