Instrument Data Lake and Automatic Experiment Data Entry into LIMS

← All case studies

Overview

Jounce Therapeutics has a bold mission to transform the treatment of cancer. They develop therapies that enable the immune system to attack tumors, bringing long lasting benefits to patients. Their unique approach prioritizes targets and identifies related biomarkers to match the right immunotherapy to the right patients. But Jounce’s approach to developing cancer therapies doesn’t stop with the science; it permeates every aspect of their business, such as core values, operations, and how they manage one of their key assets, their research data. The way Jounce manages research data, including raw instrument data, has evolved over the years - from simple file-based approaches to leveraging on-premises database driven technologies to modern Software-as-a-Service (SaaS) and Platform-as-a-Service (PaaS) cloud offerings.

Jounce engaged with TetraScience to do just that - automatically harmonize and centralize their R&D data using the TetraScience Data Integration Platform. The result from integrating the first set of instruments exceeded expectations – data is now centralized, easily searchable, and automatically moves to their LIMS without any manual effort.

Challenges

In order to achieve their goals, Jounce needed to upgrade their approach to data management and search. Jounce used multiple legacy systems for managing research data including Windows file servers, Biologics Registry, LIMS, and ELN. This complex environment made it challenging for the scientists to curate, analyze, and retrieve data.

  1. Data Management: Raw instrument data was stored in native and exported formats on a Windows file server. In some cases, scripts copied data to the file server but a large number of instruments required the users to manually copy data from the instrument to the file server. Scientists would then analyze data and update the Biologics Registry and LIMS systems manually as new data was available. The structure of the Windows file server was user dependent. This made it challenging to find data from experiments run far in the past and caused scientists to re-run experiments.
  2. Wasted Time: Scientists should be spending their time developing therapeutics. The combination of all the manual steps outlined in Challenge #1 made it time consuming to track data properly. Jounce’s scientists (along with nearly every other life sciences R&D scientist), spend at least 10% of their time manually entering result data in the LIMS.
  3. Data Integration: As mentioned above, Jounce leverages a number of best-in-class solutions for Biologics data management. While some systems were integrated to copy and move data, other systems could not be integrated or had limitations in their APIs. The Jounce Research Informatics team struggled to ensure that data was being leveraged properly using the various data management solutions.
"Integrating lab instrument data with data management tools is a challenge faced by many scientific organizations, including Jounce. Leveraging the TetraScience platform enabled Jounce to implement a vendor supported, LIMS agnostic solution for this problem, thus alleviating the need for custom integration solutions to be developed, maintained, and supported in-house."

- Shannon McCabe, Senior Business Analyst, Research Informatics, Jounce Therapeutics

  1. Human Error: The manual nature of the data capture process into the LIMS introduced ample opportunity for error. It is easy to forget to load an image, to incorrectly enter a data point, to mislabel a file, etc. Let’s consider an example where a scientist needs to archive a chromatogram image. After producing a chromatogram on a GE AKTA instrument, the scientists are required to capture a screenshot or download the chromatogram image and then copy it into the LIMS. Hopefully, the scientist uses the standard file naming convention, places the file in the correct location, includes all the correct tags, and spells everything correctly

After evaluating several vendors, we selected TetraScience because of their innovative technology platform, their instrument vendor agnostic approach, their team, and their vision for a Digital Lab. We felt that they had the most mature and scalable platform in this market.

- Bill Goode, VP of IT, Informatics, and Operations at Jounce Therapeutics

Solution

Jounce quickly realized that building individual custom data pipelines from each instrument in the lab to the LIMS was not an efficient or scalable option for their R&D IT team. They chose TetraScience as a vendor-neutral platform that had built data connectors to automatically extract data from most of the popular instruments in their labs, such as GE AKTA UNICORN, Agilent Chemstation, Azure Gel Imager, Molecular Devices Spectramax Platereader, MSD, ForteBio Octet, etc.

Upon extraction, the data is parsed into a standardized format and uploaded to an AWS hosted cloud in a matter of milliseconds. The data is also automatically enriched with metadata like user, system, sample & plate info, method, and more. Now that the TetraScience Data Integration Platform is in production, all a Jounce scientist has to do to process their AKTA data is to run the experiment. The chromatogram image is then automatically downloaded, named, labeled, tagged, enriched, and delivered to the LIMS.

Since the data connectors and pipelines are automated, there is no extra time spent and no opportunity for human error. The data is harmonized and centralized in the cloud-based data lake, and files are easy to search and query. The integration between the TetraScience pipeline and the LIMS allows the scientists to associate key biologics identifiers with the data, automatically simplifying the tracking of protein production progress.

We were able to work with TetraScience to customize each of our workflows and define what events would occur and what constitutes an error condition. We are now able to view data capture and transformation pipelines in real-time and quickly investigate any issues.

- Todd Rowe, Associate Director of Research Informatics at Jounce Therapeutics

If there is an error, the Jounce team receives an alert which they can resolve remotely. The TetraScience Customer Success team is ready to spring into action to address issues, and the Product team is continuously adding enhancements and new data connectors.

Outcome

Jounce scientists can now easily access their experimental data - they run an experiment and the data is automatically captured, cleaned, formatted, and delivered both to the LIMS and the Biologics Registry, saving time and improving data integrity. Scientists can also easily query the data in the instrument data lake, avoiding re-work and enabling analytics + insights.

The Research Informatics team sleeps easy because they know their experimental data is secure and accurate. Most importantly, Jounce is moving more rapidly towards their mission of developing novel cancer therapeutics to help patients.

"Keeping team members from a variety of functions updated as data is generated is difficult and time-consuming. The TetraScience Data Integration Platform allows our instruments to share data as it becomes available, allowing us to focus on problem-solving quickly across functions in order to push our fast-paced workflows."

- Karl Wetterhorn, Scientist, PDTO at Jounce Therapeutics

"Jounce had been storing instrument data for years using legacy technologies that made managing data tedious and required manual effort to input the data into our LIMS.

We engaged the TetraScience team to upgrade our approach by leveraging their digital platform using modern data integration and search technologies.”

- Bill Goode, VP of IT, Informatics, and Operations at Jounce Therapeutics

Data integration solutions

Tetrascience Logo

Make your R&D Lab Data Accessible and Actionable

Data integration platform
Lab monitoring application
File converter application
Waters Empower data science link application
connectors
Terms and Conditions