Marjan Vračko

Novel 2-D graphical representation of DNA sequences and their numerical characterization

We consider a novel 2-D graphical representation of DNA sequences preserving information on sequential adjacency of bases and allowing numerical characterization. The representation avoids loss of information accompanying alternative 2-D representations in which the curve standing for DNA overlaps and intersects itself. The method is illustrated on the coding sequence of the first exon of human β-globin gene.

Analysis of similarity/dissimilarity of DNA sequences based on novel 2-D graphical representation

The recently proposed 2-D graphical representation of DNA based on four horizontal lines involves an arbitrary assignment of the four types of bases to the lines. While each such assignment is legitimate, it is desirable to have a scheme free of such arbitrary choices among non-equivalent geometrical representations. We outline one such approach, which is based on the construction of a 12-component vector whose components are the leading eigenvalues of the L/L matrices associated with DNA. The examination of similarities/dissimilarities among the coding sequences of the first exon of b-globin gene of different species illustrates the utility of the approach. 2003 Elsevier Science B.V. All rights reserved.

On the similarity of DNA primary sequences.

We consider numerical characterization of graphical representations of DNA primary sequences. In particular we consider graphical representation of DNA of beta-globins of several species, including human, on the basis of the approach of A. Nandy in which nucleic bases are associated with a walk over integral points of a Cartesian x, y-coordinate system. With a so-generated graphical representation of DNA, we associate a distance/distance matrix, the elements of which are given by the quotient of the Euclidean and the graph theoretical distances, that is, through the space and through the bond distances for pairs of bases of graphical representation of DNA. We use eigenvalues of so-constructed matrices to characterize individual DNA sequences. The eigenvalues are used to construct numerical sequences, which are subsequently used for similarity/dissimilarity analysis. The results of such analysis have been compared and combined with similarity tables based on the frequency of occurrence of pairs of bases.

New public QSAR model for carcinogenicity

BackgroundOne of the main goals of the new chemical regulation REACH (Registration, Evaluation and Authorization of Chemicals) is to fulfill the gaps in data concerned with properties of chemicals affecting the human health. (Q)SAR models are accepted as a suitable source of information. The EU funded CAESAR project aimed to develop models for prediction of 5 endpoints for regulatory purposes. Carcinogenicity is one of the endpoints under consideration.ResultsModels for prediction of carcinogenic potency according to specific requirements of Chemical regulation were developed. The dataset of 805 non-congeneric chemicals extracted from Carcinogenic Potency Database (CPDBAS) was used. Counter Propagation Artificial Neural Network (CP ANN) algorithm was implemented. In the article two alternative models for prediction carcinogenicity are described. The first model employed eight MDL descriptors (model A) and the second one twelve Dragon descriptors (model B). CAESARs models have been assessed according to the OECD principles for the validation of QSAR. For the model validity we used a wide series of statistical checks. Models A and B yielded accuracy of training set (644 compounds) equal to 91% and 89% correspondingly; the accuracy of the test set (161 compounds) was 73% and 69%, while the specificity was 69% and 61%, respectively. Sensitivity in both cases was equal to 75%. The accuracy of the leave 20% out cross validation for the training set of models A and B was equal to 66% and 62% respectively. To verify if the models perform correctly on new compounds the external validation was carried out. The external test set was composed of 738 compounds. We obtained accuracy of external validation equal to 61.4% and 60.0%, sensitivity 64.0% and 61.8% and specificity equal to 58.9% and 58.4% respectively for models A and B.ConclusionCarcinogenicity is a particularly important endpoint and it is expected that QSAR models will not replace the human experts opinions and conventional methods. However, we believe that combination of several methods will provide useful support to the overall evaluation of carcinogenicity. In present paper models for classification of carcinogenic compounds using MDL and Dragon descriptors were developed. Models could be used to set priorities among chemicals for further testing. The models at the CAESAR site were implemented in java and are publicly accessible.

Compact 2-D graphical representation of DNA

Abstract We present a novel 2-D graphical representation for DNA sequences which has an important advantage over the existing graphical representations of DNA in being very compact. It is based on: (1) use of binary labels for the four nucleic acid bases, and (2) use of the ‘worm’ curve as template on which binary codes are placed. The approach is illustrated on DNA sequences of the first exon of human β-globin and gorilla β-globin.

A STUDY OF STRUCTURE-CARCINOGENIC POTENCY RELATIONSHIP WITH ARTIFICIAL NEURAL NETWORKS. THE USING OF DESCRIPTORS RELATED TO GEOMETRICAL AND ELECTRONIC STRUCTURES

This contribution is an attempt to estimate carcinogenic potency (measured in TD50 dose) of molecules using artificial neural networks (ANN) with counterpropagation learning strategy. Three kinds of descriptors have been tested:  geometrical structures of molecules, which have been described with 3D coordinates of all atoms, geometrical structures in combination with atomic charges, and energy spectra of occupied orbitals, i.e., the electronic structures. Structures or structures plus atomic charges have been represented with “spectrum-like” representation, which is suitable as input for ANN modelling. A set of 45 benzene derivatives was considered in this study. The models were able to recognize structures of training set, and a weak correlation between descriptors and carcinogenic potency was found.

Validation of counter propagation neural network models for predictive toxicology according to the OECD principles: a case study

The OECD has proposed five principles for validation of QSAR models used for regulatory purposes. Here we present a case study investigating how these principles can be applied to models based on Kohonen and counter propagation neural networks. The study is based on a counter propagation network model that has been built using toxicity data in fish fathead minnow for 541 compounds. The study demonstrates that most, if not all, of the OECD criteria may be met when modeling using this neural network approach.

Modeling toxicity by using supervised kohonen neural networks.

Counterprogation neural network is shown to be a powerful and suitable tool for the investigation of toxicity. This study mined a data set of 568 chemicals. Two hundred eighty-two objects were used as the training set and 286 as the test set. The final model developed presents high performances on the data set R(2) = 0.83 (R(2) = 0.97 on the training set, R(2) = 0.59 on the test set). This technique distinguishes itself also for the ability to give to the expert two-dimensional maps suitable for the study of the distribution/clustering of the data and the identification of outliers.

On novel representation of proteins based on amino acid adjacency matrix

A novel characterization of proteins is presented based on selected properties of recently introduced 20 × 20 amino acid adjacency matrix of proteins in which matrix elements count the occurrence of all 400 possible pair-wise adjacencies obtained by reading protein primary sequence from the left to the right. In particular we consider the characterization based on the sum and the difference of the rows and the corresponding columns, which characterize proteins by a pair of 20-component vectors. The approach is illustrated on a set of ND6 proteins of eight species. †Visitor, Emeritus, Department of Mathematics and Computer Science, Drake University, Des Moines, IA, USA.

Structure-mutagenicity modelling using counter propagation neural networks

The set of 95 aromatic amines and their mutagenic potency was treated with counter propagation neural network, which enables analysis of self-organising maps (SOMs) and also the prediction of mutagenicity. Compounds were described with four classes of descriptors: topostructural (TS), topochemical (TC), geometrical, and quantum chemical (QC). The models were tested on their prediction ability with leave-one-out (LOO) cross-validation method. The squares of correlation coefficient lie between 0.65 and 0.75 and are comparable with models obtained by linear methods. In addition, we analysed self-organising maps and found clusters of structurally similar compounds.