The biomedical computing infrastructure for pancreatic cancer (PC) research has been developed and is being supported by the staff of the PI's laboratory. The Pancreatic Cancer Collaborative Registry (PCCR) - a Web- based software system that is designed to gather, validate, and share pancreatic cancer-related data among cancer centers around the world - is at the heart of this infrastructure. At the present time, nine centers are utilizing the PCCR and the number of participating centers continues to grow. Using this infrastructure, demographic, lifestyle, quality of life, dietary preferences, physical activities, family history, genetic and medical data, as well as treatment and surgery data on more than 2,000 PC and normal (PC-free) subjects has already been collected. The main objective of this project is to significantly enhance the PC biomedical computing infrastructure to make it: (i) easily upgradeable, repairable and interoperable with other related systems;(ii) compliant with the emerging requirements for data collection, access, mining, management and sharing;and (iii) a computing platform useful for developing novel hypotheses and studies aimed at better understanding the PC phenomena, identify PC risk factors and predictors of PC survival, and determine best practices for PC therapies. To achieve this objective, the following aims will be accomplished.
Aim 1 : Development of a controlled vocabulary and ontology for the PCCR System. Several thousands of terms from the PCCR glossary will be harmonized with terms from the Cancer Data Standards Repository (caDSR), the NCI Thesaurus (NCIT), and the Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT). The ontological relationships between these terms will be created using Web Ontology Language (OWL). The PC Ontology to be created, will extend NCIT's and caDSR's coverage of the PC domain.
Aim 2 : Re-engineering the PCCR. The need for the re-engineering the PCCR is dictated by new NIH requirements on the interoperability of the systems and data sharing. The activities planned in this aim will make this infrastructure: (i) semantically and syntactically interoperable with other related tools;(ii) easily upgradeable and repairable;(iii) compliant with new standards for data collection, access, mining, management, and sharing;(iv) coupled with the caTissue Suite for managing biospecimen data;and (iv) having advanced data reporting functionality beneficial for data visualization, data mining, and statistical analysis.
Aim 3 : Development of statistical models and computing tools for the PC research. The long-term goal of this aim is to develop a solid statistical basis and computing tools that will assist researchers in: (i) better understanding the PC phenomena;(ii) determining the need to further investigate the likelihood of PC for a subject, based on the epidemiological data and findings from physical and radiological examinations;(iii);choosing the best personalized PC treatment;and (iv) prediction of the survival rate for patients with certain characteristics after being diagnosed with PC, when a particular treatment will be used.
Aim 4 : Providing continuous maintenance of the biomedical computing platform and support of the PC research. Maintenance of the systems forming the PC computing infrastructure, recruiting new centers, user training, data collection, data security, comprehensive data validation and quality control, data sharing, and data dissemination, require the development of practical methods and contributing sufficient resources.
This aim i s focused on determining the most suitable procedures necessary for continuous maintenance and support of the computing activities promoting the PC research.
Pancreatic Cancer (PC) is the fourth leading cause of cancer death in both men and women. This project is aimed to significantly enhance the biomedical computing infrastructure for PC research to make it a platform for the development of novel hypotheses and studies aimed to identify PC risk factors and predictors of PC survival, and to determine best practices for PC therapies.
|Mdzinarishvili, Tengiz; Sherman, Simon (2014) Heuristic modeling of carcinogenesis for the population with dichotomous susceptibility to cancer: a pancreatic cancer example. PLoS One 9:e100087|
|Mdzinarishvili, Tengiz; Gleason, Michael X; Kinarsky, Leo et al. (2011) Extension of cox proportional hazard model for estimation of interrelated age-period-cohort effects on cancer survival. Cancer Inform 10:31-44|
|Sherman, Simon; Shats, Oleg; Ketcham, Marsha A et al. (2011) PCCR: Pancreatic Cancer Collaborative Registry. Cancer Inform 10:83-91|