3D image registration using a fast noniterative algorithm.

This note describes the implementation of a three-dimensional (3D) registration algorithm, generalizing a previous 2D version [Alexander, Int J Imaging Systems and Technology 1999;10:242-57]. The algorithm solves an integrated form of linearized image matching equation over a set of 3D rectangular sub-volumes ('patches') in the image domain. This integrated form avoids numerical instabilities due to differentiation of a noisy image over a lattice, and in addition renders the algorithm robustness to noise. Registration is implemented by first convolving the unregistered images with a set of computationally fast [O(N)] filters, providing four bandpass images for each input image, and integrating the image matching equation over the given patch. Each filter and each patch together provide an independent set of constraints on the displacement field derived by solving a set of linear regression equations. Furthermore, the filters are implemented at a variety of spatial scales, enabling registration parameters at one scale to be used as an input approximation for deriving refined values of those parameters at a finer scale of resolution. This hierarchical procedure is necessary to avoid false matches occurring. Both downsampled and oversampled (undecimating) filtering is implemented. Although the former is computationally fast, it lacks the translation invariance of the latter. Oversampling is required for accurate interpolation that is used in intermediate stages of the algorithm to reconstruct the partially registered from the unregistered image. However, downsampling is useful, and computationally efficient, for preliminary stages of registration when large mismatches are present. The 3D registration algorithm was implemented using a 12-parameter affine model for the displacement: u(x) = Ax + b. Linear interpolation was used throughout. Accuracy and timing results for registering various multislice images, obtained by scanning a melon and human volunteers in various stationary positions, is described. The algorithm may be generalized to more general models of the displacement field, and is also well suited to parallel processing.

Classification of 1H MR spectra of human brain neoplasms: the influence of preprocessing and computerized consensus diagnosis on classification accuracy.

We study how classification accuracy can be improved when both different data preprocessing methods and computerized consensus diagnosis (CCD) are applied to 1H magnetic resonance (MR) spectra of astrocytomas, meningiomas, and epileptic brain tissue. The MR spectra (360 MHz, 37 degrees C) of tissue specimens (biopsies) from subjects with meningiomas (95; 26 cases), astrocytomas (74; 26 cases), and epilepsy (37; 8 cases) were preprocessed by several methods. Each data set was partitioned into training and validation sets. Robust classification was carried out via linear discriminant analysis (LDA), artificial neural nets (NN), and CCD, and the results were compared with histopathological diagnosis of the MR specimens. Normalization of the relevant spectral regions affects classification accuracy significantly. The spectra-based average three-class classification accuracies of LDA and NN increased from 81.7% (unnormalized data sets) to 89.9% (normalized). CCD increased the classification accuracy of the normalized sets to an average of 91.8%. CCD invariably decreases the fraction of unclassifiable spectra. The same trends prevail, with improved results, for case-based classification. Preprocessing the 1H MR spectra is essential for accurate and reliable classification of astrocytomas, meningiomas, and nontumorous epileptic brain tissue. CCD improves classification accuracy, with an attendant decrease in the fraction of unclassifiable spectra or cases.

PROANAL version 2: multifunctional program for analysis of multiple protein sequence alignments and for studying the structure--activity relationships in protein families.

A new version of the program PROANAL is described. A multiple linear regression analysis of the protein structure--activity relationship allows one to investigate the combinations of protein sites and factors influencing the activity. The program also provides the possibility to seek out protein sites, conservative or variable in variations of physicochemical characteristics, and regions with high or low values of these characteristics. PROANAL2 may be useful in the simulation of protein-engineering experiments and in the search of a number of protein regions such as functional sites, secondary structures, solvent-exposed regions, T- and B-cell antigenic determinants, etc.

Artificial protein vaccines with predetermined tertiary structure: application to anti-HIV-1 vaccine design.

