Since the completion of the Human Genome Project, it has been widely established that most DNA is not transcribed into proteins. These non-protein-coding regions are believed to be moderators within transcriptional and post-transcriptional processes, which play key roles in the onset of diseases. Long non-coding RNAs (lncRNAs) are generally lacking in conserved motifs typically used for detection and thus hard to identify, but nonetheless present certain characteristic features that can be exploited by bioinformatics methods. By combining lncRNA detection with known miRNA, RNA-binding protein and chromatin interaction, current tools are able to recognize and functionally annotate large number of lncRNAs. This review discusses databases and platforms dedicated to cataloging and annotating lncRNAs, as well as tools geared at discovering novel sequences. We emphasize the issues posed by the diversity of lncRNAs and their complex interaction mechanisms, as well as technical issues such as lack of unified nomenclature. We hope that this wide overview of existing platforms and databases might help guide biologists toward the tools they need to analyze their experimental data, while our discussion of limitations and of current lncRNA-related methods may assist in the development of new computational tools.

UNLABELLED: NetPathMiner is a general framework for mining, from genome-scale networks, paths that are related to specific experimental conditions. NetPathMiner interfaces with various input formats including KGML, SBML and BioPAX files and allows for manipulation of networks in three different forms: metabolic, reaction and gene representations. NetPathMiner ranks the obtained paths and applies Markov model-based clustering and classification methods to the ranked paths for easy interpretation. NetPathMiner also provides static and interactive visualizations of networks and paths to aid manual investigation. AVAILABILITY: The package is available through Bioconductor and from Github at http://github.com/ahmohamed/NetPathMiner.

Like all models, network feature selection models require that assumptions be made on the size and structure of the desired features. The most common assumption is sparsity, where only a small section of the entire network is thought to produce a specific phenomenon. The sparsity assumption is enforced through regularized models, such as the lasso. However, assuming sparsity may be inappropriate for many real-world networks, which possess highly correlated modules. In this paper, we illustrate two novel optimization strategies, namely, boosted expectation propagation (BEP) and boosted message passing (BMP), which directly use the network structure to estimate the parameters of a network classifier. BEP and BMP are ensemble methods that seek to optimize classification performance by combining individual models built upon local network features. Neither BEP nor BMP assumes a sparse solution, but instead they seek a weighted average of all network features where the weights are used to emphasize all features that are useful for classification. In this paper, we compare BEP and BMP with network-regularized logistic regression models on simulated and real biological networks. The results show that, where highly correlated network structure exists, assuming sparsity adversely effects the accuracy and feature selection power of the network classifier.

MOTIVATION: An observed metabolic response is the result of the coordinated activation and interaction between multiple genetic pathways. However, the complex structure of metabolism has meant that a compete understanding of which pathways are required to produce an observed metabolic response is not fully understood. In this article, we propose an approach that can identify the genetic pathways which dictate the response of metabolic network to specific experimental conditions. RESULTS: Our approach is a combination of probabilistic models for pathway ranking, clustering and classification. First, we use a non-parametric pathway extraction method to identify the most highly correlated paths through the metabolic network. We then extract the defining structure within these top-ranked pathways using both Markov clustering and classification algorithms. Furthermore, we define detailed node and edge annotations, which enable us to track each pathway, not only with respect to its genetic dependencies, but also allow for an analysis of the interacting reactions, compounds and KEGG sub-networks. We show that our approach identifies biologically meaningful pathways within two microarray expression datasets using entire KEGG metabolic networks. AVAILABILITY AND IMPLEMENTATION: An R package containing a full implementation of our proposed method is currently available from http://www.bic.kyoto-u.ac.jp/pathway/timhancock.

This paper presents research projects tackling two aspects in data mining. First, a toolbox is discussed that allows flexible and interactive data exploration, analysis and presentation using the scripting language Python. The advantages of this toolbox are that it provides the functionality to process multiple SQL queries in parallel, and enables fast data retrieval using a supervised caching mechanism for commonly used queries. These two facets of the toolbox allow for fast, efficient data access reducing the time spent on data exploration, preparation and analysis. Secondly, an approach to predictive modelling is presented that leads to scalable parallel algorithms for high dimensional data collections. This is an essential requirement for data mining algorithms as those that do not scale linearly with the data size are infeasible. These algorithms are implemented in parallel and achieve an almost ideal speedup for their respective implementations. One aim of the presented research is to integrate and combine these two different aspects of data mining into an efficient but flexible data mining toolbox that allows the experienced data miner to attack large scale problems interactively or with batch processing.