AWS Public Sector Blog

AMILI helps advance precision medicine by building microbiome library on AWS

This is a guest post by Jeremy Lim, chief executive officer (CEO) of AMILI; Vivek Ravindran, chief technology officer (CTO) of AMILI; Raghunathan Shanmugam, engineering lead at AMILI; with Edwin Sandanaraj, genomics solutions architect at Amazon Web Services (AWS).

One hundred trillion tiny organisms called microbes live in the gastrointestinal system within our body. But these are more than just bacteria; this collection of fungi and viruses—known as the microbiome—may hold the key to better health.

AMILI is a healthcare technology (HealthTech) company based in Singapore that seeks to advance precision medicine and personalized health and nutrition by harnessing the potential of the microbiome. AMILI uses artificial intelligence (AI) and machine learning (ML) on Amazon Web Services (AWS) to comprehensively quantify and characterize gut microbiomes, with the aim to build and curate the world’s largest multi-ethnic Asia microbiome database.

Improving lives begins with insights from AI and ML

By examining the different microbiomes of healthy and ill individuals, researchers today are discovering how microbiome profiles influence our susceptibility to diseases, and whether they can be modified to prevent certain diseases, such as diabetes and cancer.

AMILI recently worked with Professor Pierce Chow, senior consultant surgeon at the National Cancer Centre, to study the microbiomes of patients at a high risk of primary liver cancer, aimed at diagnosing the disease more accurately and earlier. AMILI is also involved in more than 40 other research projects, collaborating with medical and academic partners, such as the National University of Singapore, Agency for Science Technology and Research, and Monash University. These research projects aim to answer a wide range of critical questions about microbes to predict the onset of diseases and discover their potential applications against metabolic and neurological disorders.

At the core of AMILI’s work is its proprietary platform, PRIME. The platform uses AI and ML to help it better understand the microbiome’s complex composition and interactions to discover new associations.

Along with the microbiome profiles, AMILI also collects information on diet and lifestyle choices. This data is used to continually train its algorithms in PRIME and build expert models for the different areas of medical science.

Scientists, medical advisors, and data scientists use PRIME to identify and analyze microbiome profiles associated with higher risks of specific diseases such as obesity and Inflammatory Bowel Disease (IBS). PRIME is also used to study the gut health effects of food and identify food clusters based on the microbiome type, both of which are relevant to food retailers and manufacturers.

Using the power of the cloud to drive research

To build the library of microbiomes, AMILI uses next-generation sequencing technologies to analyze the genome data, before using PRIME to interrogate the outcomes and generate precision gut microbiome reports for genome-guided healthcare decisions.

As the analysis of a single microbiome can take up to 1 TB, AMILI built its platform in the AWS Cloud to deliver microbiome insights rapidly at scale, without having to worry about spending a significant amount of time maintaining its infrastructure.

AMILI receives a dynamic volume of workloads, ranging from hundreds to thousands of samples every month. The team built an agile and scalable analytic platform on AWS to handle compute-intensive tasks at scale. The current microbiome analysis platform was able to achieve the dynamic scalability of varying sample workloads seamlessly on the cloud. The microbiome analysis architecture (Figure 1) demonstrates the capability of an automated system to run microbiome analyses at scale on AWS.

Figure 1. Microbiome analysis solution architecture.Figure 1. Microbiome analysis solution architecture.

The AMILI sequencing partners, which are external sequencing laboratories, process the DNA samples into readable DNA sequence data, which is used in molecular biology for studying genomes and proteins. Sequence analysis is a broad field of research with sub-domains, and aligning sequences can reveal important information about their structural and functional sites. The database maintains integrity between different nodes to support data quality. AWS Batch automates the process of running containerized batch and analytics workloads, allowing AMILI to scale microbiome analysis.

The classic virtual machine (VM) system the AMILI team deployed, before moving to AWS, was able to handle four microbiome analyses per day due to the procurement nature and scaling challenges. Now, after moving to AWS, the current microbiome analysis architecture system can achieve 1,000 microbiome analyses in six hours using AWS serverless capabilities, such as AWS Batch, Amazon API Gateway, and Amazon DynamoDB. In addition, the AMILI team orchestrated a microbial workflow pipeline system using containers managed in Amazon Elastic Container Registry (Amazon ECR) and a more efficient deployment of container images for the applications with Amazon Elastic Container Service (Amazon ECS). This cloud-native solution supports rapid scaling and performance delivery, while Amazon Elastic Compute Cloud (Amazon EC2) Reserved Instances and Spot Instances helped reduce compute costs by 20-40%.

