Equipment & Instrumentation
Database Hardware/Security/Data Integrity
Hardware maintenance, security, data integrity and preservation is provided by the Research Information Technology (IT) Department, except for the software and hardware associated with the Beowulf-style parallel computer that the core owns. Patient confidentiality, in compliance with HIPAA security and privacy regulations, is protected.
Supercomputer Hardware
The Cancer Center has access to several Beowulf-style supercomputers of varying sizes for development, testing and performing computationally intensive analysis. A nine-node cluster at the Cancer Center and a 12-node cluster at USF are available for development and moderate computer-intensive problems. There are also 16- and 30-node clusters at the Maui Supercomputer in Hawaii which provide unlimited access for more complex computations. An additional 42-node cluster at the Cancer Center is used primarily by the HTS & Chemistry core for its data analysis. One node is reserved for managing the supercomputer and the remainders are dual processors. The total processor capability of each computer is double the number of nodes minus one.
Individual Workstations
Core members use Windows and/or Linux-based workstations for daily operations.
Software
The Cancer Center holds licenses for Spotfire (visualization tool); Matlab (library of basic numerical analysis routines); Pathway Assist from Ariadne, (pathways database and visualization tool); Metacore from Genego (pathways database and visualization tool); Ingenuity (pathways database and visualization tool); Rosetta Resolver (microarray databasing and visualization tool); and SAS (statistical analysis software).
Public domain software is available for C4.5 (rule induction algorithm to generate decision trees to classify data); GALLOPS (genetic algorithm); SVM (support vector machines algorithm); ANN (artificial neural networks algorithm); BLAST and netBLAST (basic alignment sequence tool); BRUTE (rule induction algorithm to classify data); Hierarchical Clustering; and Heatmaps and TreeView.
Usage Priority
Priority is given to Cancer Center members with peer-reviewed funding followed by those not currently involved in peer-reviewed research. The Bioinformatics Steering Committee prioritizes institutional projects and non-peer reviewed projects. If utilizing clinical data, prospective users must have IRB approval for their project prior to using this core.