A successful approach to the development of a safe and effective synthetic vaccine requires that different B and T cell epitopes of the infectious agent be included in the vaccine construction. In this paper we suggest a new approach to vaccine design in the form of an artificial protein with a predetermined tertiary structure (PTS vaccines). Based on B and T cell epitope properties, we substantiate the possible use for vaccine construction of one well-known protein spatial motif--the four-alpha-helix bundle. Antigenic determinants of cellular immunity (amphipathic alpha-helices) and humoral immunity (flexible hydrophilic loop regions) are used as blocks for vaccine design. General principles of PTS vaccine construction have been applied to anti-HIV-1 vaccine design.

Algorithm and computer program Pro__Anal for analysis of relationship between structure and activity in a family of proteins or peptides.

In this paper we introduce a computer algorithm and program Pro__Anal for analysis of the structure-activity relationship in a family of evolutionarily related (and/or artificially mutated) proteins/peptides. The program uses aligned amino acid sequences with data of their activity (pK, Km, ED50 or any other) and searches for correlations between data on activity and various physico-chemical characteristics of different regions in primary structures. In automatic mode, the program generates and verifies hypotheses on the disposition of a sequential modulating region in a protein, and key characteristics of the region. In manual mode, users can generate and analyze their own hypotheses. The program is implemented on IBM PC or compatible computers. It is designed to be easily handled by the occasional computer user and yet it is powerful enough for experienced professionals. Pro__Anal operation is demonstrated on the example of finding modulating centers in a family of disintegrins-proteins from snake venoms which inhibit fibrinogen interaction with platelet receptors. In another example it is shown that the immunogenicity of peptides is connected with their positive charge.

[Design of four-helix protein--a possible vaccine against human immunodeficiency virus (HIV-1)]. / Konstruirovanie chetyrekhspiral'nogo belka--vozmozhnoi vaktsiny protiv virusa immunodefitsita cheloveka (HIV-1).

Successful approach to the development of safe and effective synthetic vaccines requires that different B- and T-cell epitopes of the infectious agent be included into the vaccine construction. It is suggested that vaccines should be constructed as proteins with both optimal epitope composition and predetermined tertiary structure. Based on analysis of B-cell and T-cell epitope properties, a possibility to use one well-known protein spatial motif--four-alpha-helix bundle--for vaccine construction is substantiated. Antigenic determinants of cellular immunity (amphipathic alpha-helices) and humoral immunity (flexible hydrophilic loop regions) can be used as blocks for vaccine design. Nonloop B-epitopes and nonhelical T-epitopes may be introduced in the protein N- and C-terminal regions. General principles of PTS-vaccine construction have been applied to anti-HIV-1 vaccine design. Experimentally studied T- and neutralizing B-cell epitopes from HIV-1 proteins were analyzed. The sequence of one possible four-alpha-helix protein vaccine has been constructed. Predicted secondary structure and T- and B-cell epitopes of this protein coincided with the planned ones. The amino acid composition of the protein was found to be consistent with the composition of globular water-soluble proteins. The gene of the protein with codon composition optimal for expression in E. coli has been synthesized. The advantages and limitations of this approach to vaccine design are discussed.

II-Q restriction endonucleases--new class of type II enzymes.

Unique restriction endonucleases Bpu 10l and Bsil have been isolated from Bacillus pumilas and Bacillus sphaericus, respectively. The recognition sequences and cleavage points of these enzymes have been determinated as 5'-CC1TNAGC-3'/3'-GGANT1CG-5' for Bpu 10l and 5'-C1TCGTG-3'/3'-GAGCA1C-5' for Bsil. Restriction endonucleases Bpu 10l and Bsil represent a new class of enzymes which recognize non-palindromic nucleotide sequences and hydrolize DNA within the recognition sequence. Bpu 10l and Bsil recognition sequences may be regarded as quasipalindromic and the enzymes may be designated as type II-Q restriction endonucleases.

[Development of rules for vaccine engineering based on the variability of peptide fragments in closely related proteins]. / Razrabotka pravil dlia vaktsinnoi inzhenerii na osnove analiza variabel'nosti peptidnykh fragmentov v blizkorodstvennykh belkakh.

