Neural coding of categories: information efficiency and optimal population codes

This paper deals with the analytical study of coding a discrete set of categories by a large assembly of neurons. We consider population coding schemes, which can also be seen as instances of exemplar models proposed in the literature to account for phenomena in the psychophysics of categorization. We quantify the coding efficiency by the mutual information between the set of categories and the neural code, and we characterize the properties of the most efficient codes, considering different regimes corresponding essentially to different signal-to-noise ratio. One main outcome is to find that, in a high signal-to-noise ratio limit, the Fisher information at the population level should be the greatest between categories, which is achieved by having many cells with the stimulus-discriminating parts (steepest slope) of their tuning curves placed in the transition regions between categories in stimulus space. We show that these properties are in good agreement with both psychophysical data and with the neurophysiology of the inferotemporal cortex in the monkey, a cortex area known to be specifically involved in classification tasks.     

Identification and distribution of sequences having similarity to mitochondrial plasmids in mitochondrial genomes of filamentous fungi

Mitochondrial plasmids are autonomously replicating genetic elements commonly associated with fungal and plant species. Analysis of several plant and fungal mitochondrial genomes has revealed regions that show significant homology to mitochondrial plasmids, suggesting that plasmids have had a long-term association with their mitochondrial hosts. To assess the degree to which plasmids have invaded fungal mitochondrial genomes, BLAST search parameters were modified to identify plasmid sequences within highly AT-rich mtDNAs, and output data were parsed by E value, score, and sequence complexity. High scoring hits were evaluated for the presence of shared repetitive elements and location within plasmids and mtDNAs. Our searches revealed multiple sites of sequence similarity to four distinct plasmids in the wild-type mtDNA of Neurospora crassa, which collectively comprise more than 2% of the mitochondrial genome. Regions of plasmid similarity were not restricted to plasmids known to be associated with senescence, indicating that all mt plasmids can potentially integrate into mitochondrial DNA. Unexpectedly, plasmid-related sequences were found to be clustered in regions that have disproportionately low numbers of PstI palindromic sequences, suggesting that these repetitive elements may play a role in eliminating foreign DNA. A separate class of GC-rich palindromes was identified that appear to be mobile, as indicated by their occurrence within regions of plasmid homology. Sites of sequence similarity to mitochondrial plasmids were also detected in other filamentous fungi, but to a lesser degree. The tools developed here will be useful in assessing the contribution plasmids have made to mitochondrial function and in understanding the co-evolution of mitochondrial plasmids and their hosts.Electronic Supplementary Material Supplementary material is available for this article at     

An automatic method involving cluster analysis of secondary structures for the identification of domains in proteins.

With a growing number of structures available in the Brookhaven Protein Data Bank, automatic methods for domain identification are required for the construction of databases. Domains are considered to be clusters of secondary structure elements. Thus, helices and strands are first clustered using intersecondary structural distances between C alpha positions, and dendrograms based on this distance measure are used to identify domains. Individual domains are recognized by a disjoint factor, which enables the automatic identification and classification into disjoint, interacting, and conjoint domains. Application to a database of 83 protein families and 18 unique structures shows that the approach provides an effective delineation of boundaries and identifies those proteins that can be considered as a single domain. A quantitative estimate of the interaction between domains has been proposed. The database of protein domains is a useful tool for understanding protein folding, for recognizing protein folds, and for understanding structure-activity relationships.     

A morphological classification of plants,with special reference to the New Zealand alpine flora

Abstract. A plant is a complex of integrated systems (leaves, leaf groups, stems, roots, inflorescences), coexisting side by side or superimposed on each other to produce the life form (orphysiognomy) of the individual. The classic Raunkiaer classification based fundamentally on one character (apex position) is insufficient for the purpose of a functional classification. Five keys are given to determine the categories of:

• 1) the plant silhouette or general shape resulting from a combination of other systems, with 11 categories proposed;
• 2) the leaf group, with 14 categories;
• 3) the stem, with 27 categories;
• 4) the root, with 5 categories; and
• 5) the inflorescence, with 3 categories.

Each plant can be named according to the category or model of each of the five different systems that they most resemble, or by using only the name(s) of systems which are more conspicuous than others. Characters are selected primarily for their influence on form and secondarily on size. This scheme allows for detailed studies of a flora in terms of morphological characteristics (alone or in systems), expressed as frequency of occurrence of each character in the flora. Characters can be analysed separately (e.g. entire margined leaves), as a combination of characters (e.g. leaf groups) or as a combination of systems (e.g. rosettes without stems). Thus correlations between environmental variables and plants canbe made with more precision than in previous classification schemes. The classification also serves as the framework for including additional morphological data and incorporating new models. The New Zealand alpine flora is used as a test case and to exemplify the classification.     

