Biology Reference
In-Depth Information
As the functioning of ToolBus is not dependent on any data domain
type, it can easily support new data types. This allows new analysis and
visualization modules to be rapidly added to PathPort. It is scalable
both in the number of users and the services it provides, as data
sources, analysis tools, and data models are all decentralized and web
service tools, are hosted on multiple servers. PathPort uses XML,
JAVA and associated technologies, making it a platform-independent
application.
The client-side interconnect, ToolBus, provides users with a con-
sistent visualization interface and gives users greater control over data
and tools located at off-site servers, allowing them to download only
the data necessary for their work. ToolBus also allows users to create
groups of associated data; such associations may be made between dif-
ferent data types that reside in separate visualizations. Data association
is further supported by a novel function of ToolBus — group sug-
gestors. Group suggestors monitor the data in ToolBus to look for
possible associations among the data, which are then presented to the
user. Associations can be made on the basis of gene ontology assign-
ments, sequence identifiers (GenBank or Swissprot identifiers), pre-
computed orthologs (planned), and other shared attributes. ToolBus
allows the saving and sharing of work sessions. Because of the plat-
form-independent nature of the application, work sessions may be
shared among colleagues working on different operating systems.
PathPort offers a wide range of biological data analysis tools
(Table 3). Pathport covers gene prediction programs, sequence align-
ment (pairwise and multiple), sequence comparison, sequence simi-
larity searches, and microarray data analysis modules. Aside from
biological data and analysis tools, PathPort also provides manually
curated information in the form of Pathogen Information (PathoInfo)
documents and Molecular Interaction Networks (MINet). PathInfo
documents contain detailed background information on up to 46 bio-
threat agents. These manually curated documents describe the taxonomy,
lifecycle, epidemiology, and the host range of the pathogen, including
any relevant biosafety information, methods of culturing, and diagnostic
tests (Fig. 3). PathInfo documents, in the PIML format (described
above), may be accessed through ToolBus. The MINet project pro-
vides a graphical representation of the knowledge of host-pathogen
