Areas:

• Advance NLP techniques in biomedicine - With the use of computers in storing the explosive amount of biomedical information, natural language processing (NLP) approaches have been explored to make the task of managing information recorded in free text more feasible. We are interested in using corpus-based unsupervised machine learning and online resources to build NLP applications in biomedicine.
  
  
• Ontology-based database integration - Currently, there are hundreds of biomedical databases. Ontology-based methodologies for database integration promote precise communication between scientists, enable information retrieval across multiple resources, and extend the power of computational approaches to perform data exploration, inference and mining. We promote the use of data mining and text mining techniques for ontology development and relationship establishment.
  
  
• Microarray data analysis - DNA microarray technology has provided an opportunity to simultaneously monitor the expression levels of a large number of genes in response to intentional experiment perturbations such as gene disruptions and drug treatments. We conduct research on upper-stream analysis including probe set redefinition, probe-level analysis, and down-streaming analysis using Gene Ontology.
  
  

Ongoing projects:

• Knowledge acquistion: acquire lexical and semantic knowledge of biomedical terms from online resources and text. Projects include BioThesaurus and SFThesaurus (a collaboration with Protein Information Resource; supported by NSF award IIS0639062)
  
  
• Biological named entity tagging: use supervised and unsupervised machine learning techniques for biological named entity tagging (a collaboration with Protein Information Resource and Dr. Carol Friedman from Columbia University; supported by NSF award IIS0639062)
  
  
• Ontology development, implementation, and visualization: develop, implement, and visualize ontologies in the biomedical domain. Projects include DynGO and Protein Ontology (a collaboration with Protein Information Resource; supported by NIH/NIGMS/R01GM080646-01)
  
  
• Microarray data analysis: regroup Affymetrix probes into probe sets according to sequences, provide probe-level analysis, provide transcript-level probe design support, and provide down-stream analysis based on Gene Ontology. Projects include AffyProbeMiner , SpliceMiner, GoMiner , and DynGO.The first two are collaborations with Dr. John Weinstein and Dr. Barry Zeeberg from NIH/NCI/CCR/LMP
  
  
• Software engineering in biomedical informatics: collaborate with Dr. Gunes Koru in the Empirical & Applied Software Engineering Lab from UMBC to use machine learning and survival analysis to study the biomedical open source software error proneness.
  
  
• Ontology-based database integration: develop an ontology-based system for managing microarray experiments for Breast Cancer (a collaboration with Dr. Robert Clarke)