Information Technology Reference
In-Depth Information
GNU Make
GNU make [8] is commonly known as a build utility that can be used for
the automation of source code compilation processes. In the bioinformatics
community it has also become a popular tool for producing pipelines and
simple workflows using existing command-line tools as components [78]: make
is able to handle and resolve dependencies between targets - hence, defining
individual tool invocations behind the individual targets, and providing make
with a final target will cause all required tools to be executed.
While GNU make saves programmers the work of defining workflows
explicitly, makefiles are often confusing and in particular not accessible
to people without programming experience. Domain-specific wrappers like
BioMake/Skam [2] add some convenience for using GNU make for the def-
inition and execution of bioinformatics workflows, but they do not increase
the power of the approach: the automatic resolution of dependencies that is
done by GNU make is at the end of the day only able to handle simple goal
specifications and workflows of low complexity.
JOpera
JOpera [26], also sometimes called BioOpera [38], is an Eclipse-based plat-
form that provides “process support for (more than) web services” [7]. It
has been developed in the scope of a joint project of the Information and
Communication Systems Research Group (IKS) and the Computational Bio-
chemistry Research Group (CBRG) at ETH Zurich, Switzerland. The project
aimed at improving and automating the large-scale analysis of genetic data
sets [25].
JOpera applies the JOpera Visual Composition Language (JVCL) [245,
Chapter 3] as a graphical notation for processes. Its essential components
are service interfaces (representing, e.g., web services, Java methods, scripts,
or subtasks) and their parameters. JOpera uses two models for specifying a
workflow: the main process model defines the data-flow between the tasks
and parameters, while a separate second model is used for more complex
control-flow definitions between the tasks. The processes are compiled into
(Java) executable code to achieve ecient execution.
SADI/SHARE
SADI (Semantic Annotated Discovery and Integration) [342, 343], primarily
developed at the University of British Columbia in Vancouver (Canada), is a
framework of conventions and best practices for the use of standard Semantic
Web technologies for working with web services. Currently the main applica-
tions of SADI are in the life sciences, but the framework itself is conceived
to be domain-independent. SHARE (Semantic Health and Research Environ-
ment) [331] is the current reference implementation of the SADI approach,
Search WWH ::
Custom Search