Workflow and Pipelining in Cheminformatics

InforSense

The InforSense platform¹ provides a scalable, analytical middleware to build, deploy and embed analytical applications into Web portals, Customer Relationship Management (CRM) systems etc., using a Service Oriented Architecture (SOA) methodology. It is built on four sets of components: analytical workflow, analytical portal, in-database execution, and grid. Analytical workflows allow users to compose and deploy different analytical methods and data transforms without coding.

(An analytical workflow is similar to a Pipeline Pilot protocol, and a group of InforSense modules is akin to a Pipeline Pilot component collection.) InforSense has over 700 different modules that can be visually composed. The InforSense analytical portal allows workflows to be packaged as reusable, interactive Web applications. Applications built from analytical workflows can be executed in the InforSense platform or the InforSense Grid. In-database execution technology enables workflows to be embedded within a database. This has the advantages of scalability, security and data fidelity, assuming all the data are in, say, Oracle, as analytics can take place without data extraction.

There are analytical modules for data mining, and statistics and domain specific modules, GenSense, ChemSense, BioSense and ClinicalSense, for genomics, cheminformatics, bioinformatics, and clinical research, respectively. ChemSense allows combination of "best of breed" tools from multiple cheminformatics vendors, including ChemAxon,⁶ Daylight,⁷ MDL (now Symyx),⁸ Molecular Networks,⁹ and Tripos,¹⁰ and can integrate Oracle chemical cartridges. An SDK allows users to implement custom applications. The InforSense Spotfire connector provides interaction between Spotfire DecisionSite for visual analytics and the InforSense platform.

Pharmaceutical, biotechnology, consumer goods, healthcare, financial services, manufacturing and communications companies are using InforSense. AstraZeneca and GlaxoSmithKline are reportedly using the in-database technology.¹¹ Windows, Mac OS, Linux and Solaris are supported (subject to the software from third parties being able to operate on those platforms).