Marina is an OS-independent GUI tool for computing TFBS abundance
given two sets of promoter sequences.
Marina performs such computations by harnessing 7 knowledge-discovery statistical metrics and the
hypergeometric distribution so as
to infer magnitude of TFBS over-representation. A standardization algorithm known as Iterative Proportional Fitting (IPF)
enables "agreement" across these various metrics as to which TFBSs are the most over-represented and which are not.
In order to run Marina, a collection of TFBS models are required. Sample models are included.
Valid models are partitioned into two specific classes:
Marina works with TFBS models be-it PWMs and/or DNA motif as long as they follow their respective schema
outlined in the documentation.
This tool is designed for ease-of-use and interpretation; to aide the investigator deduce over/under-represented TFBSs and magnitude of.
Originally, Marina was built using Python and the PyQt4 GUI library (see Marina 0.67). We re-built Marina from the ground-up using Java and JavaFX (a GUI library now part of the Java7+ language). We encourage usage of this latest Java build over prior Python builds as it provides increased UI interaction, runtime performance, various bug-fixes and improved threading support.