The Urfold: Structural similarity just above the superfold level?

We suspect that there is a level of granularity of protein structure intermediate between the classical levels of “architecture” and “topology,” as reflected in such phenomena as extensive three‐dimensional structural similarity above the level of (super)folds. Here, we examine this notion of architectural identity despite topological variability, starting with a concept that we call the “Urfold.” We believe that this model could offer a new conceptual approach for protein structural analysis and classification: indeed, the Urfold concept may help reconcile various phenomena that have been frequently recognized or debated for years, such as the precise meaning of “significant” structural overlap and the degree of continuity of fold space. More broadly, the role of structural similarity in sequence?structure?function evolution has been studied via many models over the years; by addressing a conceptual gap that we believe exists between the architecture and topology levels of structural classification schemes, the Urfold eventually may help synthesize these models into a generalized, consistent framework. Here, we begin by qualitatively introducing the concept.     

Stowaway MITEs in Hordeum (Poaceae): evolutionary history, ancestral elements and classification

Given the lack of direct observational data relating to transposition of Stowaway miniature inverted repeat transposable elements, phylogenetic methods may provide a means of generating data that adds to our knowledge of these elements. In a phylogenetic framework the evolutionary history of homologous elements may be traced, and the nucleotide sequence of elements at or close to the time of insertion can be reconstructed. Based on a phylogeny of the diploid species of the genus Hordeum we explore evolutionary aspects of four non-homologous groups of Stowaway elements inserted into three nuclear genes: nucellin, xylose isomerase, and barley leucine zipper 1. The data illustrate how elements starting from a high degree of sequence similarity between terminal inverted repeat regions gradually degrade, and confirm previous notions about preferential insertion at particular TA target sites. It is shown how creation of consensus sequences as estimates of ancestral elements may be positively misleading. The Stowaway family of transposable elements is often further divided into subfamilies based on sequence similarity between elements. Sequence similarity data from the elements discovered in the xylose isomerase gene, and other elements found through BLAST searches in GenBank, reveal inconsistency in the rules used for classification. In order to reflect natural groups, a classification of transposable elements must be based on phylogenetic evidence rather than raw similarity.  © The Willi Hennig Society 2009.     

Function annotation for pseudoknot using structure similarity

Many raw biological sequence data have been generated by the human genome project and related efforts. The understanding of structural information encoded by biological sequences is important to acquire knowledge of their biochemical functions but remains a fundamental challenge. Recent interest in RNA regulation has resulted in a rapid growth of deposited RNA secondary structures in varied databases. However, a functional classification and characterization of the RNA structure have only been partially addressed. This article aims to introduce a novel interval-based distance metric for structure-based RNA function assignment. The characterization of RNA structures relies on distance vectors learned from a collection of predicted structures. The distance measure considers the intersected, disjoint, and inclusion between intervals. A set of RNA pseudoknotted structures with known function are applied and the function of the query structure is determined by measuring structure similarity. This not only offers sequence distance criteria to measure the similarity of secondary structures but also aids the functional classification of RNA structures with pesudoknots.     

Towards a structural classification of phosphate binding sites in protein-nucleotide complexes: an automated all-against-all structural comparison using geometric matching

A method is described for the rapid comparison of protein binding sites using geometric matching to detect similar three-dimensional structure. The geometric matching detects common atomic features through identification of the maximum common sub-graph or clique. These features are not necessarily evident from sequence or from global structural similarity giving additional insight into molecular recognition not evident from current sequence or structural classification schemes. Here we use the method to produce an all-against-all comparison of phosphate binding sites in a number of different nucleotide phosphate-binding proteins. The similarity search is combined with clustering of similar sites to allow a preliminary structural classification. Clustering by site similarity produces a classification of binding sites for the 476 representative local environments producing ten main clusters representing half of the representative environments. The similarities make sense in terms of both structural and functional classification schemes. The ten main clusters represent a very limited number of unique structural binding motifs for phosphate. These are the structural P-loop, di-nucleotide binding motif [FAD/NAD(P)-binding and Rossman-like fold] and FAD-binding motif. Similar classification schemes for nucleotide binding proteins have also been arrived at independently by others using different methods.     

