NMJ-morph reveals principal components of synaptic morphology influencing structure-function relationships at the neuromuscular junction

The ability to form synapses is one of the fundamental properties required by the mammalian nervous system to generate network connectivity. Structural and functional diversity among synaptic populations is a key hallmark of network diversity, and yet we know comparatively little about the morphological principles that govern variability in the size, shape and strength of synapses. Using the mouse neuromuscular junction (NMJ) as an experimentally accessible model synapse, we report on the development of a robust, standardized methodology to facilitate comparative morphometric analysis of synapses ('NMJ-morph'). We used NMJ-morph to generate baseline morphological reference data for 21 separate pre- and post-synaptic variables from 2160 individual NMJs belonging to nine anatomically distinct populations of synapses, revealing systematic differences in NMJ morphology between defined synaptic populations. Principal components analysis revealed that overall NMJ size and the degree of synaptic fragmentation, alongside pre-synaptic axon diameter, were the most critical parameters in defining synaptic morphology. 'Average' synaptic morphology was remarkably conserved between comparable synapses from the left and right sides of the body. Systematic differences in synaptic morphology predicted corresponding differences in synaptic function that were supported by physiological recordings, confirming the robust relationship between synaptic size and strength.

PeerWise provides significant academic benefits to biological science students across diverse learning tasks, but with minimal instructor intervention

We demonstrate that student engagement with PeerWise, an online tool that allows students to author and answer multiple-choice questions (MCQs), is associated with enhanced academic performance across diverse assessment types on a second year Genetics course. Benefits were consistent over three course deliveries, with differential benefits bestowed on groups of different prior ability. A rating scheme, to assess the educational quality of students' questions, is presented and demonstrates that our students are able intuitively to make such quality assessments, and that the process of authoring high quality questions alone does not explain the academic benefits. We further test the benefits of providing additional PeerWise support and conclude that PeerWise works efficiently with minimal intervention, and can be reliably assessed using automatically generated PeerWise scores.

Functional specialization in nucleotide sugar transporters occurred through differentiation of the gene cluster EamA (DUF6) before the radiation of Viridiplantae

Background

The drug/metabolite transporter superfamily comprises a diversity of protein domain families with multiple functions including transport of nucleotide sugars. Drug/metabolite transporter domains are contained in both solute carrier families 30, 35 and 39 proteins as well as in acyl-malonyl condensing enzyme proteins. In this paper, we present an evolutionary analysis of nucleotide sugar transporters in relation to the entire superfamily of drug/metabolite transporters that considers crucial intra-protein duplication events that have shaped the transporters. We use a method that combines the strengths of hidden Markov models and maximum likelihood to find relationships between drug/metabolite transporter families, and branches within families.

Results

We present evidence that the triose-phosphate transporters, domain unknown function 914, uracil-diphosphate glucose-N-acetylglucosamine, and nucleotide sugar transporter families have evolved from a domain duplication event before the radiation of Viridiplantae in the EamA family (previously called domain unknown function 6). We identify previously unknown branches in the solute carrier 30, 35 and 39 protein families that emerged simultaneously as key physiological developments after the radiation of Viridiplantae, including the "35C/E" branch of EamA, which formed in the lineage of T. adhaerens (Animalia). We identify a second cluster of DMTs, called the domain unknown function 1632 cluster, which has non-cytosolic N- and C-termini, and thus appears to have been formed from a different domain duplication event. We identify a previously uncharacterized motif, G-X(6)-G, which is overrepresented in the fifth transmembrane helix of C-terminal domains. We present evidence that the family called fatty acid elongases are homologous to transporters, not enzymes as had previously been thought.

Conclusions

The nucleotide sugar transporters families were formed through differentiation of the gene cluster EamA (domain unknown function 6) before Viridiplantae, showing for the first time the significance of EamA.

