Category Cross-Omics>Workflow Knowledge Bases/Systems/Tools

Abstract e-BioFlow is an advanced workflow tool.

E-BioFlow enables the life scientist to compose life science experiments by connecting web services.

E-Bioflow is a ‘control flow’ based workflow tool, in contrast with most life science workflow tools, which are ‘data flow’ oriented.

Using the control flow paradigm, e-BioFlow is able to model more advanced control structures, such as parallel execution of tasks, iteration (loops) and conditional branching (if-then-else).

The workflows designed in e-BioFlow can contain advanced ‘control flow structures’ and are independent of the available web services.

e-BioFlow supports ‘hierarchical workflows’ to deal with complex experiments and has the ability to reuse sub-workflows.

The interface of e-BioFlow provides three (3) different perspectives:

Control Flow perspective, Data Flow perspective and the Resource perspective.

1) Control Flow perspective - Defines the order of execution.

Advanced control flow structures are supported, such as iteration, conditions and hierarchy.

2) Data Flow perspective - Defines the information flow between tasks.

Data is typed based on both semantics and encoding.

3) Resource perspective - Abstract from resources (web services) until the instantiation of the workflow.

In fact, in the Resource perspective, the user defines the types of web services required to perform the tasks.

By providing these perspectives, e-BioFlow prevents the user from getting overloaded by the amount of information presented, but does Not restrict the user from creating complex workflows.

e-BioFlow key features and differences with respect to existing workflow tools:

1) e-BioFlow is a ‘control flow workflow’ tool. Tasks can be executed in sequential, parallel, as well as iterative and conditional order.

2) The information flow between tasks is as easy to model as in any data flow oriented workflow tool.

3) e-BioFlow supports late-binding. Tasks do Not directly represent services, but rather service types.

At runtime, the e-BioFlow engine looks for services that are available corresponding to the service type.

4) BioMOBY (see G6G Abstract Number 20520) services and any Web Service Description Language (WSDL) services are supported.

5) e-BioFlow provides support for different scripting languages: R, Perl and Sleep (a Perl-like language).

Support for ‘Java Beanshell’ scripting will be provided in the future.

6) The e-BioFlow engine is built on YAWL, a reliable workflow engine used in business processing.

YAWL is a Business Process Management (BPM)/Workflow Management System.

Based on a concise and advanced modeling language, YAWL handles complex data, transformations, integration with organizational resources and Web Service integration.

Built in Java, it uses XML Schema and XQuery.

7) e-BioFlow is freely available and is open source.

e-BioFlow current state and future work --

Currently, an engine is integrated into e-BioFlow that supports ‘late binding’ to create a workflow design and execution environment.

In the future, provenance support will be added to enable the scientist to browse, analyze and annotate their experiments.

System Requirements

e-BioFlow Manual


Manufacturer Web Site e-BioFlow

Price Contact manufacturer.

G6G Abstract Number 20541

G6G Manufacturer Number 104156