Domenica D'Elia

Integrating microRNA target predictions for the discovery of gene regulatory networks: a semi-supervised ensemble learning approach

BackgroundMicroRNAs (miRNAs) are small non-coding RNAs which play a key role in the post-transcriptional regulation of many genes. Elucidating miRNA-regulated gene networks is crucial for the understanding of mechanisms and functions of miRNAs in many biological processes, such as cell proliferation, development, differentiation and cell homeostasis, as well as in many types of human tumors. To this aim, we have recently presented the biclustering method HOCCLUS2, for the discovery of miRNA regulatory networks. Experiments on predicted interactions revealed that the statistical and biological consistency of the obtained networks is negatively affected by the poor reliability of the output of miRNA target prediction algorithms. Recently, some learning approaches have been proposed to learn to combine the outputs of distinct prediction algorithms and improve their accuracy. However, the application of classical supervised learning algorithms presents two challenges: i) the presence of only positive examples in datasets of experimentally verified interactions and ii) unbalanced number of labeled and unlabeled examples.ResultsWe present a learning algorithm that learns to combine the score returned by several prediction algorithms, by exploiting information conveyed by (only positively labeled/) validated and unlabeled examples of interactions. To face the two related challenges, we resort to a semi-supervised ensemble learning setting. Results obtained using miRTarBase as the set of labeled (positive) interactions and mirDIP as the set of unlabeled interactions show a significant improvement, over competitive approaches, in the quality of the predictions. This solution also improves the effectiveness of HOCCLUS2 in discovering biologically realistic miRNA:mRNA regulatory networks from large-scale prediction data. Using the miR-17-92 gene cluster family as a reference system and comparing results with previous experiments, we find a large increase in the number of significantly enriched biclusters in pathways, consistent with miR-17-92 functions.ConclusionThe proposed approach proves to be fundamental for the computational discovery of miRNA regulatory networks from large-scale predictions. This paves the way to the systematic application of HOCCLUS2 for a comprehensive reconstruction of all the possible multiple interactions established by miRNAs in regulating the expression of gene networks, which would be otherwise impossible to reconstruct by considering only experimentally validated interactions.

MitoRes: a resource of nuclear-encoded mitochondrial genes and their products in Metazoa

BackgroundMitochondria are sub-cellular organelles that have a central role in energy production and in other metabolic pathways of all eukaryotic respiring cells. In the last few years, with more and more genomes being sequenced, a huge amount of data has been generated providing an unprecedented opportunity to use the comparative analysis approach in studies of evolution and functional genomics with the aim of shedding light on molecular mechanisms regulating mitochondrial biogenesis and metabolism.In this context, the problem of the optimal extraction of representative datasets of genomic and proteomic data assumes a crucial importance. Specialised resources for nuclear-encoded mitochondria-related proteins already exist; however, no mitochondrial database is currently available with the same features of MitoRes, which is an update of the MitoNuc database extensively modified in its structure, data sources and graphical interface. It contains data on nuclear-encoded mitochondria-related products for any metazoan species for which this type of data is available and also provides comprehensive sequence datasets (gene, transcript and protein) as well as useful tools for their extraction and export.DescriptionMitoRes http://www2.ba.itb.cnr.it/MitoRes/ consolidates information from publicly external sources and automatically annotates them into a relational database. Additionally, it also clusters proteins on the basis of their sequence similarity and interconnects them with genomic data. The search engine and sequence management tools allow the query/retrieval of the database content and the extraction and export of sequences (gene, transcript, protein) and related sub-sequences (intron, exon, UTR, CDS, signal peptide and gene flanking regions) ready to be used for in silico analysis.ConclusionThe tool we describe here has been developed to support lab scientists and bioinformaticians alike in the characterization of molecular features and evolution of mitochondrial targeting sequences. The way it provides for the retrieval and extraction of sequences allows the user to overcome the obstacles encountered in the integrative use of different bioinformatic resources and the completeness of the sequence collection allows intra- and interspecies comparison at different biological levels (gene, transcript and protein).

Isolation of a 25-kDa Protein Binding to a Curved DNA Upstream the Origin of the L Strand Replication in the Rat Mitochondrial Genome

The presence of a curved DNA sequence in the gene for the NADH-dehydrogenase subunit 2 of rat mitochondrial genome, upstream from the origin of the light strand replication have been demonstrated through theoretical analysis and experimental approaches. Gel retardation assays showed that this structure makes a complex with a protein component extracted from the mitochondrial matrix. The isolation and purification of this protein is reported. With a Sepharose CL-6B and magnetic DNA affinity chromatography a polypeptide was purified to homogeneity having 25-kDa mass as shown by gel electrophoresis. To functionally characterize this protein, its capability to bind to other sequences of the homologous or heterologous DNA and to specific riboprobes was also investigated. A role for this protein as a trans-acting agent required for the expression of the mammalian mitochondrial genome is suggested.

