Discovery and Pre-Clinical Research Use Cases
Explore example use cases and learn how to AWS cloud-technology can enhance your agility and data-driven insights in discovery and pre-clinical research.
-
High performance computing
-
AI/ML in research
-
Connect data from research equipment
-
Genomics research
-
Enabling Collaboration
-
High performance computing
-
High performance computing
In early discovery and pre-clinical research, scientists need access to extensive computing power to perform tasks such as computational chemistry simulations or large-scale genomic analyses. These datasets can contain millions of data points and require billions of calculations to produce an experimental output. However, the procurement, setup, and management of on-premises high performance computing (HPC) clusters can be prohibitively costly and take months.
The AWS Cloud provides access to infrastructure without upfront investments, including powerful HPC environments that can quickly scale up during peak demand periods, and scale back down again when demand has decreased. Orchestrating services, like AWS Batch and AWS ParallelCluster, can simplify the use and management of these HPC environments, making them more accessible to everyday users. Moreover, the significant decrease in costs and time-to-results gives researchers the freedom to ask questions that weren’t previously possible in their on-premises environments.
Case studies
InhibOx Case Study
In this case study, learn how InhibOx is able to utilize virtually unlimited computing resources to power the world's largest virtual library of modeled drug candidates, while providing a high level of data security.
New York University Case Study
In this case study, learn how New York University collaborates with researchers around the globe and utilize HPC in AWS Cloud to accelerate its medical informatics research.
Baylor College of Medicine Case Study
In this case study, learn how Baylor College of Medicine was able to quickly scale up 21,000 cores for research analysis and complete a study 5x faster than with their local infrastructure.
Celgene Case Study
In this case study, learn how Celgene R&D moved two of its critical HPC workloads to AWS, which resulted in improved collaboration, increased business process agility, improved scalability, and lower total cost of ownership.
Bristol-Myers Squibb Case Study
In this case study, learn how Bristol-Myers Squibb was able to perform intensive clinical trial simulations on AWS with a 98% time savings, which helps them to shorten study time and reduce patient impact.
Related resources
HPC Training
Follow a customized learning path to launch an HPC environment on AWS and to review important concepts along the way.
HPC Well-Architected Framework
Explore common HPC scenarios and identify key elements to ensure your workloads are architected according to best practices.
AstraZeneca Webinar
In this webinar, learn how AstraZeneca was able to process 20,000 exomes of genomic data in just 20 days using AWS high-performance computing resources, while driving down the total cost of their research analysis.
HPC Resources
Access blogs, whitepapers, reference architectures and documents specific to HPC on AWS.
Blog
Building a tightly coupled molecular dynamics workflow with multi-node parallel jobs in AWS Batch.
-
AI/ML in research
-
AI/ML in research
Researchers look to artificial intelligence and machine learning (AI/ML) capabilities for performing tasks, such as toxicology predictions, workflow optimization, and compound modeling to help de-risk and advance molecules into development. However, building AI/ML algorithms is often outside the traditional skill set of pharma researchers, and their access to data scientists can be limited.
Researchers can use Amazon SageMaker to quickly and easily build, train, and deploy ML models at scale. In addition, because Amazon SageMaker supports all popular open source frameworks, researchers can also use it to build custom ML models. AWS provides on-demand access to digital trainings at no cost for scientists to learn the skills needed to take on projects, using the same curriculum used to train Amazon’s data scientists. To accelerate the development of your own ML experts through direct guidance and instruction, Amazon ML Solutions Lab can pair your team with Amazon machine learning experts to prepare data, build and train models, and put models into production.
Case studies
Celgene Article
In this article, learn how Celgene uses Amazon SageMaker to perform toxicology predictions in pre-clinical research.
Blackthorn Therapeutics Webinar
In this webinar, learn how startup Blackthorne therapeutics utilizes ML to accelerate scientific innovation.
Atomwise Webinar
In this webinar, learn how startup Atomwise built prospective validation of machine learning for small molecule drug discovery.
Insitro Webinar
In this webinar, learn how startup Insitro uses ML models to inform drug development.
Related resources
Intro to ML training course
Dive deep into the same ML curriculum used to train Amazon’s developers and data scientists in these training courses.
