Journal of Chemometrics

Cover image for Vol. 26 Issue 10

October 2012

Volume 26, Issue 10

Pages i–iii, 497–555

  1. Issue Information

    1. Top of page
    2. Issue Information
    3. Review
    4. Research Articles
    1. Issue Information (pages i–iii)

      Version of Record online: 8 OCT 2012 | DOI: 10.1002/cem.2411

  2. Review

    1. Top of page
    2. Issue Information
    3. Review
    4. Research Articles
    1. Methods of studies on quantitative structure–activity relationships for chiral compounds (pages 497–508)

      Qing-You Zhang, Li-Zhuang Xu, Jing-Ya Li, Dan-Dan Zhang, Hai-Lin Long, Ji-Yan Leng and Lu Xu

      Version of Record online: 23 AUG 2012 | DOI: 10.1002/cem.2464

      Chiral compounds are very important in drug development, organic synthesis, materials science, toxicology, or environmental chemistry. Chirality belongs to 3-D category, thus the requirements to create new chiral methods or to extend the existed two-dimensional indices to be suitable for the three-dimensional descriptions are increasing. In this paper, the algorithms of some new method for the derivations of the parameters and applications for studies on quantitative structure-activity/property relationships of chiral molecules are reviewed.

  3. Research Articles

    1. Top of page
    2. Issue Information
    3. Review
    4. Research Articles
    1. An evaluation of experimental design in QSAR modelling utilizing the k-medoid clustering (pages 509–517)

      Stefan Brandmaier, Igor V. Tetko and Tomas Öberg

      Version of Record online: 19 JUN 2012 | DOI: 10.1002/cem.2459

      We compared the performance of models derived from a selection based on the k-medoid approach with that of quantitative structure–activity relationship models derived using a number of other approaches, such as space-filling design and D-optimal design. We validated the performance on four datasets with different endpoints, representing toxicity, boiling point, bioconcentration and an adsorption coefficient. Compared with the models derived from the compounds selected by the other examined approaches, those models derived with the k-medoid selection show high reliability for experimental design, as their performance was constantly among the best for all the examined datasets.

    2. Demasking large dummy effects approach in revealing important interactions in Plackett–Burman experimental design (pages 518–525)

      Tijana Rakić, Biljana Jančić-Stojanović, Andjelija Malenović, Darko Ivanović and Mirjana Medenica

      Version of Record online: 27 JUN 2012 | DOI: 10.1002/cem.2461

      Newly developed demasking large dummy effects (DDE) approach for identification of significant factor interactions in Plackett–Burman experimental design is proposed. Dummy factors, as imaginary factors, are not expected to contribute to the model. The DDE approach presents the algorithm for revealing the important interactions hidden behind large dummy effects. The DDE approach showed to create the models that have great statistical power. These models are particularly suitable for describing the real systems, which increases their practical value.

    3. Reduction of the rotational ambiguity of curve resolution techniques under partial knowledge of the factors. Complementarity and coupling theorems (pages 526–537)

      Mathias Sawall, Christian Fischer, Detlef Heller and Klaus Neymeyr

      Version of Record online: 20 JUN 2012 | DOI: 10.1002/cem.2458

      The rotational ambiguity of curve resolution techniques can successfully be reduced if additional knowledge on the chemical system is available. Complementarity and coupling theorems are presented, which provide affine–linear restrictions on the admissible concentration profiles and on the spectra. Numerical results are given for a three-component model problem as well as for a catalytic system, namely, the formation of hafnacyclopentene.

    4. Model-free multivariate curve resolution combined with model-based kinetics: algorithm and applications (pages 538–548)

      Mathias Sawall, Armin Börner, Christoph Kubis, Detlef Selent, Ralf Ludwig and Klaus Neymeyr

      Version of Record online: 24 JUL 2012 | DOI: 10.1002/cem.2463

      Model-free multivariate curve resolution techniques suffer from the rotational ambiguity of the solution. Any additional constraints on the solution can help to reduce the non-uniqueness. In this paper, the distance functional between the concentration profiles from a self-modeling curve resolution method and those, which are computed by a kinetic model, is used to regularize the reconstruction error functional. The numerical algorithm uses an overall optimization and allows to reduce the ambiguity drastically.

    5. A descriptor of amino acids SVWG and its applications in peptide QSAR (pages 549–555)

      Jianbo Tong, Yang Chen, Shuling Liu, Ting Che and Xiameng Xu

      Version of Record online: 7 SEP 2012 | DOI: 10.1002/cem.2465

      The SVWG, principal component scores vector for weighted holistic invariant molecular descriptors and geometry, topology, and atom-weights assembly descriptors, was derived from principal component analysis of a matrix of two structural variables of coded amino acid. SVWG scales were then applied in three polypeptides. The obtained models with the correlation coefficient, cross-validation correlation coefficient were 0.893 and 0.830 for angiotensin-converting enzyme inhibitors, 0.990 and 0.968 for antimicrobial peptides, and 0.955 and 0.908 for oxytocin, respectively. Satisfactory results showed that SVWG scales can be a useful structural expression methodology for study on peptide QSAR.