No abstract

No abstract

The Medaka Expression Pattern Database (MEPD; http://mepd.cos.uni-heidelberg.de/) is designed as a repository of medaka expression data for the scientific community. In this update we present two main improvements. First, we have changed the previous clone-centric view for in situ data to a gene-centric view. This is possible because now we have linked all the data present in MEPD to the medaka gene annotation in ENSEMBL. In addition, we have also connected the medaka genes in MEPD to their corresponding orthologous gene in zebrafish, again using the ENSEMBL database. Based on this, we provide a link to the Zebrafish Model Organism Database (ZFIN) to allow researches to compare expression data between these two fish model organisms. As a second major improvement, we have modified the design of the database to enable it to host regulatory elements, promoters or enhancers, expression patterns in addition to gene expression. The combination of gene expression, by traditional in situ, and regulatory element expression, typically by fluorescence reporter gene, within the same platform assures consistency in terms of annotation. In our opinion, this will allow researchers to uncover new insights between the expression domain of genes and their regulatory landscape.

Structural learning of Bayesian networks (BNs)is an NP-hard problem which is generally addressed by means of heuristic search algorithms. Despite the fact that earlier proposals for dealing with this task were based on searching the space of Directed Acyclic Graphs (DAGs), there are some alternative approaches. One of these approaches for structural learning consists of searching the space of orderings, as given a certain topological order among the problem variables, it is relatively easy to build (and evaluate) a BN compatible with it. In practice, the latter methods make it possible to obtain good results, but they are still costly in terms of computation. In this article, we prove the correctness of the method used to evaluate each ordering, and we propose some efficient learning algorithms based on it. Our first proposal is based on the Hill-Climbing algorithm, and uses an improved neighbourhood definition. The second algorithm is an extension of the first one, and is based on the well-known Variable Neighbourhood Search metaheuristic. Finally, iterative versions of both algorithms are also proposed. The algorithms have been tested over a set of different domains, and have been compared with other methods such as HillClimbing in the space of DAGs or Greedy Equivalent Search, in order to study their behaviour in practice.

No abstract