Life sciences R&D labs often spend too much time and precious resources just managing and making information accessible to scientists. Extracting value from data (turning experimental results into actionable knowledge that drives decisions) is the fundamental challenge for lab managers struggling to deal with the enormous amount of information generated by today’s labs.

This challenge is best addressed through an information- driven approach to R&D that enables scientists to leverage the information and knowledge of others in order to improve experimental design and analysis and thus overall R&D productivity. To deliver an optimal information-driven R&D strategy, companies are investing in flexible, multi-disciplinary electronic laboratory notebooks (ELNs) that can be used internally across the enterprise or worldwide across business ventures. These latest-generation ELNs offer centralized data repositories; infrastructure for capturing, accessing and sharing experimental information; improved communication between instruments and related software; and workflow orchestration tools that support the diverse needs of different disciplines, without extensive customization. This approach is accelerating the documentation and reporting of experimentation while also enabling scientists to collaborate effectively on multi-stage projects and leverage existing information for improved experiment throughput and success.

The latest ELNs serve as a fulcrum supporting the convergence of instruments, software and R&D workflows in an electronic lab environment. This convergence is enabling scientists to work more efficiently in project teams while also better orchestrating daily experimental activities involving compound registration, laboratory execution and experiment analysis, data collection from instrumentation, sample management and tracking, materials/ inventory management, laboratory information management systems (LIMS)/scientific data management systems (SDMS) and data warehousing systems. Today’s systems even reach beyond the lab into business information areas such as enterprise resource planning and financial reporting.

The evolution of a convergent ELN solution

Michael Elliott, CEO of Atrium Research & Consulting, describes four generations of ELNs.1 Beginning with first-generation systems for basic data capture in the late 1990s, ELNs have evolved through a second generation that offered specific domain functionality such as searchable reaction databases, automated stoichiometry calculations and reaction planning for chemists, A third generation expanded this chemistry functionality and provided generic capabilities that extended the reach of the ELN to other disciplines, including biology. The third-generation ELN systems provided reusable forms and templates for consistent data entry as well as the ability to work with MS Word, MS Excel and other third-party applications directly, greatly improving their ease-of-use.

According to Elliott, today’s fourth-generation ELNs are all about convergence. Labs are demanding a single, flexible ELN solution that combines generic, multi-disciplinary functionality with mix-and-match, domain-specific modules for handling both structured and unstructured information. Elliott suggests that this requirement is moving fourth-generation ELNs directly into territory traditionally inhabited by LIMS and SDMS.

In response, vendors like Symyx are partnering with LIMS and SDMS providers such as Thermo Fisher Scientific to provide out-of-the-box, integrated ELN functionality along with enterprise ELN deployment, domain-specific notebook configurations and integrated decision support capabilities, enabling improvements in:

• Documentation of experimental design and results,
• Communication of procedures and information between instruments,
• Workflow orchestration among project team members and
• Capture and re-use of structured and unstructured information.

Managing structured and unstructured information

Approximately 20 percent of lab data is structured— that is, capturable, storable and retrievable in software applications such as LIMS. The remaining 80 percent is semi-structured (i.e., the experimental content and context typically captured in a lab notebook, including hypotheses, chemistry/materials, conditions, parameters, results and conclusions) and/or unstructured data (i.e., the mix of data from analytical laboratory instruments and business applications that typically is archived in an SDMS).

These different information repositories reflect the growing complexity and challenges of today’s R&D informatics environment. For example, the SDMS is an automated electronic repository for document and instrument data. The system automatically captures and retrieves raw data files (mass spectrum, chromatography, images, etc.) and archives reports in a software-neutral format that does not require original source software for viewing reports. While the SDMS easily can access printed reports from many sources, the system does not enable data analysis or comparison as easily. Neither does it compute additional results, as the reports are essentially static “snapshots,” not live data.

LIMS are sample management systems used to track studies, projects, batches/lots, plates and task assignments in a central analytical facility. They provide study-centric reporting functionality but do not facilitate cross-study analyses, which can be challenging and time-consuming.

Information archived in SDMS and LIMS systems is useful only if the information is accessible to scientists via powerful data access, analysis and decision support software.

Advanced decision support tools are evolving to help scientists fully leverage structured and unstructured information captured by an ELN. The enterprise notebook lets scientists pull back individual experiments for reference, enabling cross-experiment analysis and reporting using complementary decision support tools. The critical capability to store structured and unstructured data extends notebook functionality, enabling scientists to answer many challenging questions they encounter in the course of R&D workflows. What information already exists? What data searches or experiments do I need to repeat? What can I do to make sure that my first experiment is successful? What can I learn from my colleagues’ experiments? What are the best results? How can my colleagues’ data help me reach a better decision or come up with a new approach?

Indeed, the enterprise notebook is changing the way scientists interact with information. Not only is data captured in an ELN fully searchable within and across experiments, but an ELN also can automatically provide third-party information to scientists. For example, without leaving their notebooks, scientists can look for commercial compounds or known synthesis pathways and populate the experiment details. As scientists draw molecules, their notebooks can automatically inform them about in-house and commercial availability.