ML Solution Labs
A collaboration and education program that connects Amazon ML experts with AWS customers and partners.
Blog
Introducing the Amazon ML Solutions Lab
Blog
Build an online solubility prediction workflow with AWS Batch and Amazon SageMaker
-
Connect data from research equipment
-
Connect data from research equipment
Research labs need to handle large volumes of data acquired by legacy research equipment, such as microscopes and spectrometers. This data is locally stored, which limits how the data can be securely archived, processed, and shared with collaborating researchers globally.
Researchers can automate the transfer of instrument data into the cloud using AWS DataSync and AWS Storage Gateway, ensuring that important experimental data is securely stored and archived. Amazon S3 Glacier provides long-term, durable storage for data archiving at significant savings compared to on-premises solutions.
Case studies
NYU case study
In this case study, learn how New York University reliably transfers large datasets from diverse research equipment in parallel to the AWS Cloud.
ThermoFisher case study
In this case study, learn how ThermoFisher Cloud was built on AWS to help scientists securely store, analyze, and share data globally.
Celgene case study
In this case study, learn how Celgene uses Amazon Elastic File Service to store data for its R&D workloads and many of its critical applications.
Illumina case study
In this case study, learn how Illumina built its BaseSpace Sequence Hub on AWS to significantly reduce data analysis and storage costs.
Related resources
Blog
Expanding AWS hybrid cloud capabilities with block storage on Snowball Edge
Blog
AWS DataSync- automated and accelerated data transfer
Webinar
Automate and accelerate online data transfer
Webinar
Hybrid cloud storage with AWS Storage Gateway and Amazon S3
-
Genomics research
-
Genomics research
Genomic analyses can support basic scientific research, inform clinical trial cohort selection, and provide information to predict optimal treatment regimens for patients. The AWS Cloud provides dynamic resourcing and scalability necessary for fast and cost-effective genomics analysis, reducing time to insight. From building your genomics pipeline to the integration of genomic findings into diagnostic treatment patterns, AWS has a broad ecosystem of partners who can work with you to customize the right solution.
Case studies
Baylor case study
In this case study, learn how Baylor College of Medicine securely and efficiently shares terabytes of genome sequence data with collaborators globally.
Fabric Genomics case study
In this case study, learn how Fabric Genomics provides end-to-end genomic solutions for clinical labs and life sciences companies using AWS scalable compute options.
DNAnexus case study
In this case study, learn how DNAnexus powers large scale genome studies with their informatics and data management platform on AWS.
Blog
In this blog, learn how AWS and APN partners support the White House Precision Medicine Initiative.
Related resources
Blog
Building simpler genomics workflows on AWS Step Functions
Blog
Scaling high-throughput genomics workflows on AWS
Blog
Accelerating precision medicine at scale
Blog
Using Cromwell with AWS Batch
-
Enabling Collaboration
-
Enabling Collaboration
Case studies
AWS can facilitate global and multinational collaborations- for example, between pharma and contract research organizations (CROs)- by giving you the ability to create isolated access for researchers from different institutions. Organizations can protect their intellectual property through AWS fine-grained access controls and by creating and revoking encryption keys through AWS Key Management Service. Similarly, Amazon WorkSpaces provides virtual, cloud-based Microsoft Windows or Linux desktops for users, giving them access to the documents, applications, and resources they need, to help on-board new collaborators and help teams scale globally.
Johnson & Johnson case study
In this case study, learn why Johnson & Johnson decided to move to the cloud and redefine its global IT strategy, which includes using Amazon Workspaces for cloud-based desktops for its consultants and employees.
NIH STRIDES Initiative
In this blog, learn how AWS and the National Institutes of Health collaborate to foster secure scientific collaboration.
Celgene case study
In this case study, learn how Celgene R&D utilizes Amazon S3 and Amazon Glacier to store hundreds of terabytes of data, and enables teamwork by providing isolated access for researchers from different organizations.
AHA case study
In this case study, learn how the American Heart Association is supporting cardiovascular research collaborations by building its precision medicine platform on AWS.
Related resources
Whitepaper
Best practices for deploying Amazon WorkSpaces
eLearning
Learn how to provision, manage, and provide access to Amazon WorkSpaces
Webinar
Identity and Access Management Best Practices