Guillermo Palma

Finding cross genome patterns in annotation graphs

Annotation graph datasets are a natural representation of scientific knowledge. They are common in the life sciences where concepts such as genes and proteins are annotated with controlled vocabulary terms from ontologies. Scientists are interested in analyzing or mining these annotations, in synergy with the literature, to discover patterns. Further, annotated datasets provide an avenue for scientists to explore shared annotations across genomes to support cross genome discovery. We present a tool, PAnG ( P atterns in An notation G raphs), that is based on a complementary methodology of graph summarization and dense subgraphs. The elements of a graph summary correspond to a pattern and its visualization can provide an explanation of the underlying knowledge. We present and analyze two distance metrics to identify related concepts in ontologies. We present preliminary results using groups of Arabidopsis and C. elegans genes to illustrate the potential benefits of cross genome pattern discovery.

Measuring Relatedness Between Scientific Entities in Annotation Datasets

Linked Open Data has made available a diversity of scientific collections where scientists have annotated entities in the datasets with controlled vocabulary terms (CV terms) from ontologies. These semantic annotations encode scientific knowledge which is captured in annotation datasets. One can mine these datasets to discover relationships and patterns between entities. Determining the relatedness (or similarity) between entities becomes a building block for graph pattern mining, e.g., identifying drug-drug relationships could depend on the similarity of the diseases (conditions) that are associated with each drug. Diverse similarity metrics have been proposed in the literature, e.g., i) string-similarity metrics; ii) path-similarity metrics; iii) topological-similarity metrics; all measure relatedness in a given taxonomy or ontology. In this paper, we consider a novel annotation similarity metric AnnSim that measures the relatedness between two entities in terms of the similarity of their annotations. We model AnnSim as a 1-to-1 maximal weighted bipartite match, and we exploit properties of existing solvers to provide an efficient solution. We empirically study the effectiveness of AnnSim on real-world datasets of genes and their GO annotations, clinical trials, and a human disease benchmark. Our results suggest that AnnSim can provide a deeper understanding of the relatedness of concepts and can provide an explanation of potential novel patterns.

Drug-Target Interaction Prediction Using Semantic Similarity and Edge Partitioning

The ability to integrate a wealth of human-curated knowledge from scientific datasets and ontologies can benefit drug-target interaction prediction. The hypothesis is that similar drugs interact with the same targets, and similar targets interact with the same drugs. The similarities between drugs reflect a chemical semantic space, while similarities between targets reflect a genomic semantic space. In this paper, we present a novel method that combines a data mining framework for link prediction, semantic knowledge (similarities) from ontologies or semantic spaces, and an algorithmic approach to partition the edges of a heterogeneous graph that includes drug-target interaction edges, and drug-drug and target-target similarity edges. Our semantics based edge partitioning approach, semEP, has the advantages of edge based community detection which allows a node to participate in more than one cluster or community. The semEP problem is to create a minimal partitioning of the edges such that the cluster density of each subset of edges is maximal. We use semantic knowledge (similarities) to specify edge constraints, i.e., specific drug-target interaction edges that should not participate in the same cluster. Using a well-known dataset of drug-target interactions, we demonstrate the benefits of using semEP predictions to improve the performance of a range of state-of-the-art machine learning based prediction methods. Validation of the novel best predicted interactions of semEP against the STITCH interaction resource reflect both accurate and diverse predictions.

On the Selection of SPARQL Endpoints to Efficiently Execute Federated SPARQL Queries

We consider the problem of source selection and query decomposition in federations of SPARQL endpoints, where query decompositions of a SPARQL query should reduce execution time and maximize answer completeness. This problem is in general intractable, and performance and answer completeness of SPARQL queries can be considerably affected when the number of SPARQL endpoints in a federation increases. We devise a formalization of this problem as the Vertex Coloring Problem and propose an approximate algorithm named Fed-DSATUR. We rely on existing results from graph theory to characterize the family of SPARQL queries for which Fed-DSATUR can produce optimal decompositions in polynomial time on the size of the query, i.e., on the number of SPARQL triple patterns in the query. Fed-DSATUR scales up much better to SPARQL queries with a large number of triple patterns, and may exhibit significant improvements in performance while answer completeness remains close to 100i¾?%. More importantly, we put our results in perspective, and provide evidence of SPARQL queries that are hard to decompose and constitute new challenges for data management.

AnnEvol: An Evolutionary Framework to Description Ontology-Based Annotations

Existing biomedical ontologies encode scientific knowledge that is exploited in ontology-based annotated entities, e.g., genes described using Gene Ontology (GO) annotations. Ontology-based annotations correspond to building blocks for computing relatedness between annotated entities, as well as for data mining techniques that attempt to discover domain patterns or suggest novel associations among annotated entities. However, effectiveness of these annotation-based approaches can be considerably impacted by the quality of the annotations, and models that allow for the description of the quality of the annotations are required to validate and explain the behavior of these approaches. We propose AnnEvol, a framework to describe datasets of ontology-based annotated entities in terms of evolutionary properties of the annotations of these entities over time. AnnEvol complements state-of-the-art approaches that perform an annotation-wise description of the datasets, and conducts an annotation set-wise description which characterizes the evolution of annotations into semantically similar annotations. We empirically evaluate the expressiveness power of AnnEvol in a set of proteins annotated with GO using UniProt-GOA and Swiss-Prot. Our experimental results suggest that AnnEvol captures evolutionary behavior of the studied GO annotations, and clearly differentiates patterns of annotations depending on both the annotation provider and on the model organism of the studied proteins.

OnSim: A Similarity Measure for Determining Relatedness Between Ontology Terms

Accurately measuring relatedness between ontology terms becomes a building block for determining similarity of ontology-based annotated entities, e.g., genes annotated with the Gene Ontology. However, existing measures that determine similarity between ontology terms mainly rely on taxonomic hierarchies of classes, and may not fully exploit the semantics encoded in the ontology, i.e., object properties and their axioms. This limitation may conduct to ignore the stated or inferred facts where an ontology term participate in the ontology, i.e., the term neighborhood. Thus, high values of similarity can be erroneously assigned to terms that are taxonomically similar, but whose neighborhoods are different. We present OnSim, a measure where semantics encoded in the ontology is considered as a first-class citizen and exploited to determine relatedness of ontology terms. OnSim considers the neighborhoods of two terms, as well as the object properties that are present in the neighborhood facts and the justifications that support the entailment of these facts. We have extended an existing annotation-based similarity measure with OnSim, and empirically studied the impact of producing accurate values of ontology term relatedness. Experiments were run on benchmarks published by the Collaborative Evaluation of Semantic Similarity Measures (CESSM) tool. The observed results suggest that OnSim increases the Pearson’s correlation coefficient of the annotation-based similarity measure with respect to gold standard similarity measures, as well as its effectiveness is improved with respect to state-of-the-art semantic similarity measures.

Considering Semantics on the Discovery of Relations in Knowledge Graphs

Knowledge graphs encode semantic knowledge that can be exploited to enhance different data-driven tasks, e.g., query answering, data mining, ranking or recommendation. However, knowledge graphs may be incomplete, and relevant relations may be not included in the graph, affecting accuracy of these data-driven tasks. We tackle the problem of relation discovery in a knowledge graph, and devise

An authority-flow based ranking approach to discover potential novel associations between Linked Data

\mathcal {KOI}