Evolutionary computation

Evolution does not require DNA, or even living organisms. In computer science, the field known as 'evolutionary computation' uses evolution as an algorithmic tool, implementing random variation, reproduction and selection by altering and moving data within a computer. This harnesses the power of evolution as an alternative to the more traditional ways to design software or hardware. Research into evolutionary computation should be of interest to geneticists, as evolved programs often reveal properties - such as robustness and non-expressed DNA - that are analogous to many biological phenomena.     

Evolutionary grass roots for odor recognition

Considerable evidence supports the idea that odorant recognition depends on specific sequence variations in olfactory receptor (OR) proteins. Much of this emerges from in vitro screens in heterogenous expression systems. However, the ultimate proof should arise from measurements of odorant thresholds in human individuals harboring different OR genetic variants, a research vein that has so far been only scantly explored. The study of McRae et al., published in this issue of Chemical Senses, shows how the recognition of a grassy odorant depends on specific OR interindividual sequence changes. It provides a clear relevant example for the impact of genetics on olfaction and is an excellent portrayal of the power of human genomics to decipher olfactory perception.     

Cortical activity pattern computation

Neural computations are modelled in various ways, but still there is no clear understanding of how the brain performs its computational tasks. This paper presents new results about analysis of neural processes in terms of activity pattern computations. It is shown that it is possible to extract from high-resolution EEG data a first order Markov approximation of a neural communication system employing pattern computations, which is significantly different from similar purely random systems. In our view this result shows that it is likely that neural activity patterns measurable at the macro-level by EEG are correlated with underlying neural computations.     

Automated identification of SUMOylation sites using mass spectrometry and SUMmOn pattern recognition software

Tandem mass spectrometry (MS/MS) allows for the rapid identification of many types of post-translational modifications (PTMs), especially those that can be detected by a diagnostic mass shift in one or more peptide fragment ions (for example, phosphorylation). But some PTMs (for example, SUMOs and other ubiquitin-like modifiers) themselves produce multiple fragment ions; combined with fragments from the modified target peptide, a complex overlapping fragmentation pattern is thus generated, which is uninterpretable by standard peptide sequencing software. Here we introduce SUMmOn, an automated pattern recognition tool that detects diagnostic PTM fragment ion series within complex MS/MS spectra, to identify modified peptides and modification sites within these peptides. Using SUMmOn, we demonstrate for the first time that human SUMO-1 multimerizes in vitro primarily via three N-terminal lysines, Lys7, Lys16 and Lys17. Notably, our method is theoretically applicable to any type of modification or chemical moiety generating a unique fragment ion pattern.     

Multiple cis-acting sites positively regulate Escherichia coli nrd expression

Regulation of nrd expression in Escherichia coli by cis -acting elements was found to be more complex than previously reported. At least five upstream sites appear to positively regulate nrd expression including a Fis binding site, a DnaA binding site, an AT-rich region, an inverted repeat and a 10 bp site between the AT-rich region and the inverted repeat. Double mutants defective in these sites indicate that all sites tested act independently when regulating nrd expression. As the decrease in nrd expression in exponentially growing cultures paralleled the decrease observed in DNA synthesis-inhibited cultures for all single and double mutants, we concluded that nrd is regulated by the same mechanism in these physiological states. As mutants unable to induce nrd expression during inhibition of DNA synthesis also fail to exhibit cell cycle-regulated nrd expression, we conclude that cell cycle nrd regulation is controlled by these same sites. Site-directed mutagenesis was used to show that the absence of an increase in nrd expression during DNA inhibition previously observed for deletion of the AT-rich region results from deletion of both the Fis binding site and the AT-rich region.     

A novel method for automatic genotyping of microsatellite markers based on parametric pattern recognition

Genetic mapping of loci affecting complex phenotypes in human and other organisms is presently being conducted on a very large scale, using either microsatellite or single nucleotide polymorphism (SNP) markers and by partly automated methods. A critical step in this process is the conversion of the instrument output into genotypes, both a time-consuming and error prone procedure. Errors made during this calling of genotypes will dramatically reduce the ability to map the location of loci underlying a phenotype. Accurate methods for automatic genotype calling are therefore important. Here, we describe novel algorithms for automatic calling of microsatellite genotypes using parametric pattern recognition. The analysis of microsatellite data is complicated both by the occurrence of stutter bands, which arise from Taq polymerase misreading the number of repeats, and additional bands derived form the non-template dependent addition of a nucleotide to the 3 end of the PCR products. These problems, together with the fact that the lengths of two alleles in a heterozygous individual may differ by only two nucleotides, complicate the development of an automated process. The novel algorithms markedly reduce the need for manual editing and the frequency of miscalls, and compares very favourably with commercially available software for automatic microsatellite genotyping.     

cis-acting sites required for osmoregulation of ompF expression in Escherichia coli K-12.

OmpF and OmpC are major outer membrane proteins which form passive diffusion pores in Escherichia coli K-12. The expression of the structural genes for these proteins, ompF and ompC, is influenced by medium osmotic strength and requires the products of two regulatory genes, ompR and envZ. We have constructed a series of ompF-lacZ fusions containing different regions of ompF to determine sites involved with osmoregulation. These fusions were crossed onto a specialized transducing phage and integrated into the bacterial chromosome in unit copy. By measuring the fluctuations of beta-galactosidase activity in lysogens grown in high versus low osmolarity, we have identified three regions which are necessary. Furthermore, we have determined that, although the OmpR activation site is not sufficient, OmpR is probably essential for ompF osmoregulation.     