Threshold-mediated mechanisms in mutagenesis: implications in the classification and regulation of chemical mutagens

Chemical mutagens are currently regulated and labelled on the basis of their hazardous properties defined in hazard classification schemes. The strength and type of experimental evidence is used as the only criterion for classification in categories which express different levels of concern for the possibility of adverse effects - notably transmissible genetic alterations - in humans. Differently from the classification of carcinogens, no consideration is given to potency, nor to the mechanism of action. The rationale of such hazard based classification is that the hazardous property of a chemical is an intrinsic feature, which is expressed independently of dosing. Changing of dose level results in a mere change in the probability to observe an adverse effect, but not in its potential occurrence. The lack of theoretical threshold underlying this approach can be envisaged, in principle, for stochastic processes such as DNA damage, which can be triggered by single molecular interactions. On the other hand, indirect mechanisms of genotoxicity, involving multiple interactions with non-DNA targets, are expected to show a threshold. At variance to DNA reactive agents, chemicals acting with threshold-mediated mechanism do change also qualitatively their toxic properties depending on the dose level. Possible problems arising in the application of hazard based schemes for the evaluation of chemicals with threshold-mediated mechanism of action are discussed, using the spindle poisons benzimidazole fungicides as an example.     

Evidence for a relationship between the Bombyx mori middle repetitive Bm1 sequence family and U1 snRNA

Several genomic library equivalents of Bombyx mori were constructed in the EMBL-4 lambda derivative. The genomic bank was screened with purified Bombyx mori U1 RNA and twenty positive clones for the U1 gene were isolated. Three U1-related sequences were subcloned and sequenced. Two of the sequences are U1 pseudogenes while a third sequence represents a member of the Bm1 family of repetitive elements of B.mori with significant sequence similarity to U1 small nuclear RNA. The U1-related Bm1 element exhibits 82% sequence similarity with the Bm1 consensus sequence and, under less stringent computer comparison parameters, 60% similarity with a composite B.mori/Drosophila melanogaster U1 gene. The Bm1 family consensus sequence exhibits 53% sequence similarity with the composite U1 gene. The two pseudogenes possess highly conserved sequences with the B.mori U1 gene only for the first 101 nucleotides. These findings are indicative of at least two different categories of U1-related sequences in B.mori, one with a possible evolutionary relationship to the Bm1 family of repetitive elements and the other representing characteristic processed pseudogenes with retroposon mode of dispersion and target selection for the TTTA hotspot. In addition, the U1-related Bm1 element may demonstrate for the first time that a family of retroposons is ultimately derived from a U snRNA.This article is dedicated in memory of Ms. Deborah Lampert who helped so much in the preparation of this paper.     

PCA disjoint models for multiclass cancer analysis using gene expression data

MOTIVATION: Microarray expression profiling appears particularly promising for a deeper understanding of cancer biology and to identify molecular signatures supporting the histological classification schemes of neoplastic specimens. However, molecular diagnostics based on microarray data presents major challenges due to the overwhelming number of variables and the complex, multiclass nature of tumor samples. Thus, the development of marker selection methods, that allow the identification of those genes that are most likely to confer high classification accuracy of multiple tumor types, and of multiclass classification schemes is of paramount importance. RESULTS: A computational procedure for marker identification and for classification of multiclass gene expression data through the application of disjoint principal component models is described. The identified features represent a rational and dimensionally reduced base for understanding the basic biology of diseases, defining targets for therapeutic intervention, and developing diagnostic tools for the identification and classification of multiple pathological states. The method has been tested on different microarray data sets obtained from various human tumor samples. The results demonstrate that this procedure allows the identification of specific phenotype markers and can classify previously unseen instances in the presence of multiple classes.     

Phylogenetic relationships between cultivated and wild species of the genusBeta revealed by DNA “fingerprinting”

Forty-one accessions of the genusBeta representing wild and cultivated species of all sections were analyzed by DNA fingerprinting. Four sugar beet minisatellite DNA probes revealed characteristic banding patterns with Southern-hybridizedBeta DNA restricted withHindIII. A total of 111 polymorphic RFLP bands were scored across all accessions. Cluster analysis based on genetic similarity estimates for all 820 combinations of accessions revealed the following results. (1) All accessions could unambiguously be identified by a characteristic RFLP banding pattern. (2) The sugar beet cultivars examined displayed a low level of genetic diversity; they showed high similarity toB. Vulgaris ssp.maritima but low genetic similarity to the other wild species of section I. (3) In most cases, the present taxonomic classification of the genusBeta was confirmed. Species of sections II, III, and IV were clearly distinguishable from those of section I except forB. Macrocarpa, which showed high similarity to wild species of section II. In a second experiment, 108 single-copy RFLP probes from sugar beet were Southern hybridized withB. procumbens DNA. A surprisingly low degree of homology (34%) was found. The results are discussed with regard to the taxonomic classification of the genusBeta.     

