The Challenge
Data at Resilience was siloed across multiple instruments and sites, making access to critical information slow and complex. With over 100 instruments and the need for Next-Generation Sequencing (NGS) runs taking 7-10 weeks, the lack of centralized data access hindered research progress. Additionally, multiple file versions and uncertain data lineage complicated data management, causing inefficiencies in finding and filing accurate information.
The Solution
Quilt provided Resilience with a transformative approach to managing and accessing scientific data by leveraging Quilt Packages, Amazon S3 integration, and advanced version control mechanisms. Through Quilt’s seamless interface, Resilience scientists were able to browse millions of files either programmatically or through the web app, significantly improving their ability to retrieve and work with data without needing direct access to research instruments. Quilt’s data packages allowed for deep-indexing of content, enabling free-text search across millions of files with results returned instantaneously.
Quilt provides Resilience with the optimal Scientific Data Management System that enables one-click access to data with a snappy and intuitive GUI. The contents of our data are deep-indexed allowing us to free-text search millions of files, with the results returned instantaneously. The Quilt API allows us to automatically post the raw data links back to our electronic notebook for a given experiment ensuring data traceability and capturing precious intellectual property.
- Brian McNatt, Director of Digital R&D at Resilience
One of the key features that Quilt introduced was automated data ingestion from instruments to scientists. The API allowed Resilience to automatically post raw data links back to their electronic notebooks for experiments, ensuring data traceability and preserving intellectual property. Additionally, Quilt’s version control and immutability features created an audit trail for datasets as they moved through research, analysis, and downstream processes, including GXP and manufacturing stages, ultimately preparing the data for regulatory submission.
Quilt allows us to easily browse through millions of files stemming from our research sites, either programmatically or through the web app. It also allows us to frontend the data to users, without compromising on the key engineering efforts behind the scenes to provide file version control, immutability, and organizational efficiency. Leveraging Quilt Packages, users may also add custom analytics & visualizations and collaborate seamlessly to create well-structured data products shared across the company.
- Adam Mendez, Head of Data Engineering at Resilience
Quilt's integration with Amazon Web Services (AWS) provided a secure and scalable solution, ensuring cost-effective storage and collaboration across Resilience's vast data lake. Scientists could now share well-structured data products and collaborate in real time, closing the gap between lab teams and data science teams. By reducing manual effort and ensuring data accuracy, Quilt enabled Resilience to scale their biomanufacturing operations, supporting the rapid development of novel therapies
The Results
- 200+ scientists onboarded within the first month
- Data access time reduced from weeks to 5 minutes
- NGS run times reduced to under 1 hour
- A comprehensive audit trail for datasets across research, analysis, and manufacturing
- Improved collaboration between lab and data science teams
