Logo
 

Welcome to the OMIT Project


The identification and characterization of important roles microRNAs (miRNAs) played in human cancer is an increasingly active area. MiRNAs are a class of small non-coding RNAs capable of regulating gene expression. They have been demonstrated to be involved in diverse biological functions, and miRNAs' expression profiling has identified them associated with clinical diagnosis and prognosis of several major tumor types. Unfortunately, the prediction of the relationship between miRNAs and their target genes still remains a challenging task. On the other hand, ontologies are formal, declarative knowledge representation models, playing a key role in defining formal semantics in traditional knowledge engineering. The most successful example of applying ontological techniques into biological research is the Gene Ontology (GO) project.


We propose a framework based on the Ontology for MiRNA Target (OMIT) to handle the aforementioned challenge, and our overall objective is to explore a computing framework that will facilitate knowledge acquisition from existing, and usually geographically distributed and heterogeneous sources. We aim to assist cancer biologists in unraveling important roles of miRNAs in human cancer by synthesizing data from source miRNA databases into a comprehensive conceptual model. This model will permit an emphasis on data semantics rather than on the forms in which the data was originally represented. Consequently, a more accurate, complete view of miRNAs' biological functions can be acquired. Therefore, users will be provided with a single query engine that takes their needs in a nonprocedural specification format.


There are many challenges to be handled in this research effort., some of which are envisioned below:

  • How to build and maintain a comprehensive, up-to-date knowledge representation model in miRNA area?
  • How to efficiently and effectively align our OMIT ontology with existing knowledge bases, the GO for example?
  • How to deal with the tradeoff between a "shallow" semantic annotation and a "deep" one?
  • What will be an appropriate format to specify the mapping rules during the semantic annotation?
  • Is a centralized data warehouse a good choice for annotated source databases?
  • How to balance the pros and cons of GAV (Global-As-View) and LAV (Local-As-View) approaches during the data integration?
  • How to achieve a unified query-answering for system users?

We are hoping that our investigation in all these challenges will result in paradigm-shifting advances in the fields of (1) medical informatics (in particular, the human cancer research), (2) ontological techniques, and (3) semantic annotation & integration.