Amazon Web Services ブログ

Bio-IT World Expo 2024 AWS関連セッションのお知らせ

こちらのBlogは、”Join the AWS Genomics Team at Bio-IT World Expo 2024“を和訳し、一部修正を加えたものです。

2024年4月15日から17日までボストンで開催されるBio-IT World Conference and Expo(※当カンファレンスは閉会しました)のゴールドスポンサーとして、AWSゲノミクスチームは、オミクスデータから洞察を得るために特別に設計されたサービスであるAWS HealthOmicsが、研究を変革し、精密医療の未来を加速させる方法をご紹介することをを心待ちにしています。このプレミアイベントでは、複雑な課題を解決し、科学的なブレークスルーを促進することに尽力するライフサイエンスやテクノロジーの専門家が一堂に会する予定です。

この業界の進化の最前線に立ち、AWSは、サービスとパートナー連携がデータの流動性を高め、R&D ワークフローを再構築し、人工知能(AI)をシームレスに組み込んで、科学的および臨床的な洞察を具体的な患者のアウトカムに変換する方法を実演します。3日間のカンファレンスを通して、AWSのお客様とパートナーは、創薬研究から臨床現場までのイノベーションを推進するためにクラウドを活用した実例を共有します。参加者は、組織がオミクスデータ解析のスケーリング、創薬のための生成AIの展開、国家規模の多様ながん研究プラットフォームの構築に関連するIT上および科学上の障壁を克服するためにAWSを活用している方法を学んでいただけます。

このブログでは、AWSが主催するレセプションに参加する方法や、パートナー主導のデモを体験するためのプライベートデモスイートを訪れる方法など、包括的なBio-ITカンファレンスのアジェンダをまとめています。これらのセッションをブックマークし、デモに参加し、カンファレンス中にAWSゲノミクスチームと個別に会議を設定するためにAWSアカウントチームに連絡することをお勧めします。ボストンでみなさんにお会いできるのを楽しみにしています。

AWS Speaking Sessions

Overcoming the Scientific and IT Challenges Associated with Scaling Omics Analysis
Tuesday, April 16, 12:25 PM ET | AWS and Basepair | Track: Cloud Computing

Discover how a combination of AWS HealthOmics and partner solutions from Basepair can help life sciences organizations maximize productivity, reduce costs, and adhere to IT best practices when scaling omics workflows.

Generative AI Distributed Training for Drug Discovery with AWS and NVIDIA
Wednesday, April 17, 12:40 PM ET | AWS and NVIDIA | Track: AI for Drug Discovery and Development

This session covers deploying NVIDIA’s BioNeMo on AWS ParallelCluster and Amazon SageMaker. Learn how to use BioNeMo’s containers for generative AI and protein structure visualization on these AWS services.

Utilizing Generative AI for Drug Discovery in the Cloud
Wednesday, April 17, 1:20 PM ET | Luncheon | AWS | Track: AI for Drug Discovery and Development

At this session, Evozyne will showcase how they leveraged generative AI on NVIDIA’s BioNeMo platform and AWS to create two new proteins with supernatural function. The developed LLMs outperform traditional protein engineering approaches by enabling the production of functional synthetic molecules with hundreds of mutations in a single round. The role of large, pre-trained models and libraries to unlock such generative design will be reviewed alongside the benefits cloud computing provides in storage, automation, and parallelization.

AWS Hosted Reception

Bio-IT World 2024 Opening Night Reception

Hosted By: Rescale and AWS
Program Time: Monday, April 15, 7:30 pm – 8:30 pm ET
Location: Abstract Room located on the 3rd floor of the Omni Hotel Seaport (conference center)

Bio-IT World 2024 is a marathon, not a sprint! Recharge on day 1 with Rescale and AWS at our opening night networking event and reception, fueled by complimentary cocktails and bites.

R&D modernization, from data to drug discovery applications

Hosted By: Deloitte, NVIDIA, and AWS
Program Time: Tuesday, April 16, 5:00 pm – 7:30 pm ET | Reception: 7:30 pm – 8:30 pm
Location: AWS Seaport Offices (BOS21), 55 Pier 4 Blvd., Boston, MA 02210

We are excited to invite you to an insightful event on R&D modernization from data to drug discovery applications, hosted by Deloitte, NVIDIA, and AWS. If you’re attending the Bio-IT World conference or are in the Boston area during the evening of April 16 and are interested in connecting with your peers, we’d love for you to join us!

AWS & Tennex BioIT Cocktail Reception  

Hosted By: Tennex and AWS
Program Time: Wednesday, April 17, 5:00 pm – 7:00 pm ET
Location: Trillium Brewing – Fort Point, 50 Thompson Place, Boston, MA 02210

Join the AWS Startups Life Sciences Team and Tennex for a BioIT Happy Hour on Wednesday, April 17 from 5pm to 7pm at Trillium!

AWS HealthOmics Demo Suite Agenda (Suite-12004)

Join us in our demo suite, where AWS partners will showcase cutting-edge solutions and provide hands-on technical guidance to help you build for the future of genomics. To attend, simply reach out to your AWS account team or speak with representatives from the AWS Genomics team onsite at the conference. You can find us at any of the speaking sessions and reception listed above. Don’t miss this opportunity to explore innovative partner demonstrations and connect with AWS genomics experts.