On the necessity of an archetypal concept in morphology: With special reference to the concepts of “structure” and “homology”

Morphological elements, or structures, are sorted into four categories depending on their level of anatomical isolation and the presence or absence of intrinsically identifying characteristics. These four categories are used to highlight the difficulties with the concept of structure and our ability to identify or define structures. The analysis is extended to the concept of homology through a discussion of the methodological and philosophical problems of the current concept of homology. It is argued that homology is fundamentally a similarity based concept rather than a phylogenetic concept, and a proposal is put forth to return to a comparative context for homology. It is shown that for both the concepts of structure and homology ana priori assumption of stable underlying patterns (i.e. archetypes) is essential.     

Quantifying spatial classification uncertainties of the historical Wisconsin landscape (USA)

Landscape feature can be classified by creating categories based on aggregation of spatially explicit information. However, many landscape features appear continuous rather than discrete. The aggregation process likely involves loss of information and introduces a variety of uncertainties whose degree and extent may differ spatially. Since landscape classifications have found wide application in e.g. natural resource policies or ecological research, assessments of spatial classification uncertainties are required.
We present a quantitative framework to identify the degree of landscape continuity (fuzziness) and structure (categorization) based on fuzzy classification and offer measures to quantify uncertainties originating from aggregating features into categories. Fuzzy classification is a non-hierarchical, quantitative method of assessing class definitions using degrees of association between features and class. This results in classes which are well defined and compositionally distinct, as well as classes which are less clearly defined but which, to various degrees, share characteristics with some or all classes. The spatial variation in the degree of class definition on the landscape is used to assess classification uncertainties. The two aspects of uncertainty investigated are the degree of association of a feature with the overall class definitions (membership diffusion), and the class-specific degree of association of each pixel on the landscape with each class (membership saturation).
Three classification scenarios, one fuzzy and one discrete, of the historical landscape of Wisconsin (USA) were compared for spatial classification uncertainties. Membership diffusion is highest in topographically heterogeneous environments, or areas characterized by many species occupying similar ecological niches. Classification uncertainties for individual classes show that differentiated species distributions can be identified, not only distribution centers.     

<Emphasis Type="Italic">hATpin</Emphasis>, a Family of MITE-like <Emphasis Type="Italic">hAT</Emphasis> Mobile Elements Conserved in Diverse Plant Species that Forms Highly Stable Secondary Structures

We identified a 178 bp mobile DNA element in lettuce with characteristic CGAGC/GCTCG repeats in the subterminal regions. This element has terminal inverted repeats and 8-bp target site duplications typical of the hAT superfamily of class II mobile elements, but its small size and potential to form a single-stranded stable hairpin-like secondary structure suggest that it is related to MITE elements. In silico searches for related elements identified 252 plant sequences with 8-bp target site duplications and sequence similarity in their terminal and subterminal regions. Some of these sequences were predicted to encode transposases and may be autonomous elements; these constituted a separate clade within the phylogram of hAT transposases. We demonstrate that the CGAGC/GCTCG pentamer maximizes the hairpin stability compared to any other pentamer with the same C + G content, and the secondary structures of these elements are more stable than for most MITEs. We named these elements collectively as hATpin elements because of the hAT similarity and their hairpin structures. The nearly complete rice genome sequence and the highly advanced genome annotation allowed us to localize most rice elements and to deduce insertion preferences. hATpin elements are distributed on all chromosomes, but with significant bias for chromosomes 1 and 10 and in regions of moderate gene density. This family of class II mobile elements is found primarily in monocot species, but is also present in dicot species. Electronic supplementary material Electronic supplementary material is available for this article at and accessible for authorised users.     

Associated LAH Numbers,Factorial Series Distributions and Unbiased Estimation of a Size Parameter

A finite population consists of kN individuals of N different categories with k individuals each. It is required to estimate the unknown parameter N, the number of different classes in the population. A sequential sampling scheme is considered in which individuals are sampled until a preassigned number of repetitions of already observed categories occur in the sample. Corresponding fixed sample size schemes were considered by Charalambides (1981). The sequential sampling scheme has the advantage of always allowing unbiased estimation of the size parameter N. It is shown that relative to Charalambides' fixed sample size scheme only minor adjustments are required to account for the sequential scheme. In particular, MVU estimators of parametric functions are expressible in terms of the C-numbers introduced by Charalambides.     

