Five Ways a LIMS Accelerates Next-Generation Sequencing Research

Transcription

1 Five Ways a LIMS Accelerates Next-Generation Sequencing Research An Essential Business Tool for Genomics Labs Just over 10 years ago, genomics was an elite discipline, practiced by select labs associated mainly with marquee efforts such as the Human Genome Project. How things change! Today, the technologies developed to support early genomics research have developed primarily for pharmaceutical, chemical, or clinical labs they have not been built specifically to address the needs of basic biological research or the unprecedented throughput and experimental complexity associated with next-generation sequencing. entered the mainstream. Labs everywhere can adopt high throughput, next generation sequencing. But in doing so, they often realize that the ability to sequence entire genomes in a matter of weeks means that bottlenecks impeding research aren t cost or lack of technology, but the ability to rapidly and accurately interpret and exploit sequencing data. This paper will review five ways that a robust lab information management system (LIMS) accelerates modern genomics and next-generation sequencing. It will focus particularly on the business reasons that make a well-built, commercial lab information management software system designed specifically for sequencing is a smarter investment than Since the 1970s, research labs have turned to laboratory building a system from scratch. information and data management systems to streamline data throughput and reporting. The first lab information management systems (LIMS) were homegrown, and commercial systems emerged in Today, LIMS is a mature class of life sciences software encompassing well over thirty open-source and commercial products capable of integrating instrumentation; automating workflows; and managing samples, users, and lab standards. Most of these systems, however, were Page 1 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

2 In those workflows, where everything is mixed, you ll never know if you ve done something wrong until you send a customer the wrong data and you can t afford to make that type of mistake even once. On this volume of data, the only way you can guarantee you are getting things right is through a LIMS. Kenny Beckman, Biomedical Genomics Center, University of Minnesota Systems specifically built for genomics and nextgeneration sequencing can help investigators quickly set up a lab; manage the samples, data, technologies, staff, and workflows essential to modern sequencing research; Today, businesses are competing to sequence an individual s entire genome for about $10,000, and experts are claiming that the thousand-dollar genome is within sight. conduct speedier, more collaborative research; and publish findings sooner. Commercial Lab Information Management Software Accelerates Next Gen Sequencing 1. A LIMS Provides Scale to Keep Cutting-edge Labs on Pace in a Rapidly Evolving Discipline The pace of change in the next generation sequencing market is staggering. In just a single decade, sequencing costs have dropped and data volume has increased by several orders of magnitude. The Human Genome Project This pace has been made possible by technology. As Batley and Edwards point out, genome sequencing has been revolutionized since the introduction of commercial pyrosequencing and the Roche 454 GS20, which could produce 20 Mbp in a single run. The three primary manufacturers of next-gen sequencing instrumentation (Life Technologies, Illumina, and Roche) continue to push performance. In 2008, the SOLiD TM system by Applied Biosystems was capable of handling 20 gigabases per run; Titanium methods released by Roche in 2008 have made possible read lengths of over 400 bp for a throughput of between million base pairs per sequencing run; and the Illumina GAII boasted a throughput of up to 6.5 Gb per day for a 2 x 100 bp run and was capable of reading up to 320 million clusters and up to 640 paired-ends per day. took 10 years and $2-3 billion to complete. Page 2 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

3 The only other technology where costs and processing Figure 1 speed have dropped as fast as in DNA sequencing has been in the computer processor industry. And even the well-known Moore s Law (the suggestion first made in Gordon Moore s 1965 paper that computer processing speed will double every 2 years) is looking like a flat line compared to DNA sequencing (see Figure 1). The speed of change in sequencing technology is also evidenced by tracking the number of genomes entered into GenBank each year (see Figure 2). It s important to note that when this chart was published in 2009, over 5,340 additional projects were underway and therefore not reflected in this data. More impressively, the 1000 Genomes Project, the first large project to capitalize on next-generation sequencing technologies, deposited twice as much raw sequencing data into the GenBank archives in its first six months Historical trends in storage prices versus DNA sequencing costs. The blue squares describe the historic cost of disk prices in megabytes per US dollar. The long-term trend (blue line, which is a straight line here because the plot is logarithmic) shows exponential growth in storage per dollar with a doubling time of roughly 1.5 years. The cost of DNA sequencing, expressed in base pairs per dollar, is shown by the red triangles. It follows an exponential curve (yellow line) with a doubling time slightly slower than disk storage until 2004, when next generation sequencing (NGS) causes an inflection in the curve to a doubling time of less than 6 months (red line). These curves are not corrected for inflation or for the fully loaded cost of sequencing and disk storage, which would include personnel costs, depreciation and overhead. of operation as had been deposited into GenBank in the 30 years since its inception. The implication for the modern sequencing lab is that it s Figure 2 quite possible to produce more data than can be efficiently and accurately managed and analyzed. Traditional lab and data management infrastructures which rely heavily on spreadsheets, s, and paper lab notebooks to manage and trace samples and record and report data are at best inefficient and at worst completely fail under this onslaught of sequencing data. A 2010 J.P. Morgan survey confirms this: Lab directors cite data storage, data management, Number of genomes entered into GenBank by year as of September and informatics as the biggest collective hurdle to expanding next-generation sequencing operations. Page 3 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