High-resolution mapping of sequence-directed nucleosome positioning on genomic DNA

We have mapped in vitro nucleosome positioning on the sheep beta-lactoglobulin gene using high-throughput sequencing to characterise the DNA sequences recovered from reconstituted nucleosomes. This methodology surpasses previous approaches for coverage, accuracy and resolution and, most importantly, offers a simple yet rapid and relatively inexpensive method to characterise genomic DNA sequences in terms of nucleosome positioning capacity. We demonstrate an unambiguous correspondence between in vitro and in vivo nucleosome positioning around the promoter of the gene; identify discrete, sequence-specific nucleosomal structures above the level of the canonical core particle-a feature that has implications for regulatory protein access and higher-order chromatin packing; and reveal new insights into the involvement of periodically organised dinucleotide sequence motifs of the type GG and CC and not AA and TT, as determinants of nucleosome positioning-an observation that supports the idea that the core histone octamer can exploit different patterns of sequence organisation, or structural potential, in the DNA to bring about nucleosome positioning.

Genome-scale relationships between cytosine methylation and dinucleotide abundances in animals

In mammalian genomes CpGs occur at one-fifth their expected frequency. This is accepted as resulting from cytosine methylation and deamination of 5-methylcytosine leading to TpG and CpA dinucleotides. The corollary that a CpG deficit should correlate with TpG excess has not hitherto been systematically tested at a genomic level. I analyzed genome sequences (human, chimpanzee, mouse, pufferfish, zebrafish, sea squirt, fruitfly, mosquito, and nematode) to do this and generally to assess the hypothesis that CpG deficit, TpG excess, and other data are accountable in terms of 5-methylcytosine mutation. In all methylated genomes local CpG deficit decreases with higher G + C content. Local TpG surplus, while positively associated with G + C level in mammalian genomes but negatively associated with G + C in nonmammalian methylated genomes, is always explicable in terms of the CpG trend under the methylation model. Covariance of dinucleotide abundances with G + C demonstrates that correlation analyses should control for G + C. Doing this reveals a strong negative correlation between local CpG and TpG abundances in methylated genomes, in accord with the methylation hypothesis. CpG deficit also correlates with CpT excess in mammals, which may reflect enhanced cytosine mutation in the context 5'-YCG-3'. Analyses with repeat-masked sequences show that the results are not attributable to repetitive elements.

In silico approaches reveal the potential for DNA sequence-dependent histone octamer affinity to influence chromatin structure in vivo

Nucleosome positioning signals embedded within the DNA sequence have the potential to influence the detailed structure of the higher-order chromatin fibre. In two previous studies of long stretches of DNA, encompassing the chicken beta-globin and ovine beta-lactoglobulin genes, respectively, we mapped the relative affinity of every site for the core histone octamer. In both cases a periodic arrangement of the in vitro positioning sites suggests that they might influence the folding of a nucleosome chain into higher-order structure; this hypothesis was borne out in the case of the beta-lactoglobulin gene, where the distribution of the in vitro positioning sites is related to the positions nucleosomes actually occupy in sheep liver cells. Here, we have exploited the in vitro nucleosome positioning datasets to simulate nucleosomal organisation using in silico approaches. We use the high-resolution, quantitative positioning maps to define a one-dimensional positioning energy lattice, which can be populated with a defined number of nucleosomes. Monte Carlo techniques are employed to simulate the behaviour of the model at equilibrium to produce a set of configurations, which provide a probability-based occupancy map. Employing a variety of techniques we show that the occupancy maps are a sensitive function of the histone octamer density (nucleosome repeat length) and find that a minimal change in this property can produce dramatic localised changes in structure. Although simulations generally give rise to regular periodic nucleosomal arrangements, they often show octamer density-dependent discontinuities, which tend to co-localise with sequences that adopt distinctive chromatin structure in vivo. Furthermore, the overall organisation of simulated chromatin structures are more closely related to the situation in vivo than is the original in vitro positioning data, particularly at a nucleosome density corresponding to the in vivo state. Although our model is simplified, we argue that it provides a unique insight into the influence that DNA sequence can have in determining chromatin structure and could serve as a useful basis for the incorporation of other parameters.

