Workflow Tools

This section contains descriptions of a cross-section of the workflow-based tools designed to manage computations in the discovery arena. This is not an exhaustive list of such products, or even of just those with a life science focus. In addition, although there are interesting workflow tools focused on modeling and statistical analysis, they are beyond the scope of this chapter (e.g., the Clementine workbench from SPSS [DM1] and Insightful Miner from Insightful Corp. [DM2]). 

A number of collaboration and workflow tools are available to solve visibility problems both within the organization and between organizations. Technologies like barcode, EDI, and enterprise resource planning (ERP) have played a critical role in improving visibility. Emerging technolo-... 

Thorsten M, Wiswedel B, Berthold, Michael R (2012) Workflow tools for managing biological and chemical data. In Guha R, Bender A (eds) Computational approaches in chemoinformatics and bioinformatics, pp 179-209... 

Initially, it was planned to implement a separate interactive front-end for the CIDB to allow for a native access to the database contents. However, our experience shows that such a front-end is rarely used in practice and that it is sufficient to provide access via workflow tools (e.g., Pipeline Pilot [6] and KNIME [7]). Extracts of the data are made available to end users through the project data marts described in Section 13.2.5. 

We currently use a generic mechanism to deploy such standardized descriptor calculations, QS AR-models, and other services that have been implemented in different workflow tools to a number of different front-ends. This ensures that the end user sees identical descriptor values regardless of the front-end. The mechanism works completely asynchronously and uses a database as intermediate communication infrastructure. The database layer ensures that a request is stored persistently until it is processed—it thus automatically provides a buffering mechanism for large requests or computationally expensive calculations and guarantees that no requests are lost even if one of the calculation engines should be unavailable. 

The identification of controlled substances is a good example where the usage of workflow tools is extremely helpful to automate well defined and recurrent processes. Once all data flows are defined and the workflow is set up and productive, the annotation is done completely automated. But in the end it still is the human individual who finally has to decide how to proceed with the findings. 

If a new Cl method is established internally, an important question is whether it should remain an expert technology that has to be applied and interpreted by the computational chemist as a service for colleagues from other disciplines or if it is suitable for deployment to a wider user base. This discussion is facilitated by the fact that modem workflow tools allow for a simple exchange of workflows and protocols... 

A further insight is that the best workflow depends on a combination of factors that can in many cases be expressed in closed mathematical form, allowing very rapid graphical feedback to users of what then becomes a visualization rather than a stochastic simulation tool. This particular approach is effective for simple binary comparisons of methods (e.g., use of in vitro alone vs. in silico as prefilter to in vitro). It can also be extended to evaluation of conditional sequencing for groups of compounds, using an extension of the sentinel approach [24]. 

The ultimate goal of all scientists is to analyze their data thoroughly until they are sure that it is valid and to then analyze it in a more global context and discuss it with their colleagues. This workflow requires enterprise level IT tools that can effectively compare and correlate multiple HTS campaigns that generated millions of results from hundreds of thousands of compounds, recognize and chart trends and hierarchies of association and help the scientist visualize them, annotate them, and render the visualizations in media that can be used to share that vision with other members of the team. 

Meier MAR, Schubert US (2005) Integration of MALDl-TOEMS as high-throughput screening tool into the workflow of combinatorial polymer research. Rev Sci Instrum 76 062211... 

There is a long history of efforts to find simple and interpretable /i and fi functions for various activities and properties (29, 30). The quest for predictive QSAR models started with Hammett s pioneer work to correlate molecular structures with chemical reactivities (30-32). However, the widespread applications of modern predictive QSAR and QSPR actually started with the seminal work of Hansch and coworkers on pesticides (29, 33, 34) and the developments of various powerful analysis tools, such as PLS (partial least squares) and neural networks, for multivariate analysis have fueled these widespread applications. Nowadays, numerous publications on guidelines, workflows, and... 

Key words Drug discovery, chem-informatics, molecular design, combinatorial chemistry, combinatorial library, synthesis protocol, PGVL, reactant, product, enumeration, filtering integration, workflow, streamline, desktop tool, software deployment. 

---