4 The very nature of next-generation sequencing requires iterative work that depends heavily on contextual information about library preparation and sequencing methods and research goals and parameters before analysis can even begin. Scientists and bioinformaticians need to be able to rapidly access and gather the right experimental information to answer such questions as 2. A LIMS Can Get Labs Up and Running Faster Next-generation sequencing is nearly synonymous with speed. And the expectation of speed extends to the initiation, operation, and management of a lab. Grants have deadlines, and meeting those deadlines requires labs to be able to quickly ramp up to accommodate What are the research goals? What is the taxonomy of the sample? Where is all the disparate data located? next-generation sequencing technologies and workflows so that they can meet expected research goals and publish research results faster. How was the data generated? Which analysis tools should I use, and which algorithms and parameters should be applied? What quality threshold should be applied to the dataset? In this fast-paced research environment, labs simply don t have time to build infrastructure software such as a lab information management system. Such a project can easily take between months an eternity in a discipline where cycle times are doubling faster than Ultimately, too much information is the same thing as no information. Because data and information can only be useful if it is delivered in a form that can be rapidly interpreted and acted upon, next-generation sequencing labs must aim to completely eliminate manual, one-at-atime processes. A robust lab information management Moore s Law. Consider also that once the initial system is in place, it must be maintained and updated to accommodate new research objectives, protocols, workflows, and technologies. Can a research center dedicated to modern genomics afford to also become an expert in sequencing research informatics? system centralizes sample management and workflows. Lab processes, protocols, and analyses can be specified, defined, integrated, automated, and pipelined to create an end-to-end lab workflow. Most importantly, labs can readily swap out methodologies, parameters, and instrumentation as technologies and workflows inevitably evolve. This scalability leaves labs well positioned to provide best-in-class service to collaborators and customers now and in the future. Implementing commercial lab information management software designed specifically for genomics and sequencing can significantly shorten timeframes between lab set up and production. In as little as two or three months, a commercial system can be installed, tracking samples, and providing real value to researchers. With a commercial system, lab personnel receive support and training to ensure that the system is implemented appropriately to match their Page 4 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

5 Figure 3 Figure 4 PHYSICAL SAMPLE RECEIVED ISOLATE NUCLEIC ACID PREPARE LIBRARY AMPLIFY & SEQUENCE ANALYZE DATA VALIDATE RESULTS Sample and clinical information Imported sample QC UV spectrophotometer results Alignment result file Geneus LIMS SAMPLE DATA UPLOADED SAMPLE PREPARATION AND WORKFLOW INTEGRATED SAMPLE TRACKING AUTOMATED DATA COLLECTION RESULTS SHARING Project Management and Collaboration Typical sequencing workflow. (and hard to trace) collection of s, conference calls, and cryptically labeled samples received from collaborators that typically kick off a sequencing research project. Within the sequencing lab, the commercial lab information management system provides end-to-end workflow management to ensure that samples are easy to trace and tied directly to relevant experimental details (see Figure 3). Additionally, interfaces enable lab staff to check on project progress in real-time and rapidly publish End-to-end next-generation sequencing sample and data genealogy. specific results and reports to external collaborators or customers. Robust lab information management software gets a lab up and running quickly and keeps it running specific workflows and processes. Training will often focus on system configuration and the sophisticated application program interfaces (APIs) that enable programmers and bioinformaticians to customize the system s capabilities (with rapid scripting) to ensure the system will meet future needs. A commercial lab information management system should also specify best practices and common workflow steps that keep labs on task and producing accurate, traceable data from the very first project. A self-service web portal captures, organizes, and centralizes sample submission, eliminating the need to sort through the often confusing and operating with best-in-class efficiency. 3. A LIMS Improves the Integrity of Results with Unmatched Sample Traceability The unprecedented throughput enabled by modern sequencing instruments and techniques such as multiplexing make it impossible for labs to rely on manual workflows and disorganized, disparate desktop software or white boards to track samples. Clear sample traceability from the moment samples enter the lab to the point when the results are Page 5 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