<Emphasis Type="Italic">Pearl</Emphasis>, a Novel Family of Putative Transposable Elements in Bivalve Mollusks

Abstract While genome sequencing projects have discovered numerous types of transposable elements in diverse eukaryotes, there are many taxa of ecological and evolutionary significance that have received little attention, such as the molluscan class Bivalvia. Examination of a 0.7-MB genomic sequence database from the cupped oyster Crassostrea virginica revealed the presence of a common interspersed element, CvA. CvA possesses subterminal inverted repeats, a tandemly repeated core element, a tetranucleotide microsatellite region, and the ability to form stable secondary structures. Three other less abundant repetitive elements with a similar structure but little sequence similarity were also found in C. virginica. Ana-1, a repetitive element with similar features, was discovered in the blood ark Anadara trapezia by probing a genomic library with a dimeric repeat element contained in intron 2 of a minor globin gene in that species. All of these elements are flanked by the dinucleotide AA, a putative target-site duplication. They exhibit structural similarity to the sea urchin Tsp family and Drosophila SGM insertion sequences; in addition, they possess regions of sequence similarity to satellite DNA from several bivalve species. We suggest that the Crassostrea repetitive elements and Ana-1 are members of a new MITE-like family of nonautonomous transposable elements, named pearl. Pearl is the first putative nonautonomous DNA transposon to be identified in the phylum Mollusca.     

Classification of doubly wound nucleotide binding topologies using automated loop searches

A classification is presented of doubly wound α/β nucleotide binding topologies, whose binding sites are located in the cleft formed by a topological switch point. In particular, the switch point loop nearest the N-terminus is used to identify specific structural classes of binding protein. This yields seven structurally distinct loop conformations, which are subsequently used as motifs for scanning the Protein Data Bank. The searches, which are effective at identifying functional relationships within a large database of structures, reveal a remarkable and previously unnoticed similarity between the coenzyme binding sites of flavodoxin and tryptophan synthetase, even though there is no sequence or topological similarity between them.     

PCBOST: Protein classification based on structural trees

In this paper, we present the protein classification based on structural trees (PCBOST). This is a novel hierarchical classification of proteins that is primarily based on similarity of overall folds of proteins as well as on the modeled folding pathways of proteins. Amino acid sequences, functions of proteins and their evolutionary relationship are not taken into account in this classification. To date the database includes 3847 proteins and domains grouped into six categories having structural similarity and forming six structural trees (total 10,547 PDB-entries). The work on extension of the database and construction of novel structural trees is in progress. The service is free for all users and available at the URL <http://strees.protres.ru/>.     

Assessing annotation transfer for genomics: quantifying the relations between protein sequence, structure and function through traditional and probabilistic scores

Measuring in a quantitative, statistical sense the degree to which structural and functional information can be "transferred" between pairs of related protein sequences at various levels of similarity is an essential prerequisite for robust genome annotation. To this end, we performed pairwise sequence, structure and function comparisons on approximately 30,000 pairs of protein domains with known structure and function. Our domain pairs, which are constructed according to the SCOP fold classification, range in similarity from just sharing a fold, to being nearly identical. Our results show that traditional scores for sequence and structure similarity have the same basic exponential relationship as observed previously, with structural divergence, measured in RMS, being exponentially related to sequence divergence, measured in percent identity. However, as the scale of our survey is much larger than any previous investigations, our results have greater statistical weight and precision. We have been able to express the relationship of sequence and structure similarity using more "modern scores," such as Smith-Waterman alignment scores and probabilistic P-values for both sequence and structure comparison. These modern scores address some of the problems with traditional scores, such as determining a conserved core and correcting for length dependency; they enable us to phrase the sequence-structure relationship in more precise and accurate terms. We found that the basic exponential sequence-structure relationship is very general: the same essential relationship is found in the different secondary-structure classes and is evident in all the scoring schemes. To relate function to sequence and structure we assigned various levels of functional similarity to the domain pairs, based on a simple functional classification scheme. This scheme was constructed by combining and augmenting annotations in the enzyme and fly functional classifications and comparing subsets of these to the Escherichia coli and yeast classifications. We found sigmoidal relationships between similarity in function and sequence, with clear thresholds for different levels of functional conservation. For pairs of domains that share the same fold, precise function appears to be conserved down to approximately 40 % sequence identity, whereas broad functional class is conserved to approximately 25 %. Interestingly, percent identity is more effective at quantifying functional conservation than the more modern scores (e.g. P-values). Results of all the pairwise comparisons and our combined functional classification scheme for protein structures can be accessed from a web database at http://bioinfo.mbb.yale.edu/alignCopyright 2000 Academic Press.