In Vitro and in Vivo nucleosome positioning on the ovine beta-lactoglobulin gene are related

Although positioned nucleosomes are known to play a direct, localised role in regulating access to DNA sequence, they also have the potential, through their long-range distribution, to affect the detailed structure of the higher-order chromatin fibre. To investigate this possibility, we firstly mapped, in vitro, the sequence-dependent positions that the core histone octamer adopts when reconstituted onto DNA containing the ovine beta-lactoglobulin gene. These positioning sites are discussed in terms of their relative affinity for the histone octamer, their locations with respect to the gene sequence and their periodic distribution throughout the gene region. Secondly, we mapped, in vivo, the sites that nucleosomes occupy on the same sequence in liver nuclei, where the gene is transcriptionally inactive. Although the sequence is largely packaged into regularly spaced nucleosomes, reflecting a fibre of uniform higher-order structure, this organisation is disrupted by a number of unusual chromatin structures in a region stretching from the second to the third introns of the gene. A comparison of the in vitro and in vivo nucleosome positioning data shows that they are qualitatively and quantitatively related, suggesting that the structure of the higher-order chromatin fibre containing the beta-lactoglobulin gene is determined, in part, by the long-range organisation of the non-coding sequences within which the gene is embedded.

An ontology of human developmental anatomy

Human developmental anatomy has been organized as structured lists of the major constituent tissues present during each of Carnegie stages 1-20 (E1-E50, approximately 8500 anatomically defined tissue items). For each of these stages, the tissues have been organized as a hierarchy in which an individual tissue is catalogued as part of a larger tissue. Such a formal representation of knowledge is known as an ontology and this anatomical ontology can be used in databases to store, organize and search for data associated with the tissues present at each developmental stage. The anatomical data for compiling these hierarchies comes from the literature, from observations on embryos in the Patten Collection (Ann Arbor, MI, USA) and from comparisons with mouse tissues at similar stages of development. The ontology is available in three versions. The first gives hierarchies of the named tissues present at each Carnegie stage (http://www.ana.ed.ac.uk/anatomy/database/humat/) and is intended to help analyse both normal and abnormal human embryos; it carries hyperlinked notes on some ambiguities in the literature that have been clarified through analysing sectioned material. The second contains many additional subsidiary tissue domains and is intended for handling tissue-associated data (e.g. gene-expression) in a database. This version is available at the humat site and at http://genex.hgu.mrc.ac.uk/Resources/intro.html/), and has been designed to be interoperable with the ontology for mouse developmental anatomy, also available at the genex site. The third gives the second version in GO ontology syntax (with standard IDs for each tissue) and can be downloaded from both the genex and the Open Biological Ontology sites (http://obo.sourceforge.net/).

Nucleosome positioning signals in the DNA sequence of the human and mouse H19 imprinting control regions

We have investigated the sequences of the mouse and human H19 imprinting control regions (ICRs) to see whether they contain nucleosome positioning information pertinent to their function as a methylation-regulated chromatin boundary. Positioning signals were identified by an in vitro approach that employs reconstituted chromatin to comprehensively describe the contribution of the DNA to the most basic, underlying level of chromatin structure. Signals in the DNA sequence of both ICRs directed nucleosomes to flank and encompass the short conserved sequences that constitute the binding sites for the zinc finger protein CTCF, an essential mediator of insulator activity. The repeat structure of the human ICR presented a conserved array of strong positioning signals that would preferentially flank these CTCF binding sites with positioned nucleosomes, a chromatin structure that would tend to maintain their accessibility. Conversely, all four CTCF binding sites in the mouse sequence were located close to the centre of positioning signals that were stronger than those in their flanks; these binding sites might therefore be expected to be more readily incorporated into positioned nucleosomes. We found that CpG methylation did not effect widespread repositioning of nucleosomes on either ICR, indicating that allelic methylation patterns were unlikely to establish allele-specific chromatin structures for H19 by operating directly upon the underlying DNA-histone interactions; instead, epigenetic modulation of ICR chromatin structure is likely to be mediated principally at higher levels of control. DNA methylation did, however, both promote and inhibit nucleosome positioning at several sites in both ICRs and substantially negated one of the strongest nucleosome positioning signals in the human sequence, observations that underline the fact that this epigenetic modification can, nevertheless, directly and decisively modulate core histone-DNA interactions within the nucleosome.