Run enterprise-scale Generative AI Models for Drug Discovery using BioNeMo NVIDIA Inference Microservice (NIM) on AWS
Tuesday, April 16, 1:00 PM – 1:45 PM ET | Demo Session | NVIDIA

NVIDIA BioNeMo is a generative AI platform for machine learning in chemistry and biology. It accelerates drug discovery’s most time-consuming and costly stages by streamlining the development and deployment of AI models. There are two parts of BioNeMo. The BioNeMo Framework is a freely available docker container that provides tools for model training scalable to multiple GPUs and nodes, with integrations for different AWS services. BioNeMo microservices are extensively optimized to scale inference to thousands of GPUs and are packaged as NVIDIA NIMs for easy deployment. As part of the broader cross-domain NVIDIA NIM catalog, each container is standardized by packaging an inference-optimized model with NVIDIA AI Enterprise base image, an NVIDIA triton server, and a cloud API. Users can flexibly compose workflows with multiple NIMs to predict properties of small molecules, proteins, and nucleic acids, predict interactions between some of these modalities, or generate novel molecules. AWS and NVIDIA have joined forces to offer high-performance, low-cost inference for NIMs, which will soon be available on AWS HealthOmics and SageMaker. In this demo we will introduce you to BioNeMo NIMs and example workflow of property-guided small molecule generation.

Basepair and AWS HealthOmics: A Point & Click Way of Running a Managed AWS Service
Tuesday, April 16, 2:00 PM – 2:45 PM ET | Demo Session | Basepair

Wednesday, April 17, 12:00 PM – 12:45 PM ET (Repeat Session)

Basepair’s SaaS platform accelerates the migration, deployment, and scaling of AWS HealthOmics workflows in any AWS account, anywhere in the world. Its point and click interface then enables research scientists to perform routine analyses, interactively visualize and explore genomic data on their own so they can collaborate with bioinformaticians on an informed question.

Building Custom Genomics and AI/ML Workflows on Amazon HealthOmics
Tuesday, April 16, 3:00 PM – 3:45 PM ET | Demo Session | Clovertex

Clovertex will showcase how to build custom workflows for genomics and AI/ML analysis using HealthOmics, which enables storing, analyzing, and querying large-scale genomic, transcriptomic, and other omics data. Clovertex will provide guidance on migrating workflows to HealthOmics, optimization strategies, and lessons learnt. Attendees will also learn how to leverage HealthOmics for building AI/ML functionalities for data analysis.

Managing Genomics Computing with EPAM CORA™ for AWS HealthOmics
Wednesday, April 17, 10:00 AM – 10:45 AM ET | Demo Session | EPAM

Come see how you can democratize access to genomics and other high-compute solutions for your scientists, maintain security and control over costs, and integrate these calculations into larger business workflows. EPAM Collaborative Omics Research Accelerator™ (EPAM CORA™) for AWS HealthOmics, recently announced by EPAM, will make it easy to leverage AWS HealthOmics using EPAM’s proven Cloud Pipeline framework within your AWS VPC.

Intel Open Omics: Ushering in Faster/Cheaper MultiOmics Pipelines on AWS HealthOmics
Wednesday, April 17, 11:00 AM – 11:45 AM ET | Demo Session | Intel

Intel’s Open Omics Acceleration Framework (in short, Open Omics) is a community-driven full stack high throughput framework for accelerating omics pipelines. This demo will highlight examples of seamlessly realizing industry leading performance for key pipelines on AWS HealthOmics.

AWS Partner and Customer Speaking Sessions

Application of Quantum Computing to Cyclic-Peptide Docking
Monday, April 15, 10:20 AM ET | Chugai Pharmaceutical | Track: Quantum Computing

Chugai has developed original mid-size molecular drug discovery technologies, which can generate orally bioavailable cyclic peptides. We are exploring invaluable applications of quantum computing technology to our mid-size molecular research. We are planning and conducting POC studies for examining its potential for enhancing our mid-size molecular research. As a case study, we will introduce a POC study of cyclic-peptide docking simulation using a quantum computing-inspired optimization solution.

Using AI and Machine Learning (AI/ML) to Power Predictive Drug Discovery
Tuesday, April 16, 10:25 AM ET | Vertex Pharmaceuticals | Track: AI for Drug Discovery and Development

In the drug discovery process, testing compound libraries is time-consuming and costly. Join this session to learn how Vertex accelerates compound testing by employing predictive modeling for rapid activity predictions. Discover how they built a modern MLOps platform on AWS services like Amazon SageMaker and AWS Batch to train, deploy, and manage thousands of activity prediction models at scale. Explore how this scalable infrastructure enables daily model retraining, optimizes small molecule testing processes, and fosters collaboration across teams.

