International Journal of Computational Biology and Drug Design最新文献

The chemical universe of molecules reachable from a set of start compounds by iterative application of a finite number of reactions is usually so vast, that sophisticated and efficient exploration strategies are required to cope with the combinatorial complexity. A stringent analysis of (bio)chemical reaction networks, as approximations of these complex chemical spaces, forms the foundation for the understanding of functional relations in Chemistry and Biology. Graphs and graph rewriting are natural models for molecules and reactions. Borrowing the idea of partial evaluation from functional programming, we introduce partial applications of rewrite rules. A framework for the specification of exploration strategies in graph-rewriting systems is presented. Using key examples of complex reaction networks from carbohydrate chemistry we demonstrate the feasibility of this high-level strategy framework. While being designed for chemical applications, the framework can also be used to emulate higher-level transformation models such as illustrated in a small puzzle game.

MicroRNAs are small (approx. 22nt) non-coding RNAs that regulate the expression of genes by either degrading messenger-RNA (mRNA) that has already been transcribed or by repressing the translation of mRNA, thus inhibiting protein production. This mechanism of gene regulation by binding of the miRNA to 3-prime-untranslated region of target mRNAs has been recently discovered. This sequence-specific post-transcriptional gene regulation process affects large set of genes involved in number of biological pathways. Mapping of 7nt long miRNA seed sequence to the target gene has been a standard way of predicting miRNA targets. In this study, we develop a framework to enrich the human miRNA-mRNA relationship based on genomic and structural information.

Bayesian-like operational taxonomic unit examiner (BOTUX) is a new tool for the classification of 16S rRNA gene sequences into operational taxonomic units (OTUs) that addresses the problem of overestimation caused by errors introduced during PCR amplification and DNA sequencing steps. BOTUX utilises a grammar-based assignment strategy, where Bayesian models are built from each word of a given length (e.g., 8-mers). de novo analysis is possible with BOTUX as it does not require a training set, and updates probabilistic models as new sequences are recruited to an OTU. In benchmarking tests performed with real and simulated datasets of 16S rDNA sequences, BOTUX accurately identifies OTUs with comparable or better clustering efficiency and lower execution times than other OTU algorithms tested. BOTUX is the only OTU classifier, which allows incremental analysis of large datasets, and is also adept in clustering both 454 and Illumina datasets in a reasonable timeframe.

This paper presents a method of separating cells that are connected to each other forming clusters. The difference to many other publications covering similar topics is that the cell types we are dealing with form clusters of highly varying morphology. An advantage of our method is that it can be universally used for different cell types. The segmentation method is based on a growth simulation starting from the nuclei areas. To start the evaluation, the cells need to be made visible with a histological stain, in our case with the May-Grünwald solution. After the staining process has been completed, the nuclei areas can be distinguished from the other cell areas by a histogram backprojection algorithm. The presented method can, in addition to histological stained cells, also be applied to fluorescent-stained cells.

The GC-rich bacterial species, Aeromonas salmonicida, is parasitised by both GC-rich phages (Aeromonas phages - phiAS7 and vB_AsaM-56) and GC-poor phages (Aeromonas phages - 25, 31, 44RR2.8t, 65, Aes508, phiAS4 and phiAS5). Both the GC-rich Aeromonas phage phiAS7 and Aeromonas phage vB_AsaM-56 have nearly identical codon usage bias as their host. While all the remaining seven GC-poor Aeromonas phages differ dramatically in codon usage from their GC-rich host. Here, we investigated whether tRNA encoded in the genome of Aeromonas phages facilitate the translation of phage proteins. We found that tRNAs encoded in the phage genome correspond to synonymous codons overused in the phage genes but not in the host genes.

Availability of complete human genome is a crucial factor for genetic studies to explore possible association between the genome and complex diseases. Haplotype, as a set of single nucleotide polymorphisms (SNPs) on a single chromosome, is believed to contain promising data for disease association studies, detecting natural positive selection and recombination hotspots. Various computational methods for haplotype reconstruction from aligned fragment of SNPs have already been proposed. This study presents a novel approach to obtain paternal and maternal haplotypes form the SNP fragments on minimum error correction (MEC) model. Reconstructing haplotypes in MEC model is an NP-hard problem. Therefore, our proposed methods employ two fast and accurate clustering techniques as the core of their procedure to efficiently solve this ill-defined problem. The assessment of our approaches, compared to conventional methods, on two real benchmark datasets, i.e., ACE and DALY, proves the efficiency and accuracy.

Augmented path eccentric connectivity topochemical indices (reported in part-1 of the manuscript) along with 42 diverse non-correlating molecular descriptors (shortlisted from a large pool of 2D and 3D MDs) were successfully utilised for the development of models through decision tree, random forest and moving average analysis for the prediction of antitubercular activity of aza and diazabiphenyl analogues of active compound (6S)-2-Nitro-{[4-(trifluoromethoxy)benzyl]oxy}-6,7-dihydro-5H-imidazo[2,1-b][1,3] oxazine (PA-824). The statistical significance of the proposed models was assessed through overall accuracy of prediction, intercorrelation analysis, sensitivity, specificity and Matthew's correlation coefficient (MCC). The accuracy of prediction of the proposed models varied from a minimum of 81% to a maximum of ∼99%. High accuracy of prediction amalgamated with high MCC values clearly indicates robustness of the proposed models. The said models offer a vast potential for providing lead structures for the development of potent antitubercular drugs.

The heat shock response is a general mechanism by which organisms deal with physical insults such as sudden changes in temperature, osmotic and oxidative stresses, and exposure to toxic substances. Plasmodium falciparum is exposed to drastic temperature changes as a part of its life cycle and maintains an extensive repertoire of heat shock response-related proteins. As these proteins serve to maintain the parasite in the face of anti-malarial drugs as well, better understanding of the heat shock-related systems in the malaria parasite will lead to therapeutic approaches that frustrate these systems, leading to more effective use of anti-malarials. Here we use protein association networks to broaden our understanding of the systems impacted by and/or implicated in the heat shock response.

Topological analysis of protein-protein interaction (PPI) networks has been widely applied to the investigation on cancer mechanisms. However, there is still a debate on whether cancer proteins exhibit more topological centrality compared to the other proteins in the human PPI network. To resolve this debate, we first identified four sets of human proteins, and then mapped these proteins into the yeast PPI network by homologous genes. Finally, we compared these proteins' properties in human and yeast PPI networks. Experiments over two real datasets demonstrated that cancer proteins tend to have higher degree and smaller clustering coefficient than non-cancer proteins. Experimental results also validated that cancer proteins have larger betweenness centrality compared to the other proteins on the STRING dataset. However, on the BioGRID dataset, the average betweenness centrality of cancer proteins is larger than that of disease and control proteins, but smaller than that of essential proteins.

In this paper, we introduce an active contour based scheme to localise Drosophila embryos in RGB images. An active contour (initiated as a closed one) maybe converge to an open contour, e.g., in the case that a targeting embryo is touched by a neighbouring one. We propose an algorithmic strategy to detect and restore open active contours. The experiment results show the promise of the proposed scheme.