Soybean miRNA Functional Network

                          miRNA Network | Target Network
                                          miRNA-Target datasets |3'-UTR |Target Annotations|Target Pathway
                                                            Functional Similarity | miRNA Network | Statistics | Visit Log
                                                                                                     Contact | Privacy policy
Note:

This site will no longer be updated and maintained from now on. The current services will continue to be provided and updated at the new database: SoyFN. We apologize for the inconvenience.

About GO Semantic Similarity

GO Semantic Similarity

1. Gene Ontology (GO)

Gene OntolgoyThe Gene Ontology project is a major bioinformatics initiative with the aim of standardizing the representationofgene and gene product attributesacross species and databases. The project provides a controlled vocabulary of terms for describing gene product characteristics and gene product annotation data from GOConsortium members, as well as tools to access and process this data. Read more about the Gene Ontology...



2. Semantic similarity

GFSATSemantic similarity measures have been applied and developed in biomedical ontologies, namely, the Gene Ontology (GO). They are mainly used to compare genes and proteins based on the similarity of their functions rather than on their sequence similarity, but they are also being extended to other bioentities, such as chemical compounds, anatomical entities and diseases.These comparisons can be done using our tools freely available on the web: GFSAT



3. What is SSDD?

 The Shortest Semantic Differentiation Distance (SSDD) algorithm measures semantic similarity between GO terms from a novel perspective. In SSDD, a pair of terms is represented as overlapping directed acyclic graphs, which is then viewed as a semantic genealogy. The semantic heredity from a parent to its children is regarded as a process of semantic differentiation. Then semantic distance between two terms is calculated by the capacity of redifferentiation from one term to the other. In comprehensive evaluations either against human rating or using a benchmark dataset, SSDD compares favorably with other methods and performs slightly better than simUI, another intrinsic method. SSDD addresses the issues of shallow and identical annotation and can furthermore distinguish sibling semantic similarity, in addition to its intrinsic to GO. It provides an alternative to both methods that use external resources and methods “intrinsic” to GO with comparable performance.



4. Where is SSDD better than other methods?

        (1) SSDD is a completely novel insight into GO semantic similarity.Borrowing from the biological process of cellular differentiation, the semantic of each GO term is represented as semantic totipotency, and so the semantic heredity from a parent to its children is regarded as a process of semantic differentiation.

       (2) SSDD jumps out of the reliance on external sources of data, e.g. the GOA datasets, thus to be intrinsic to the ontology;

      ( 3) SSDD, most prominently, can overcome the issues of shallow annotations and identical annotations, and furthermore distinguish the similarity of sibling terms.

About Us | Site Map | Privacy Policy | Contact Us | ©2012 Natural Computation Lab.