Accelerating Discovery with On-demand LIMS

Published: January 7, 2011  |  Source: scientificcomputing.com
116

Cutting-edge spinal cord research team eliminates critical bottlenecks

The Miami Project to Cure Paralysis, part of University of Miami, Florida, comprises a dedicated team of scientists who are researching a variety of treatment strategies for individuals with spinal cord injury. Dr. John Bixby and I head one of the Project’s laboratories, The Laboratory for Axon Growth and Guidance. At our laboratory, also known as the LemBix Laboratory, we conduct high-content screening (HCS) of neurons with the aim of finding new or improved treatments for spinal cord injury and other neurological disorders. HCS offers an ideal way to identify genes, molecular pathways and, ultimately, drugs that can promote the regeneration of central nervous system nerve cells.

Our team focuses on understanding how nerve cells extend their long processes, called axons, to their targets, why young neurons do it so successfully and why older neurons have lost their ability to do this. This is an immensely complex problem, and it is clear that any single molecular signal can not mediate this.

The challenge
The massive data sets available from the various genome projects combined with sophisticated image analysis and automated microscopes have created an opportunity to revolutionize the study of the development and function of the nervous system. We have developed methods that allow the testing of hundreds of genes in hundreds of thousands of neurons each week and obtain quantitative data on scores of parameters about cell morphology and gene expression from each neuron. We accomplish this by acquiring thousands of images within hours and analyzing them in real time. This high throughput capability allows us to answer questions about development, regeneration and synapse formation using systems biology approaches.

For several years, we have used paper-based worksheets and manual notebooks. Because of the nature of the work, using different compound libraries and gene libraries, we were overwhelmed with all the experimental details and data we had to oversee. Furthermore, we were moving from conducting traditional academic research activities, where it was easier to keep track of information manually in notebooks, to a different kind of workload that involved more complicated processes and more people. Keeping track of what each one of us was doing through the workflow of preparing the cells and perturbations, staining and analysis was a challenge. We realized we needed an automated electronic data management system that could easily be used by anyone who needed to enter and retrieve information.

The solution
In 2008, after narrowing our choices among three laboratory information management system (LIMS) vendors, we requested detailed demonstrations and trial installations to better determine the most appropriate solution for our specific research needs. Following an evaluation process, we selected Thermo Scientific Nautilus, due to its ease of developing workflows and sophisticated plate handling. Our confidence in the solution also was enhanced by on-site academic user experience, as another department at the University of Miami (Genomics) also employed the solution.

For cell culture and screening, we use complicated media and many different types of reagents. Workflow, reagent and stock tracking are imperative, since experiments often fail due to problems with the reagents or stocks. If an experiment fails, we need to understand why, and a LIMS greatly facilitates this. The LIMS keeps track of stocks and reagents throughout the different workflows and facilitates our ability to identify which reagents are worth progressing or need to be discarded. The system improves our laboratory’s productivity by providing improved quality control and more efficient triaging of bad reagents.

Implementation and results
We installed an enterprise-level LIMS in 2008, using an Oracle database on the central University servers. The University’s IT infrastructure was charged with the task of monitoring installation and processes related to validation, system upgrades, managing licenses, hardware requirements and maintenance issues.

In 2010, we beta-tested an on-demand model after we had been struggling with in-house maintenance issues associated with the University’s servers which, until then, could only be maintained by the University IT staff, sometimes taking weeks for an issue to be resolved. The on-demand model, which uses communication with the LIMS via Web browsers, simultaneously eliminated the platform-specific requirement of a client-server solution. As a result, any computer in the lab, whether PC, Mac or Linux, could now connect to the LIMS. We decided to migrate to an on-demand LIMS in order to remove the IT support and OS bottlenecks.

We had already invested in workflow development and stock and reagent tracking for our Nautilus LIMS, and the ability to very rapidly push this over to a SaaS system was very appealing. Data were quickly transferred from the client-based system to the on-demand solution. Within one afternoon, we moved from the existing client/server system to the on-demand model with minimal issues. Our team no longer has to worry about the servers or maintain the Oracle database, and updates are automatic. The on-demand model has removed the hassles of back-end management, while providing a high level of security and confidence.

On-demand software requires no hardware purchases, software installations, network setup or ongoing system upgrades from us. The solution is delivered via an application service provider (ASP) on a recurring subscription model hosted from a secure datacenter. Database tuning, backups, critical operating system and LIMS updates are all managed and scheduled by the vendor. In addition, the hosted LIMS environment is very secure. Data centers are monitored by closed-circuit television (CCTV) and onsite guards and use biometric two-factor authentication to ensure no unauthorized personnel can obtain access. Data are backed-up to separate geographic locations to ensure data safety.

For our laboratory, making the switch from a client/server environment to a Web system was seamless. We can now keep track of the reagents used in our experiments, and we are no longer dependant on the PC platform which the client/server system utilized. Our team can access the database from anywhere via any computer or laptop, increasing productivity time as we spend less time configuring workflows.

Conclusions
With productivity and efficiency in our mind, we have equipped our laboratory with the technologies and infrastructure that allow us to deliver consistent support to ensure the quality of our workflows. To deliver this cutting-edge research, we have adopted LIMS-on-Demand, meaning that we no longer have to rely on the university’s IT team to maintain our servers; we no longer have database licensing issues; and we can easily use any computer to interface with the LIMS. In other words, we can focus on our core business of finding a cure for spinal cord injury.

The on-demand LIMS solution has provided us with several benefits, including all the advantages of a fully installed LIMS, such as built-in workflows; a powerful database; the ability to capture, store and analyze lab data, monitor resources and integrate with instrumentation; and reporting templates. All this can be achieved while minimizing the need for IT resources and eliminating expensive hardware and software. The solution offers data integrity, guaranteed system availability and performance, allowing our team to control our data, time and budget. Only with LIMS-on-Demand can we continue to identify and treat novel mechanisms of injury following brain and spinal cord injury and efficiently translate these discoveries to the clinic.

Vance Lemmon, Ph.D.

Vance Lemmon is Walter G. Ross Distinguished Chair in Developmental Neuroscience, Professor of Neurological Surgery at The Miami Project to Cure Paralysis Center for Computational Sciences, John P. Hussman Institute for Human Genomics, University of Miami Miller School of Medicine.