6 ready to be reported and published is a must have for sequencing labs (see Figure 4). In fact, many of the most prestigious grants and research projects require labs to be able to guarantee sample traceability it s essential when dealing with the often limited DNA supplies associated with certain clinical sample cohorts. A commercial lab information management system designed specifically for sequencing can address some of the more onerous aspects of tracking samples, including Correlating wet lab sample information and experimental data one-to-one. The ability to linked results to original sample data enables reliable in silico work and downstream analysis. Validating and setting up multiplexed sequencing runs. Any mismatch between a DNA barcode, sample, and the loading pattern on a sequencer can invalidate the results of an entire run, but tracking this information in a spreadsheet often leads to transcription errors. A lab information management system can be configured to automatically generate sample sheets or run definition files for sequencing instruments, simplifying and speeding run-times. Organizing, labeling, and tracking samples. The data volumes and experimental complexity associated with next-generation sequencing can tax information associated with analyzed samples. Lab information management software specifically designed for sequencing can accurately track individual components of complex sample pools from the initiation of a project through to its conclusion so that labs can fully benefit from their investments in sample management infrastructure. Making information easy to store, manage, and access. Too often, crucial sample and experimental information is spread across paper lab notebooks, s, electronic documents, and spreadsheets, which may in turn be stored on different workstations or lab instrumentation. Lab information management software for sequencing centralizes this information so that sample preparation details, workflows, QC results, associated experimental details, and data analysis results remain accessible and retrievable years after projects are complete. By implementing a best-in-class, commercial lab information management system, labs can consistently deliver accurate, high quality results experiment after experiment. The ability to prevent sample mix-ups and accurately trace data back to the original samples saves time and money and helps labs guarantee and defend their reputations. traditional barcoding and indexing systems, which can struggle to reliably trace samples back through complex sample pools to the specific phenotypic Page 6 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

7 Why One Lab Purchased a Commercial LIMS In a presentation delivered at the XGen Congress Meeting in San Diego in March 2010, Ms. Johanna Swanson, previously a scientific programmer from the Northwest Genomics Center at the University of Washington, explained some of the business drivers behind this lab s decision to purchase a commercial LIMS for next-generation sequencing. We decided that we really needed a LIMS and with the short time period that we had to scale up, we decided to purchase a commercial LIMS, said Swanson. Various vendors were reviewed, and we actually decided to purchase The GenoLogics LIMS system. There were several reasons that the center considered a LIMS important. First, Swanson noted that the center would be analyzing DNA from well phenotyped heart and lung cohorts such as the Framingham Heart Study and the Women s Health Initiative. These are very precious DNAs and some are available in limited quantities, Swanson explained. Getting [the samples] into our system and properly handling them is crucial. We had to be able to manage these samples and know we wouldn t lose any of [them]. We had to be able to trace them from the moment they came in the lab, until the day we were done, and then serve [the data] back to the investigators. 4. A LIMS Manages Workflows and Internal Communications to Improve Lab Efficiency The number of next-generation sequencing instruments and sequencing strategies that can be applied to projects is enough to overwhelm even the most experienced laboratory personnel. It is quite possible in modern sequencing for lab managers, technicians, analysts, and bioinformaticians to spend the bulk of their day performing repetitive, administrative tasks associated with organizing and managing the data generated by their labs. Busy labs can also struggle to ensure that samples received from clients and collaborators are appropriately labeled and that all vital experimental context is passed on efficiently and accurately to experimentalists. Lack of workflow clarity can lead to miscommunications and inefficient sample handoffs between groups that have the potential to waste precious samples, time, and reagents. Moreover, inefficiencies can negatively impact a lab s credibility, particularly if they result in delays in achieving critical milestones for funded research programs. A LIMS designed for sequencing centralizes all communications and workflow tasks. It also improves lab efficiency by eliminating many of the manual, tedious, repetitive tasks associated with day-to-day lab operations. A configurable user interface lets lab personnel quickly and easily link together common lab processes to create workflows that mirror various protocols within the lab, such as those associated with sample preparation or data analysis. Time consuming steps, such as representing and Page 7 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

8 Why One Lab Purchased a Commercial LIMS Workflow management was another key driver behind the decision to purchase a system. What we needed was a LIMS that we could scale and adapt to changing workflows, Swanson said. We had to be able to track the tubes, the plates, and everything that comes through the lab. We had to be able to label everything and give them unique IDs so we wouldn t have missing names or confusion on the plates. Also we needed to tracking library pooling interactions, demultiplexing pooled samples, generating sample sheets for sequencers, or tracking critical quality control data and resulting pass/fail decisions, can be automated and even assigned to specific personnel. Reporting and communications are also streamlined, because the system tracks and centralizes all of the information associated with sequencing projects. integrate robotics instruments with our lab and integrate with our plates and tubes. Lastly, we needed to keep track of key measurements that are made on our system such as PicoGreen assay, and all the steps that were done along the way. Workflow management ensures sample traceability, standardizes best laboratory practices, and ensures that personnel are always working on high priority tasks. When codified in commercial lab information management software designed specifically for sequencing, workflow management Finally, the information had to be accessible when needed, either during research or at a later time. We needed to be able to find this information months or years later information which is sometimes spread across notebooks, Excel spreadsheets, and even Google docs, said Swanson. It gets harder and harder and harder to find stuff and recover it. Especially if texts were sent to the lab, or legacy information has been passed on. enables labs to focus on their core research objectives and establish their reputations as best-in-class facilities. We had to feel confident that the system we chose for our nextgeneration sequencing center could be deployed on an accelerated timeline, while maintaining the ability to adapt to constantly evolving next-generation sequencing technology and workflows. We chose the GenoLogics LIMS, and we have been processing samples for several months with no interruptions. Dr. Mark Rieder, Principal Investigator, Northwest Genomics Center, University of Washington Page 8 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

