We have implemented in Python the COmparative GENomic Toolkit, a fully integrated and thoroughly tested framework for novel probabilistic analyses of biological sequences, devising workflows, and generating publication quality graphics. PyCogent includes connectors to remote databases, built-in generalized probabilistic techniques for working with biological sequences, and controllers for third-party applications. The toolkit takes advantage of parallel architectures and runs on a range of hardware and operating systems, and is available under the general public license from http://sourceforge.net/projects/pycogent. RationaleThe genetic divergence of species is affected by both DNA metabolic processes and natural selection. Processes contributing to genetic variation that are undetectable with intraspecific data may be detectable by inter-specific analyses because of the accumulation of signal over evolutionary time scales. As a consequence of the greater statistical power, there is interest in applying comparative analyses to address an increasing number and diversity of problems, in particular analyses that integrate sequence and phenotype. Significant barriers that hinder the extension of comparative analyses to exploit genome indexed phenotypic data include the narrow focus of most analytical tools, and the diverse array of data sources, formats, and tools available. Theoretically coherent integrative analyses can be conducted by combining probabilistic models of different aspects of genotype. Probabilistic models of sequence change underlie many core bioinformatics tasks, including similarity search, sequence alignment, phylogenetic inference, and ancestral state reconstruction. Probabilistic models allow usage of likelihood inference, a powerful approach from statistics, to establish the significance of differences in support of competing hypotheses. Linking different analyses through a shared and explicit probabilistic model of sequence change is thus extremely valuable, and provides a basis for generalizing analyses to more complex models of evolution (for example, to incorporate dependence between sites). Numerous studies have established how biological factors representing metabolic or selective influences can be represented in substitution models as specific parameters that affect rates of interchange between sequence motifs or the spatial occurrence of such rates [1][2][3][4]. Given this solid grounding, it is desirable to have a toolkit that allows flexible parameterization of probabilistic models and interchange of appropriate modules.There are many existing software packages that can manipulate biological sequences and structures, but few allow specification of both truly novel statistical models and detailed workflow control for genome scale datasets. Traditional phylogenetic analysis applications [5,6] typically provide a number of explicitly defined statistical models that are difficult to modify. One exception in which the parameterization of entirely novel substitution models was poss...

By inserting synthetic oligonucleotides into a highly expressed gene in E. coli it has been shown that unfavourable codon usage can reduce the maximum translation rate of a protein. However, in the case of the codon used (AGG), a significant effect on translation was only seen at very high transcription rates from a gene containing multiple copies of the unfavourable codon.

The syntheses of 1,2-dideoxy-D-ribofuranose and 1,2-dideoxy-1-phenyl-beta-D-ribofuranose are described. Oligodeoxynucleotides containing these analogues have been synthesised and hybridized to their complementary strands. Hypochromicity studies have shown that these duplices are less stable than either the totally complementary duplex or those containing A.C and G.T mismatches.

No abstract

The quantitative and reversible compaction of open circular plasmid DNA (7676 bp) into toroids containing one to 19 molecules by sequential treatment with spermine and an excess of uranyl acetate is reported. The toroidal DNA structure was proven by cryoelectron microscopy. Linearized and supercoiled variants of the DNA also gave toroids under these conditions, but yields were significantly lower. In the presence of spermine alone no toroids were found. Open circular plasmid B-DNA helix was converted into the C-type helical form upon compaction as was shown by CD spectroscopy (negative peak at 255 nm) and electron microscopy (1.8-nm interduplex distance instead of 2.9 nm). Addition of uranyl salt to the DNA−spermine complexes resulted in the formation of netlike assemblies which further compacted to give toroids. A model containing a hexagonal arrangement of DNA strands with extensive strand crossings is proposed. Curvature and thus toroid formation is thought to be induced by the hydrophobic DNA coating of spermine methylene groups.