Application Refactoring · IT Operations Automation · Real-time Analytics · System Migration
Modern ML offers a complementary approach to predicting costly experimental outcomes and biasing molecular design toward more drug-like properties by learning historic experimental trends, which are published in scientific literature, patent applications, and proprietary datasets. “It’s expensive and time-consuming to synthesize and test compounds,” says Dan Price, vice president of Computational Chemistry and Struct
Reported impact
Nimbus has successfully automated a suite of ML models to predict different aspects of drug exposure and safety, which are actively influencing design across all discovery projects. The predictions provided through LiveDesign democratize the technology across all Nimbus scientists, regardless of discipline or technical background. However, the infrastructure isn’t specific to LiveDesign.
With established expertise in computational chemistry and machine learning (ML), Nimbus Therapeutics (Nimbus) sought to further accelerate its drug discovery engine by automating the operational aspects of ML (MLOps). Minimizing manual intervention in model training and deployment would help its scientists to focus on Nimbus’s core mission: the design of breakthrough medicines. Learn how computational chemists at Nimbus Therapeutics built an agile and effective ML pipeline using Amazon SageMaker. Nimbus turned to Amazon Web Services (AWS) to use Amazon SageMaker , which lets organizations build, train, and deploy ML models for any use case with fully managed infrastructure, tools, and workflows. To orchestrate communication with LiveDesign, the team used Amazon API Gateway , a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure APIs at any scale. Amazon API Gateway used the serverless
processing runs on Amazon SageMaker, AWS Lambda; and outputs are exposed through Amazon API Gateway for the cloud migration workload.
Many data sources in the pharmaceutical domain are small, inherently heterogenous, and missing key structured metadata.
Modern ML offers a complementary approach to predicting costly experimental outcomes and biasing molecular design toward more drug-like properties by learning historic experimental trends, which are published in scientific literature, patent applications, and proprietary datasets. “It’s expensive and time-consuming to synthesize and test compounds,” says Dan Price, vice president of Computational Chemistry and Struct
Nimbus considered it critical to codify domain-expert rules for metadata extraction and criteria for inclusion and exclusion to sustainably deliver performant models.
The pipeline would provide near real-time predictions through Schrödinger’s LiveDesign, an application where new molecular designs are captured and evaluated.
The team at Nimbus imagined an automated pipeline that would regularly curate data from a variety of sources and retrain and redeploy ML models.
The team needed to quickly demonstrate whether the available data, combined with its curation and modeling strategies, could meaningfully impact programs and warrant a larger investment in time and resources.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
The official website footer lists '22 Boston Wharf Road, Floor 9, Boston, MA 02210 USA,' supporting a United States headquarters.
The official contact page lists "22 Boston Wharf Road, Floor 9 Boston, MA 02210 USA."
LinkedIn's Organization JSON-LD lists addressLocality "Boston", addressCountry "US", and numberOfEmployees value 103.
Related
AWS Sample Mar 5, 2024 Mercedes-Benz Consulting With millions of internal documents to search through, getting relevant answers quickly was a standing challenge for employees. Additionally, if an employee left the company or a colleague with specific knowledge was… AWS Glue · AWS Lambda · Amazon OpenSearch Service AWS Sample Feb 1, 2024 Coca-Cola Andina Coca-Cola Andina has been using AWS since 2020, and in 2021, the company built a data lake on AWS and began migrating all its data from its on-premises storage to the cloud. The migration required a scalable solution… Instead of updating once per day, Thanos updates its data every 15 minutes. “In Thanos, we have information… AWS Lambda · Amazon RDS · Amazon S3 AWS Sample Oct 29, 2024 Vertex Pharmaceuticals After designing an experiment, they manually trained the base ML model about what they hoped to identify from the images, labeling each area of interest by hand. In the legacy system architecture, all scientists shared… 11x accelerated image segmentation AWS Fargate · AWS Lambda · AWS Step Functions 