9 5. A LIMS Helps Labs Publish Faster The ultimate goal for principal investigators in academic labs and research centers is publication. Two factors speed publication: 1) The speed at which experiments are conducted, analyzed and reported; and 2) the quality of the data that is produced. Both of these factors are supported in a scalable, adaptable lab information management system that centralizes sample information and standardizes workflows. Faster, automated lab procedures lead to faster and more efficient experiments and more reliable results. And when reliable results are received quickly, research findings can be published sooner. But to truly enable the publication step, any informatics solution must be as good at helping scientists get data out of the system as it was at helping scientists get the data in. A lab information management system maximizes instrument efficiency; automates critical data capture, quality control, and analysis pipelines; and keeps samples flowing through the lab while tracking associated data and keeping information easily accessible. It should also offer a variety of options for accessing stored data, from ad hoc searches to sophisticated reporting frameworks that can also be used to create custom queries and reports. A robust API can also help scientists and internal programmers extract data into the format needed for downstream analysis. And once that analysis is complete, scientists can choose any number of ways to summarize and communicate results. Because all project-related the same system to generate both operational reports for executive management, PIs, or clients and experimental reports to, for instance, compare key experimental parameters (such as PCR cycles) with downstream quality metrics. Internal and external reporting can be streamlined through the use of a site-branded web portal that enables lab staff and their customers to rapidly and securely share and collaborate in near real-time. Commercial Lab Information Management Software is Essential for Modern Sequencing Facilities The unprecedented data volumes and experimental complexity associated with modern sequencing make a commercial LIMS essential to labs seeking to establish their reputations as best-in-class facilities capable of high powered research. A commercial lab information management system provides the scale labs need to adapt to the changing technologies and methodologies in this field. It enables them to quickly ramp up facilities to produce results. It offers the sample traceability funding agencies demand on premiere sequencing projects. And it provides workflow management, data analysis, and reporting tools to ensure that labs run efficiently and collaboratively. All of these capabilities can have a tremendous impact on labs, enabling them to information is stored in the system, lab managers can use Page 9 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.

10 genologics.com GenoLogics Life Sciences Software 4464 Markham Street, Suite 2302, Victoria, BC, Canada V8Z 7X8 Toll free Main Fax Reduce costs by improving reagent use and eliminating the experimental errors and doubt that lead to repeated experiments Get a new sequencing center up, running, and processing samples in just a few months Accommodate the constant (sometimes daily) changes that are de rigueur in labs conducting next-generation sequencing Produce better quality results consistently, speeding the publication process and impact factor Communicate better and more effectively with customers and collaborators Focus more on research by eliminating many manual tasks and implementing, best-practice-based workflows Conduct more research using current staffing levels 1 Gibbon, G. A Brief History of LIMS. Laboratory Automation and Information Management 1996, 32, Batley, J.; Edwards, D. Genome sequences data: Management, storage, and visualization. BioTechniques 2009, 46, Karow, J. ABI Announce SOLiD, 454 Launches Titanium, Illumina Sets Specs for End announces-solid launchestitanium-illumina-sets-specs-end- %E2%80%9808 (accessed December 8, 2010). 4 Media / Media Releases. med_dia_ htmroche (accessed November 25, 2010). 5 Systems / Genome Analyzer IIx. analyzer_iix.ilmn (accessed November 25, 2010). 6 Stein, L. D. The case for cloud computing in genome informatics. Genome Biology 2010, 11, Martinez, D. A.; Nelson, M. A. The Next Generation Becomes the Now Generation: Terabyte has become the New Gigabyte. PLoS Genetics 2010, 6. 8 Stein, L. D. The case for cloud computing in genome informatics. Genome Biology 2010, 11, J.P. Morgan Equity Research. Next Generation Sequencing Survey and instrumentation, reducing facility overhead cost Build a reputation as a best-in-class facility, attracting more business and securing high profile studies and additional grant money Page 10 GenoLogics Life Sciences Software, Inc, an Illumina company 2016.