E-grocery offers customers an alternative to traditional brick-and-mortar grocery retailing. Customers select e-grocery for convenience, making use of the home delivery at a selected time slot. In contrast to brick-and-mortar retailing, in e-grocery on-stock information for stock keeping units (SKUs) becomes transparent to the customer before substantial shopping effort has been invested, thus reducing the personal cost of switching to another supplier. As a consequence, compared to brick-and-mortar retailing, on-stock availability of SKUs has a strong impact on the customer's order decision, resulting in higher strategic service level targets for the e-grocery retailer. To account for these high service level targets, we propose a suitable model for accurately predicting the extreme right tail of the demand distribution, rather than providing point forecasts of its mean. Specifically, we propose the application of distributional regression methods -so-called Generalised Additive Models for Location, Scale and Shape (GAMLSS) -to arrive at the cost-minimising solution according to the newsvendor model. As benchmark models we consider linear regression, quantile regression, and some popular methods from machine learning. The models are evaluated in a case study, where we compare their out-of-sample predictive performance with regard to the service level selected by the e-grocery retailer considered.

Protein interaction networks are important for the understanding of regulatory mechanisms, for the explanation of experimental data and for the prediction of protein functions. Unfortunately, most interaction data is available only for model organisms. As a possible remedy, the transfer of interactions to organisms of interest is common practice, but it is not clear when interactions can be transferred from one organism to another and, thus, the confidence in the derived interactions is low. Here, we propose to use a rich set of features to train Random Forests in order to score transferred interactions. We evaluated the transfer from a range of eukaryotic organisms to S. cerevisiae using orthologs. Directly transferred interactions to S. cerevisiae are on average only 24% consistent with the current S. cerevisiae interaction network. By using commonly applied filter approaches the transfer precision can be improved, but at the cost of a large decrease in the number of transferred interactions. Our Random Forest approach uses various features derived from both the target and the source network as well as the ortholog annotations to assign confidence values to transferred interactions. Thereby, we could increase the average transfer consistency to 85%, while still transferring almost 70% of all correctly transferable interactions. We tested our approach for the transfer of interactions to other species and showed that our approach outperforms competing methods for the transfer of interactions to species where no experimental knowledge is available. Finally, we applied our predictor to score transferred interactions to 83 targets species and we were able to extend the available interactome of B. taurus, M. musculus and G. gallus with over 40,000 interactions each. Our transferred interaction networks are publicly available via our web interface, which allows to inspect and download transferred interaction sets of different sizes, for various species, and at specified expected precision levels. Availability: http://services.bio.ifi.lmu.de/coin-db/.

No abstract

BackgroundMany large data compendia on context-specific high-throughput genomic and regulatory data have been made available by international research consortia such as ENCODE, TCGA, and Epigenomics Roadmap. The use of these resources is impaired by the sheer size of the available big data and big metadata. Many of these context-specific data can be modeled as data derived regulatory networks (DDRNs) representing the complex and complicated interactions between transcription factors and target genes. These DDRNs are useful for the understanding of regulatory mechanisms and helpful for interpreting biomedical data.ResultsThe Cross-species Conservation framework (CroCo) provides a network-oriented view on the ENCODE regulatory data (CroCo network repository), convenient ways to access and browse networks and metadata, and a method to combine networks across compendia, experimental techniques, and species (CroCo tool suite). DDRNs can be combined with additional information and networks derived from the literature, curated resources, and computational predictions in order to enable detailed exploration and cross checking of regulatory interactions. Applications of the CroCo framework range from simple evidence look-up for user-defined regulatory interactions to the identification of conserved sub-networks in diverse cell-lines, conditions, and even species.ConclusionCroCo adds an intuitive unifying view on the data from the ENCODE projects via a comprehensive repository of derived context-specific regulatory networks and enables flexible cross-context, cross-species, and cross-compendia comparison via a basis set of analysis tools.The CroCo web-application and Cytoscape plug-in are freely available at: http://services.bio.ifi.lmu.de/croco-web. The web-page links to a detailed system description, a user guide, and tutorial videos presenting common use cases of the CroCo framework.

SummaryThe automated annotation of data from high throughput sequencing and genomics experiments is a significant challenge for bioinformatics. Most current approaches rely on sequential pipelines of gene finding and gene function prediction methods that annotate a gene with information from different reference data sources. Each function prediction method contributes evidence supporting a functional assignment. Such approaches generally ignore the links between the information in the reference datasets. These links, however, are valuable for assessing the plausibility of a function assignment and can be used to evaluate the confidence in a prediction. We are working towards a novel annotation system that uses the network of information supporting the function assignment to enrich the annotation process for use by expert curators and predicting the function of previously unannotated genes. In this paper we describe our success in the first stages of this development. We present the data integration steps that are needed to create the core database of integrated reference databases (UniProt, PFAM, PDB, GO and the pathway database AraCyc) which has been established in the ONDEX data integration system. We also present a comparison between different methods for integration of GO terms as part of the function assignment pipeline and discuss the consequences of this analysis for improving the accuracy of gene function annotation. The methods and algorithms presented in this publication are an integral part of the ON-DEX system which is freely available from http://ondex.sf.net/.