A First Look at National-Scale Multimodal Cancer Research in Practice
Tuesday, April 16, 10:55 AM ET | Genomics England | Track: AI for Oncology, Precision Medicine, and Health

Genomics England and its partners are currently building the world’s largest multimodal research platform for cancer. The platform uniquely brings together clinical WGS and healthcare data—from Genomics England’s work enabling the NHS to deliver precision medicine through the Genomic Medicine Service—with hundreds of thousands of matched pathology and radiology images being digitized by the National Pathology Imaging Consortium and made accessible through AWS. Join this session to hear about the progress and challenges of making 100PB of data queryable in the cloud; a first look at the sort of insights this platform can provide; and how industry can use it and partner with Genomics England and its participants to advance new therapy.

Scalable Cloud Infrastructure Design
Tuesday, April 16, 11:25 AM ET | Vertex Pharmaceuticals | Track: Cloud Computing

This talk dives deep into how Vertex Pharmaceuticals modernized its image segmentation platform using AWS Serverless technologies and explains how it benefited scientists in their research & development work. This session will review how Vertex’s legacy system worked and pain points associated with it, discuss the new cloud-based system using AWS Serverless, benefits gained, and lessons learned.

Quilt: The Open Solution to Data Chaos
Tuesday, April 16, 12:10 PM ET | Quilt Data, Inc. | Track: Data Management

The vast majority of Life Sciences companies fail to capture the full value of their data investment due to the organizational friction caused by data silos. We examine the underlying causes of this “data chaos” and explore how an open-source universal data abstraction–data packages–can help organizations rewrite the tradeoff between individual productivity and organizational velocity.

Slalom Presents the xCures Platform, Built on AWS Employing AI/ML for Seamless Data Acquisition & Real-Time Access
Tuesday, April 16, 1:05 PM ET | Luncheon | Slalom | Track: Data Science and Analytics Technologies

Slalom presents the xCures Platform, built on AWS to employ AI/ML technologies that seamlessly acquire, extract, organize, and annotate medical records, providing both longitudinal views and real-time access to the source-verifiable clinical data

How to Reliably Run Omics Pipelines on AWS Spot Instances with MMCloud
Tuesday, April 16, 1:05 PM ET | Luncheon | MemVerge | Track: Bioinformatics

In the rapidly evolving landscape of genomic research, the demand for cost-effective and scalable computational resources is paramount. This talk explores the innovative integration of MMCloud (Memory Machine Cloud) with AWS Spot instances to reliably run omics pipelines, addressing the core challenges of cost, performance, and stability in high-throughput genomic analysis. MMCloud, with its unique SpotSurfer technology, offers a revolutionary approach to utilizing the economic advantages of AWS Spot instances while mitigating the risk of instance termination. We delve into the dynamics of SpotSurfer’s checkpointing mechanism, ensuring seamless pipeline execution by preserving computational states during Spot instance interruptions. Additionally, the talk examines MMCloud’s real-time resource optimization features like WaveRider and WaveWatcher, which dynamically right-size computational resources and provide comprehensive observability, thereby enhancing performance and reducing costs. By harnessing these capabilities, MMCloud emerges as a robust solution for running omics pipelines on AWS, offering researchers and bioinformaticians an unprecedented balance of affordability and reliability in cloud-based genomic computation. This talk not only presents a technical overview but also demonstrates the practical application and benefits through a series of benchmark tests, underlining MMCloud’s potential to transform the landscape of genomic research in the cloud.

Life Science Organizations in the Cloud—SNAFUs, FUBARs, and OMG Moments
Tuesday, April 16, 2:30 PM ET | Flagship Pioneering | Track: Modern Data Platforms and Storage Infrastructure

This talk will provide some balance to the abundance of cloud adoption success stories. We’ll explore real-world examples from start-ups and big pharma of things that didn’t go quite as expected. Some stories may be familiar, some not, but they all contribute to our understanding of how life science organizations can make the best use of the cloud. The experiences presented are with AWS but generalizable to other cloud providers.

AION Labs, a Venture Studio Empowering Entrepreneurs to Create or Grow Their Company with Pharma and AWS
Tuesday, April 16, 3:00 PM ET | AION Labs | Track: Generative AI

AION Labs is an alliance of global pharma companies, Amazon, and venture capital firms that have come together to co-develop and adopt groundbreaking new AI technologies that will transform drug discovery and development. AION Labs provides access to funding, data, validation, and technologies in a co-development model from day one. In this talk, I will share our model for creating or joining AI companies in the drug discovery and development space. We invest significantly in defining articulating the problem to be solved. The speaker will share examples from antibody discovery and optimization challenges that we launched and describe the companies we launched.

PANEL DISCUSSION: Digital Leadership Lessons: Reflecting and Correcting
Tuesday, April 16, 3:00 PM ET | Panel Discussion | Flagship Pioneering | Track: Modern Data Platforms and Storage Infrastructure

This discussion will convene executive leaders to share their experiences about technology, data, and cultural decisions that led to surprising outcomes. From unanticipated obstacles to valuable lessons learned, the panelists delve into the real stories that shaped the journey of start-ups and big companies alike. Join us for a light-hearted conversation about serious topics.

こちらのブログはヘルスケア・ライフサイエンス事業開発部 片岡 が翻訳しました。