Based on the protein sequence data bank (PIR), the "variable fragment" bank, comprising pairs of closely related proteins, containing one or more strongly differing sites of primary structures was formed. The bank includes 465 "variable fragments" of 383 protein pairs. Amino acid residues composition of "variable fragments" was examined and indexes of potential amino acid residues variability was formed. An analysis of amino acid fragments replaceability was carried out by substituting the N-, C-terminal, or middle part of a chain), the fragments length differences and physico-chemical properties of residues, such as volume, hydrophobicity, polarity, isoelectric point, etc. Some general empirical rules of peptide insertions in carrier-proteins were created based on these analyses. The rules are directed for performing modifications maintaining the common structure and function of the carrier-protein molecule. The selection scheme for determining the regions suitable for modification and the criteria for defining the width of acceptable modifications in this regions were suggested. The use of potential variability profile for detecting regions suitable for peptide insertion was considered on the model of hepatitis B surface protein.

Protein fragment variability analysis and some principles of protein engineering of vaccines.

Based on protein sequence databank (PIR), the 'variable fragment' bank, comprising pairs of closely-related proteins, containing one or more strongly differing sites of primary structures, was formed. The bank includes 465 'variable fragments' in 383 protein pairs. The amino acid composition of 'variable fragments' was examined and indices of potential amino acid residue variability were formed. An analysis of the interchangeability of amino acid fragments depending on the substitution site (N- or C-terminal, or middle part of a chain), the fragment length differences and physico-chemical properties of residues, such as volume, hydrophobicity, polarity and isoelectric point, was carried out. Based on this analysis some general empirical rules of peptide insertions in carrier proteins were created. The rules are directed at performing modifications leaving the general structure and function of the carrier protein molecule unchanged. The selection scheme for the regions suitable for modification and the criteria for determination of the range of acceptable variations in these regions were suggested. The use of the potential variability profile for detecting regions suitable for peptide insertion was considered using surface protein of hepatitis B virus as an example.

[BsiI--a new unusual restriction endonuclease]. / BsiI--novaia neobychnaia éndonukleaza restriktsii.

The restriction endonuclease BsiI from Bacillus sphaericus was isolated. The recognition sequence and cleavage point of enzyme BsiI have been determined as (sequence: see text). This restriction endonuclease is not an isoschizomer of any known restriction endonucleases and differs from other enzymes: it hydrolyses DNA into unsymmetrical recognition sequence.

[Determination of the substrate specificity of Bpu101 restrictase with an unusual recognition segment]. / Opredelenie substratnoi spetsifichnosti restriktazy BPu101 s neobychnym uchastkom uznavaniia.

A new enzyme Bpu10I was isolated from Bacillus pumilus. This enzyme is not an isoschizomer of any known restriction endonucleases. The search of possible recognition sequences was carried out in sequences ABCNiDEF (i = 0.6) on substrate DNA lambda CI857, T7, pBR322. The recognition sequence and cleavage sites of restriction endonuclease Bpu10I have been determined as CCTNAGC. GGANTCG

[Structural and physico-chemical properties of amino acids of functionally important regions of proteins]. / Strukturnye i fiziko-khimicheskie svoistva aminokislot funktional'no vazhnykh uchastkov belkov.

A correlation of the localization of functionally important regions with places having low and high values on twelve profiles built on a basis of amino acid sequences was analysed using a broad set of proteins. The profiles of hydrophilicity, resemblance to the sequences of human proteins, flexibility, mutability and others were considered. The resemblance profile was plotted by the program fixing short similar fragments in the testing protein and 92 human ones. The active centres were shown to be located in the primary structure regions having relatively low values on the resemblance profiles. Similar effect was observed in the mutability and alpha-helicity profiles. The potential functionally important sites of the human leukocyte interferon and interleukin-2 isolated on the basis of the analysis of this profiles were in accord with the available literary data.