Computational annotation of UTR cis-regulatory modules through Frequent Pattern Mining

BackgroundMany studies report about detection and functional characterization of cis-regulatory motifs in untranslated regions (UTRs) of mRNAs but little is known about the nature and functional role of their distribution. To address this issue we have developed a computational approach based on the use of data mining techniques. The idea is that of mining frequent combinations of translation regulatory motifs, since their significant co-occurrences could reveal functional relationships important for the post-transcriptional control of gene expression. The experimentation has been focused on targeted mitochondrial transcripts to elucidate the role of translational control in mitochondrial biogenesis and function.ResultsThe analysis is based on a two-stepped procedure using a sequential pattern mining algorithm. The first step searches for frequent patterns (FPs) of motifs without taking into account their spatial displacement. In the second step, frequent sequential patterns (FSPs) of spaced motifs are generated by taking into account the conservation of spacers between each ordered pair of co-occurring motifs. The algorithm makes no assumption on the relation among motifs and on the number of motifs involved in a pattern. Different FSPs can be found depending on different combinations of two parameters, i.e. the threshold of the minimum percentage of sequences supporting the pattern, and the granularity of spacer discretization. Results can be retrieved at the UTRminer web site: http://utrminer.ba.itb.cnr.it/. The discovered FPs of motifs amount to 216 in the overall dataset and to 140 in the human subset. For each FP, the system provides information on the discovered FSPs, if any. A variety of search options help users in browsing the web resource. The list of sequence IDs supporting each pattern can be used for the retrieval of information from the UTRminer database.ConclusionComputational prediction of structural properties of regulatory sequences is not trivial. The presented data mining approach is able to overcome some limits observed in other competitive tools. Preliminary results on UTR sequences from nuclear transcripts targeting mitochondria are promising and lead us to be confident on the effectiveness of the approach for future developments.

A 67-kDa protein binding to the curved DNA in the main regulatory region of the rat mitochondrial genome

We have purified, by sequence-specific affinity chromatography, a mitochondrial (mt) matrix protein which binds to the curved DNA located between the replication origin (ori) of the leading strand (ori-H) and the two transcription promoters in the rat mt genome. The protein was characterized by gel electrophoresis as a 67-kDa polypeptide and seems to be involved in the DNA contact on the mt light strand. This protein differs (in the size and location of its DNA-binding site) from other DNA-binding proteins studied so far in animal mt systems. We suggest a role for the 67-kDa protein, assisted by other proteins, in regulating the initiation of leading-strand replication.

Hierarchical and overlapping co-clustering of mRNA: iRNA interactions

microRNAs (miRNAs) are an important class of regulatory factors controlling gene expressions at post-transcriptional level. Studies on interactions between different miRNAs and their target genes are of utmost importance to understand the role of miRNAs in the control of biological processes. This paper contributes to these studies by proposing a method for the extraction of co-clusters of miRNAs and messenger RNAs (mRNAs). Different from several already available co-clustering algorithms, our approach efficiently extracts a set of possibly overlapping, exhaustive and hierarchically organized co-clusters. The algorithm is well-suited for the task at hand since: i) mRNAs and miRNAs can be involved in different regulatory networks that may or may not be co-active under some conditions, ii) exhaustive co-clusters guarantee that possible co-regulations are not lost, iii) hierarchical browsing of co-clusters facilitates biologists in the interpretation of results. Results on synthetic and on real human miRNA:mRNA data show the effectiveness of the approach.

KEYnet: a keywords database for biosequences functional organization

KEYnet is a database where gene and protein names are hierarchically structured. Particular care has been devoted to the search and organisation of synonyms. The structuring is based on biological criteria in order to assist the user in the data search and to minimise the risk of loss of information. Links to the EMBL data library by the entry name and the accession number have been implemented. KEYnet is available through the World Wide Web at the following site: http://www.ba.cnr.it/keynet.html. Recently KEYnet has incorporated specific gene name classifications, which can be browsed starting from the above-mentioned KEYnet home page: the Mitochondrial Gene Names classification and the Rat Gene Names classification. KEYnet database has also been structured in a flatfile format and can be queried through SRS (http://bio-www.ba.cnr.t:8000/srs).

Network reconstruction for the identification of miRNA: RNA interaction networks

Network reconstruction from data is a data mining task which is receiving a significant attention due to its applicability in several domains. For example, it can be applied in social network analysis, where the goal is to identify connections among users and, thus, sub-communities. Another example can be found in computational biology, where the goal is to identify previously unknown relationships among biological entities and, thus, relevant interaction networks. Such task is usually solved by adopting methods for link prediction and for the identification of relevant sub-networks. Focusing on the biological domain, in [4] and [3] we proposed two methods for learning to combine the output of several link prediction algorithms and for the identification of biological significant interaction networks involving two important types of RNA molecules, i.e. microRNAs (miRNAs) and messenger RNAs (mRNAs). The relevance of this application comes from the importance of identifying (previously unknown) regulatory and cooperation activities for the understanding of the biological roles of miRNAs and mRNAs. In this paper, we review the contribution given by the combination of the proposed methods for network reconstruction and the solutions we adopt in order to meet specific challenges coming from the specific domain we consider.