Umeå University's logo

The aim of this study was to develop a structure-property model for membrane partitioningof oligopeptides using statistical design methods and multivariate data analysis. A set of 20tetrapeptides with optional N-methylations at residues 2 and 4 was designed by a D-optimaldesign procedure. After synthesis and purification, the membrane partitioning abilities of thepeptides were tested in two chromatographic systems with phospholipids as the stationaryphase: immobilized artificial membrane chromatography (IAM) and immobilized liposomechromatography (ILC). The relationship between these measures and three different sets ofcalculated descriptors was analyzed by partial least-squares projection to latent structures(PLS). The descriptors used were the molecular surface area, Molsurf parameters, and Volsurfparameters. All three models were of good statistical quality and supported that a largehydrogen-bonding potential and the presence of a negative charge impair membrane partitioning,whereas hydrophobic parameters promote partitioning. The findings are in accordancewith what has been found for absorption of known drugs and have implications for the designof peptide-like drugs with good oral bioavailability.

In this thesis, attention has been focused on the quantitative description of nucleic acids, proteins and peptides. The strategy was to use multivariate chemometrical methods for improving the understanding of the complex structural codes of these kinds of biological molecules. Tools have been developed that enable quantitative modelling of biological molecules, i.e. models based on data that quantitatively describes their properties. The advantage of such models is that they provide interpretations in terms of chemical characteristics for complex features such as similarity, dissimilarity and potency.

By a multivariate physical-chemical characterization of the building blocks of nucleic acids and proteins, i.e. nucleosides and amino acids, descriptive scales have been developed, so called principal properties. The scales give a description of the intrinsic properties of these building blocks. The multivariate characterization results in a multi-property matrix. A principal component analysis of the multi-property matrix gives a small number of latent variables which are considered as the principal properties of the characterized molecules.

The principal property scales may be used for a wide range of different purposes, such as detecting trends and groupings in large sequence data sets, and for analyzing quantitative relationships between structure and function. In statistical experimental design, the descriptors are well suited as design variables to select combinations of amino acids in such a way that they span a wide range of properties.

The use of these principal property descriptors is demonstrated in the quantitative modelling of relationships between structure and activity of various peptide series, DNA-promoters and in the quantitative modelling of transfer ribonucleic acid sequence data (tRNA).

Twenty nucleosides occurring in transfer ribonucleic acid (tRNA) have been characterized using 21 experimentally determined (HPLC, TLC, NMR, etc.) and calculated (log P, van der Waals surface area, ionization potential, etc.) variables. Principal component analysis (PCA) was performed on the data set and four statistically significant components or principal properties (PPs) were extracted. The PPs described 68·4% of the variance in the data. The PP values are discussed in terms of similarity and dissimilarity among the nucleosides. The loading vectors from the PCA are used for an interpretation of the nature of the PP vectors. Application of the PPs in sequence-activity modelling is demonstrated with 25 DNA-promoter sequences originating from E. coli.

A multivariate quantitative physicochemical characterization of the five bases adenine (A), cytosine (C), guanine (G), thymine (T) and uracil (U), followed by principal component analysis, shows that the relative dissimilarities between the bases of DNA (A, C, G and T) are almost the same (i.e. balanced). In contrast, mRNA (containing U instead of T) has a considerably larger relative physicochemical similarity between C and U than between all other pairs of bases and is therefore inherently more unbalanced. These results provide a physicochemical explanation of the presence of thymine instead of uracil as an element of DNA. The principal component scores enable a quantitative description of nucleic acid sequence data to be made for structure-activity modelling purposes.

Biopolymer sequences (e.g., DNA, RNA, proteins and polysaccharides) and chemical processes (e.g., a batch or continuous polymer synthesis run in a chemical plant) have close similarities from the modelling point of view. When a set of sequences or processes is characterized by multivariate data, a three-way data matrix is obtained. With sequences the position and with processes the time is one direction in this matrix. The multivariate modelling of this matrix by principal component analysis (PCA) or partial least-squares (PLS) methods for the following purposes is discussed: classification of sequences; quantitative relationships between sequence and biological activity or chemical properties; optimizing a sequence with respect to selected properties; process diagnostics; and quantitative relationships between process variables and product quality variables. To obtain good models, a number of problems have to be adequately dealt with: appropriate characterization of the sequence or process; experimental design (selecting sequences or process settings); transforming the three-way into a two-way matrix; and appropriate modelling and validation (modelling interactions, periodicities, “time series” structures and “neighbour effects”). A multivariate approach to sequence and process modelling using PCA and PLS projections to latent structures is discussed and illustrated with several sets of peptide and DNA promoter data.

A novel quantitative multivariate approach for describing and analyse tRNA sequence data is presented. This approach is based on a multivariate chemical description of each nucleoside in the sequence. 30 theoretically calculated descriptors were used to characterize 63 nucleosides, and principal component analysis was used to extract the main variation from this multivariate description. The resulting four principal properties were interpreted as (PPa) size/bulk of the nucleoside, (PPb) polarity/hydrophobicity of the nucleoside, (PPc) electronic properties of the nucleoside and (PPd)polarity and size of the ribose moiety. These principal properties may be used to translate the tRNA letter sequence data into a quantitative chemical representation. We demonstrate the use of this quantitative description with a multivariate analysis of a set of tRNA sequences. This analysis gives models that are interpretable in terms of wich sequence positions, and nucleoside properties that discriminate the different isoacceptors. This approach is applicable on all kinds of RNA sequence data and gives information that is complementary to current sequence analysis techniques.