Classification and identification of toxic polypeptides using a pattern recognition method

A crucial rule for both rapid identification and classification of toxins by the relative content of the five most informative amino acids (glycine, cysteine, threonine, alanine and tryptophane) was established by using a computer algorithm of pattern recognition based on amino acid sequences of three families of toxic polypeptides.     

Multiclass Decision Forest--a novel pattern recognition method for multiclass classification in microarray data analysis

The wealth of knowledge imbedded in gene expression data from DNA microarrays portends rapid advances in both research and clinic. Turning the prodigious and noisy data into knowledge is a challenge to the field of bioinformatics, and development of classifiers using supervised learning techniques is the primary methodological approach for clinical application using gene expression data. In this paper, we present a novel classification method, multiclass Decision Forest (DF), that is the direct extension of the two-class DF previously developed in our lab. Central to DF is the synergistic combining of multiple heterogenic but comparable decision trees to reach a more accurate and robust classification model. The computationally inexpensive multiclass DF algorithm integrates gene selection and model development, and thus eliminates the bias of gene preselection in crossvalidation. Importantly, the method provides several statistical means for assessment of prediction accuracy, prediction confidence, and diagnostic capability. We demonstrate the method by application to gene expression data for 83 small round blue-cell tumors (SRBCTs) samples belonging to one of four different classes. Based on 500 runs of 10-fold crossvalidation, tumor prediction accuracy was approximately 97%, sensitivity was approximately 95%, diagnostic sensitivity was approximately 91%, and diagnostic accuracy was approximately 99.5%. Among 25 genes selected to distinguish tumor class, 12 have functional information in the literature implicating their involvement in cancer. The four types of SRBCTs samples are also distinguishable in a clustering analysis based on the expression profiles of these 25 genes. The results demonstrated that the multiclass DF is an effective classification method for analysis of gene expression data for the purpose of molecular diagnostics.     

Dissecting protein-RNA recognition sites

We analyze the protein–RNA interfaces in 81 transient binary complexes taken from the Protein Data Bank. Those with tRNA or duplex RNA are larger than with single-stranded RNA, and comparable in size to protein–DNA interfaces. The protein side bears a strong positive electrostatic potential and resembles protein–DNA interfaces in its amino acid composition. On the RNA side, the phosphate contributes less, and the sugar much more, to the interaction than in protein–DNA complexes. On average, protein–RNA interfaces contain 20 hydrogen bonds, 7 that involve the phosphates, 5 the sugar 2′OH, and 6 the bases, and 32 water molecules. The average H-bond density per unit buried surface area is less with tRNA or single-stranded RNA than with duplex RNA. The atomic packing is also less compact in interfaces with tRNA. On the protein side, the main chain NH and Arg/Lys side chains account for nearly half of all H-bonds to RNA; the main chain CO and side chain acceptor groups, for a quarter. The 2′OH is a major player in protein–RNA recognition, and shape complementarity an important determinant, whereas electrostatics and direct base–protein interactions play a lesser part than in protein–DNA recognition.     

Dissecting protein-protein recognition sites

The recognition sites in 70 pairwise protein-protein complexes of known three-dimensional structure are dissected in a set of surface patches by clustering atoms at the interface. When the interface buries <2000 A2 of protein surface, the recognition sites usually form a single patch on the surface of each component protein. In contrast, larger interfaces are generally multipatch, with at least one pair of patches that are equivalent in size to a single-patch interface. Each recognition site, or patch within a site, contains a core made of buried interface atoms, surrounded by a rim of atoms that remain accessible to solvent in the complex. A simple geometric model reproduces the number and distribution of atoms within a patch. The rim is similar in composition to the rest of the protein surface, but the core has a distinctive amino acid composition, which may help in identifying potential protein recognition sites on single proteins of known structures.     

Routine libraries for pattern recognition in quasispecies

The results obtained through biological research usually need to be analyzed using computational tools, since manual analysis becomes unfeasible due to the complexity and size of these results. For instance, the study of quasispecies frequently demands the analysis of several, very lengthy sequences of nucleotides and amino acids. Therefore, bioinformatics tools for the study of quasispecies are constantly being developed due to different problems found by biologists. In the present study, we address the development of a software tool for the evaluation of population diversity in quasispecies. Special attention is paid to the localization of genome regions prone to changes, as well as of possible hot spots.     

Ensemble classifier for protein fold pattern recognition

MOTIVATION: Prediction of protein folding patterns is one level deeper than that of protein structural classes, and hence is much more complicated and difficult. To deal with such a challenging problem, the ensemble classifier was introduced. It was formed by a set of basic classifiers, with each trained in different parameter systems, such as predicted secondary structure, hydrophobicity, van der Waals volume, polarity, polarizability, as well as different dimensions of pseudo-amino acid composition, which were extracted from a training dataset. The operation engine for the constituent individual classifiers was OET-KNN (optimized evidence-theoretic k-nearest neighbors) rule. Their outcomes were combined through a weighted voting to give a final determination for classifying a query protein. The recognition was to find the true fold among the 27 possible patterns. RESULTS: The overall success rate thus obtained was 62% for a testing dataset where most of the proteins have <25% sequence identity with the proteins used in training the classifier. Such a rate is 6-21% higher than the corresponding rates obtained by various existing NN (neural networks) and SVM (support vector machines) approaches, implying that the ensemble classifier is very promising and might become a useful vehicle in protein science, as well as proteomics and bioinformatics. AVAILABILITY: The ensemble classifier, called PFP-Pred, is available as a web-server at http://202.120.37.186/bioinf/fold/PFP-Pred.htm for public usage.     

Evolutionary perspective on innate immune recognition

Analysis of human and Drosophila genomes demonstrates an ancient origin of innate immunity and the diversity of the mechanisms of innate immune recognition.