Sequence analysis of transposable elements in the sea squirt, Ciona intestinalis

A systematic search of 1 Mb of genomic sequences from the sea squirt, Ciona intestinalis, revealed the presence of six families of transposable elements. The Cigr-1 retrotransposon contains identical 245-bp long terminal repeats (LTRs) and a 3,630-bp open reading frame (ORF) encoding translation products in the same order as the domains characteristic of gypsy/Ty3-type LTR retrotransposons. The closest homologs of the reverse transcriptase domain were in gypsy elements from Drosophila and the sushi element from the pufferfish. However, the capsid-nucleocapsid region shows the clearest homology to an echinoderm element, Tgr1. Database searches also indicated two classes of non-LTR retrotransposon, named Cili-1 and Cili-2. The Cili-1 sequences show matches to regions of the ORF2 product of mammalian L1 elements. The Cili-2 sequences possess similarity to the RNaseH domain of Lian-Aa1, a mosquito non-LTR retrotransposon. The most abundant element was a short interspersed nucleotide element named Cics-1 with a copy number estimated at 40,000. Cics-1 consists of two conserved domains separated by an A-rich stretch. The 172-bp 5' domain is related to tRNA sequences, whereas the 110-bp 3' domain is unique. Cics-1 is unusual, not just in its modular structure, but also in its lack of a 3' poly(A) tail or direct flanking repeats. A second abundant element, Cimi-1, has an A+T-rich 193-bp consensus sequence and 30-bp terminal inverted repeats (TIRs) and is usually flanked by A+T-rich 2-4-bp putative target site duplications-characteristics of miniature inverted-repeat transposable elements found in plants and insects. A single 2,444-bp foldback element was found, possessing long TIRs containing an A+T-rich internal domain, an array of subrepeats, and a flanking domain at the TIR ends; this is the first example of a chordate foldback element. This study provides the first systematic characterization of the families of transposable elements in a lower chordate.

CpG island libraries from human chromosomes 18 and 22: landmarks for novel genes

CpG islands are found at the 5' end of approximately 60% of human genes and so are important genomic landmarks. They are concentrated in early-replicating, highly acetylated gene-rich regions. With respect to CpG island content, human Chrs 18 and 22 are very different from each other: Chr 18 appears to be CpG island poor, whereas Chr 22 appears to be CpG island rich. We have constructed and validated CpG island libraries from flow-sorted Chrs 18 and 22 and used these to estimate the difference in number of CpG islands found on these two chromosomes. These libraries contain normalized collections of sequences from the 5' end of genes. Clones from the libraries were sequenced and compared with the sequence databases; one third matched ESTs, thus anchoring these ESTs at the 5' end of their gene. However, it was striking that many clones either had no match or matched only existing CpG island clones. This suggests that a significant proportion of 5' gene sequences are absent from databases, presumably either because they are difficult to clone or the gene is poorly expressed and/or has a restricted expression pattern. This point should be taken into consideration if the currently available libraries are those used for the elucidation of complete, as opposed to partial, gene sequences. The Chr 18 and 22 CpG island libraries are a sequence resource for the isolation of such 5' gene sequences from specific human chromosomes.

Nonmethylated transposable elements and methylated genes in a chordate genome

The genome of the invertebrate chordate Ciona intestinalis was found to be a stable mosaic of methylated and nonmethylated domains. Multiple copies of an apparently active long terminal repeat retrotransposon and a long interspersed element are nonmethylated and a large fraction of abundant short interspersed elements are also methylation free. Genes, by contrast, are predominantly methylated. These data are incompatible with the genome defense model, which proposes that DNA methylation in animals is primarily targeted to endogenous transposable elements. Cytosine methylation in this urochordate may be preferentially directed to genes.