Delivering convergence for lab software

Informatics providers today are partnering to deliver complete workflows that meet scientists’ needs. As previously mentioned, Symyx and Thermo Fisher Scientific are partnering to deliver customer-requested integration among LIMS, ELN and decision support software to improve the design, execution, analysis and reporting of laboratory experiments. Integrating Thermo Scientific Watson™ LIMS (and eventually other LIMS systems from Thermo) with an enterprise ELN and decision support software from Symyx will reduce or eliminate manual transcription and data manipulation that can cause costly schedule overruns, process errors and regulatory compliance breakdowns. A typical integrated workflow might involve setting up Watson DMPK study parameters in the ELN, transferring the DMPK setup to Watson LIMS, executing the study and analysis in Watson LIMS, and reporting the results back through the ELN or a decision support tool that enables cross-study comparison and analysis.

Delivering convergence for lab equipment

The enterprise ELN further improves the collection, analysis and reporting of experimental results by integrating with lab equipment such as balances and instruments, including HPLC, GC, LC/MS and NMR. Further convergence is achieved through integration with third-party applications such as spreadsheets, statistical analysis packages, kinetic modeling/data visualization tools, and chromatography and scientific data management software.

Integration is a two-way street. The ELN can push information out to laboratory equipment and pull it back into a digital repository, where it is searchable and sharable within and across project teams. For example, integration with balances accelerates high-volume sample analysis by enabling scientists to send weights directly from lab balances to the notebook. Other efficiency boosters include the ability to run daily balance checks from the notebook, retrieve up-to-date calibration information in the context of an experiment and update instrument status on the fly. The ELN further simplifies data entry with effective copy/paste, fill-down and import/export capabilities.

Delivering convergence for collaborative R&D

As R&D organizations expand partnerships and outsourcing, globally dispersed “virtual teams” are proliferating. The enterprise ELN facilitates collaboration with partners and contract research organizations by supporting

• Workflow coordination across geographic and business boundaries,
• The authoring of experiments by multiple scientists
• Data capture and data access across the globe and between business networks,
• Secure control of read/write access to experiment sections
• The agility to move between projects and
• Streamlined regulatory compliance.

The enterprise ELN also streamlines the exchange of information among virtual research teams by integrating with laboratory operational informatics (LOI) software and systems. Integrated LOI capabilities can streamline lab operations and improve R&D productivity by consolidating into the ELN workflow sample management and tracking (i.e., work request management), instrument management, inventory management and metrology. A work request system manages the submission, assignment, tracking and history of laboratory work orders, enabling the creation, routing, approval and completion of cross-site analytical requests. A materials system provides the ability to register compounds and search the inventory for laboratory materials and their properties. An equipment and metrology system manages laboratory hardware and tracking equipment status as well as calibration and maintenance histories.

The following simple example illustrates how the enterprise ELN with integrated LOI capabilities improves information capture, analysis and reporting for virtual teams. Scientist A creates a synthesis request in LOI using ELN messaging or a document workflow. Scientist B is able to access the experiment directly from his/her inbox. Scientist B checks out the experiment, executes the reaction, records the procedure (including searching for chemicals, linking to instruments and viewing instrument information in LOI) and creates results and conclusions. Scientist B can access any section of the experiment and add new sections during editing to capture relevant analytical data as needed. In the meantime, scientist A, at the requesting organization, easily can monitor scientist B’s progress. When the experiment is complete, scientist B checks it back into the ELN repository and routes it back to the requester. The requester then can take advantage of configurable ELN reporting templates to create and distribute required reports, further supporting project team collaboration. The result is more efficient scheduling and notification, real-time access to results, and instant feedback and adjustments— culminating in better results, decreased costs and improved productivity.

The enterprise ELN is not the hub for capturing every piece of laboratory data. There always will be a need for LIMS, SDMS and other informatics repositories that are specifically designed to handle structured or unstructured information. Today’s enterprise ELN is, however, leading the way into a new era of converging functionalities. Information-driven R&D—a single, unified data management solution—is converging around the fourthgeneration ELN described by Elliott to help scientists execute experiment workflows and manage structured and unstructured information more efficiently. This laboratory informatics strategy also is automating and better integrating lab workflows around compound registration, decision support, lab logistics and even LIMS/ SDMS capabilities. Finally, it is improving R&D decision making by enabling scientists to access rich, high-value scientific content in the context of their workflows.

Looking to the future, Symyx believes that the fifthgeneration ELN will extend fourth-generation capabilities with hosted services, placing the ELN “in the cloud,” so that virtual project teams distributed across disparate geographic locations and business boundaries easily can share information and communicate using a common notebook. The fifth-generation ELN will help more companies benefit from having an ELN, reduce cost of ownership and enhance operational agility—enabling organizations to expand and contract R&D resources and IT infrastructure as necessary to meet changing business needs.

References

1. Michael H. Elliott, “What You Should Know Before Selecting an ELN,” ScientificComputing.com, Spring 2009