Plus, the current platform is architected with modularized docker images managed in Amazon ECR and Amazon ECS and fully integrated with AWS Batch. This enables AMILI’s research and development team to quickly experiment with novel tools and integrate with pipelines in order to accelerate innovations to market. Their scientists explore novel pipelines, so an agile environment for continuous integration with the platform is an essential need for enhancing business outcomes.

As the microbiome library grows exponentially, business operations at AMILI require an extremely durable, flexible, and cost-effective storage environment to manage its precise microbiome data. Amazon Simple Storage Service (Amazon S3) helps the team meet scalable microbiome data storage requirements. With Amazon S3’s 99.999999999% of data durability, the team designed their platform to be secure and recoverable in case of disaster scenarios. The flexible storage tier system supports a wide range of data objects in appropriate storage tiers and helps achieve more cost efficiency. The server-side encryption at Amazon S3 helps encrypt data objects at rest to protect the microbiome data.

Meeting compliance requirements for the health and research industry with AWS

Business operations at AMILI handle private and sensitive patient information, so the microbiome platform is required to meet regulatory requirements. By building the microbiome platform on AWS, the team quickly designed the platform with the essential services needed to adopt ISO 27001 certification for the microbiome workflows. Importantly, AWS Key Management Service (AWS KMS) is an integral component of the microbiome analysis solution to encrypt sensitive data across AWS workloads using cryptographic keys. In addition, Amazon CloudWatch monitors the AWS services in the platform, and completely tracks the metrics of resource usage for auditing and review purposes. The CloudWatch logs are vital support systems for the team to maintain compliance with programs like the Health Insurance Portability and Accountability Act (HIPAA), payment card industry (PCI) compliance, Federal Risk and Authorization Management Program (FedRAMP), and System and Organization Controls (SOC).

Learn more about AWS for healthcare

Using AWS, the team developed AMILI with a scalable, elastic, and cost-efficient infrastructure, allowing them to focus on building its product and data pipelines. This helps drive the discovery of insights that will, in turn, advance precision medicine in the clinical realm and personalized health and nutrition in the consumer segment. Find out more about AMILI here.

For more resources regarding AWS in public sector healthcare, visit the public sector healthcare hub. Learn how to get started with AWS for health here. Curious how AWS can help your healthcare organization innovate with new technology solutions? Contact us to learn more.

Read more about AWS for healthcare:

Subscribe to the AWS Public Sector Blog newsletter to get the latest in AWS tools, solutions, and innovations from the public sector delivered to your inbox, or contact us.

Please take a few minutes to share insights regarding your experience with the AWS Public Sector Blog in this survey, and we’ll use feedback from the survey to create more content aligned with the preferences of our readers.

Dr. Jeremy Lim

Dr. Jeremy Lim

Professor Jeremy Lim is chief executive officer (CEO) and co-founder of AMILI, the first gut microbiome company centered in Singapore serving the region. Jeremy is active in the community: he chairs the steering committee of NIHA (NUS Initiative to Improve Health in Asia), an initiative to strengthen health policy research and education in Asia

Vivek Ravindran

Vivek Ravindran

As the chief technology officer (CTO) of a healthcare startup, Vivek brings a unique blend of technical expertise and industry knowledge to drive the development and commercialization of cutting-edge solutions in the field of microbiome science. Vivek is passionate about using technology to unlock the full potential of microbiome science and is excited to continue to drive innovation in this field to improve patient outcomes.

Raghunathan Shanmugam

Raghunathan Shanmugam

Raghu is a senior Amazon Web Services (AWS) solution architect at AMILI with experience in designing and deploying solutions on AWS. He leads a team of developers and other solution architects on complex projects and possesses a thorough understanding of best practices for migrating and integrating on-premise solutions to AWS.

Edwin Sandanaraj

Edwin Sandanaraj

Edwin Sandanaraj is the genomics solution architect at Amazon Web Services (AWS), and supports healthcare genomics opportunities in Asia-Pacific and Japan. He has over 15 years of experience in high throughput genomics data management and analysis. Edwin is passionate about supporting precision genomics efforts that help improve healthcare outcomes with cloud-based genomics solutions.