Gene number in an invertebrate chordate, Ciona intestinalis

Gene number can be considered a pragmatic measure of biological complexity, but reliable data is scarce. Estimates for vertebrates are 50-100,000 genes per haploid genome, whereas invertebrate estimates fall below 25,000. We wished to test the hypothesis that the origin of vertebrates coincided with extensive gene creation. A prediction is that gene number will differ sharply between invertebrate and vertebrate members of the chordate phylum. A gene number estimation method requiring limited sequence sampling of genomic DNA was developed and validated by using data for Caenorhabditis elegans. Using the method, we estimated that the invertebrate chordate Ciona intestinalis has 15,500 protein-coding genes (+/-3,700). This number is significantly lower than gene numbers of vertebrate chordates, but similar to those of invertebrates in distantly related phyla. The data indicate that evolution of vertebrates was accompanied by a dramatic increase in protein-coding capacity of the genome.

Pattern retrieval in threshold-linear associative nets

Networks of threshold-linear neurons have previously been introduced and analysed as distributed associative memory systems. Here, results from simulations of pattern retrieval in a large-scale, sparsely connected network are presented. The storage capacity lies near a = 0.8 and 1.2 for binary and ternary patterns respectively, in reasonable accordance with theoretical estimates. The system is capable of retrieving states strongly correlated with one of the stored patterns even when the initial state is a highly degraded version of one of these patterns. This pattern completion ability holds for an extensive number of memory patterns, up to α ≈ α_c/2, thereby increasing the credibility of the model as an effective associative memory.

Multidimensional Scaling of Binary Dissimilarities: Direct and Derived Approaches

Given a matrix of dissimilarities, it has been debated whether researchers should perform multidimensional scaling on this original matrix or on a new one derived by comparing rows in the original matrix. Careful comparison studies (Drasgow & Jones, 1979; Van der Kloot & Van Herk, 1991) in the context of sorting data indicated that most of the initial enthusiasm for the derivative approach was unfounded. The current work, a Monte Carlo study of structured binary data derived from known two-dimensional configurations using ALSCAL, complements and extends the previous studies. We discuss a weakness in the squared difference (δ) row-comparison rule used previously and propose an alternative row-comparison measure based on the Jaccard coefficient. Scaling the binary data directly gave better performance, as gauged by Procrustes statistics, than did scaling A data across a range of noise levels. The quality of solutions obtained by scaling Jaccard data was always better or essentially equal to that from scaling δ data, and in certain parameter regions improved upon that of direct scaling. Another alternative approach, applying the δ rule after first row-centering the binary data, was found to be generally ineffective. These findings are pertinent to the analysis not just of stimulus sorting data but of coarse dissimilarities generally, for example from direct pairwise judgment tasks and in fields outside statistical psychology.

An evaluation of the use of multidimensional scaling for understanding brain connectivity

A large amount of data is now available about the pattern of connections between brain regions. Computational methods are increasingly relevant for uncovering structure in such datasets. There has been recent interest in the use of non-metric multidimensional scaling (NMDS) for such analysis. NMDS produces a spatial representation of the 'dissimilarities' between a number of entities. Normally, it is applied to data matrices containing a large number of levels of dissimilarity, whereas for brain connectivity data there is a very small number. We address the suitability of NMDS for this case. Systematic numerical studies are presented to evaluate the ability of this method to reconstruct known geometrical configurations from dissimilarity data processing few levels. In this case there is a strong bias for NMDS to produce annular configurations, whether or not such structure exists in the original data. For the case of a connectivity dataset derived from the primate cortical visual system, we demonstrate that great caution is needed in interpreting the resulting configuration. Application of an independent method that we developed also strongly suggests that the visual system NMDS configuration is affected by an annular bias. We question the strength of support that an NMDS analysis of the visual system data provides for the two streams view of visual processing.

Parameter Sensitivity of the Elastic Net Approach to the Traveling Salesman Problem

Durbin and Willshaw's elastic net algorithm can find good solutions to the TSP. The purpose of this paper is to point out that for certain ranges of parameter values, the algorithm converges into local minima that do not correspond to valid tours. The key parameter is the ratio governing the relative strengths of the two competing terms in the elastic net energy function. Based on recent work by Durbin, Szeliski and Yuille, the parameter regime in which the net may visit some cities twice is examined. Further analysis predicts the regime in which the net may fail to visit some cities at all. Understanding these limitations allows one to select the parameter value most likely to avoid either type of problem. Simulation data support the theoretical work.