Streamlining Microbiome Research on a Secure Data Platform

Efficient Handling of Microbiome Data Speeds Up Microbial Research, Drug Trials and Discovery

In recent decades, the healthcare industry has been undergoing a rapid transformation, from a one‐drug‐fits‐all approach, to more personalized medicine. The customization of healthcare, with medical decisions, treatment, practices, and products that are tailored to a subgroup of patients, is reported to offer such advantages as:

• Efficient preemptive disease detection and treatment
• Considerable reductions in time, cost, and failure rates of pharmaceutical clinical trials
• Fewer trial-and-error inefficiencies that inflate healthcare costs and undermine patient care

As a biopharmaceutical company with 10 years of experience in microbiome research, Second Genome is at the forefront of the “right patient, right therapy” transformation, helping to identify responder/non-responder populations and determine the optimal approach to therapy.

Second Genome has developed a proprietary data platform — Microbiome Drug Discovery and Development Platform — that uses machine learning to rapidly translate clinical and molecular data into actionable insights for novel biomarkers and drug discovery programs. With more than 400 structured databases of public and proprietary microbiome datasets available on SGKnowledgeBaseTM, the platform helps pharmaceutical companies augment their research on genetic variations and enable novel drug discovery by identifying precision biomarkers.

Second Genome joined forces with Provectus in an effort to enhance its data platform by making it faster, more scalable, secure and compliant. It was critical for Second Genome to raise the bar for data security compliance, to encourage premium pharmaceutical companies, as well as government agencies and academia, to conduct their drug research and development on its data platform.

Provectus collaborated with Second Genome on the design and development of a new secure, cloud-native data infrastructure on AWS, starting with the improvement of the existing data ingestion and staging components of its platform’s data pipeline. The teams also partnered to build a Machine Learning solution, to contextualize the IO biomarkers of specific therapeutic programs.

Building a Secure and Compliant Data Infrastructure for Industrial-Scale Drug Candidate Discovery

Provectus reviewed the data ingestion and staging portions of the data pipeline of Second Genome’s data platform. Along the way, the team looked into such issues as data quality, error monitoring and logging, handling of hard-coded variables, and running API tests on sample data.

To deliver a new secure cloud-native data infrastructure for Second Genome’s data platform, Provectus applied DevOps best practices to automate patch management, centralized logging and disaster recovery, and to introduce and improve CI/CD. The CI/CD work helped drive efficiencies in the handling of features and feature environments, deployment to production and delivery pipelines, and monitoring and reporting.

From a security standpoint, data in transit and at rest were secured in Amazon S3 with TLS. Access logs were securely stored in Amazon S3, with Amazon VPC flow logs collected as well. All open security groups were eliminated, with MFA ensured for most users. Penetration tests were run to confirm that the delivered infrastructure meets AWS security requirements.

On the development side, Provectus helped anonymize Neo4J Data and optimize Django settings. The pipeline was thoroughly tested and made ready for new features to be added as the platform develops.

As part of the data infrastructure building process, Provectus and Second Genome are now in the process of revamping ML infrastructure and MLOps, to further enable Second Genome to contextualize its IO biomarker program to its therapeutic program, with the help of machine learning.

Microbiome Data Available on a Safe and Secure Data Platform Accelerates and Scales Drug Trials, Advances Precision Medicine

The new data infrastructure enabled Second Genome to enhance its Microbiome Drug Discovery and Development Platform, making it possible to run petabyte-scale projects for biomarker research, drug trials, and drug development.

The data infrastructure meets the strict requirements and high standards of AWS for security and operational efficiency in the cloud. Data pipelines are fully automated, with CI/CD in place, to ensure that Second Genome’s microbiome datasets are collected, stored, and used in a compliant manner.

Thanks to the new secure cloud-native data infrastructure on AWS, Second Genome is now able to more easily onboard premium pharmaceutical companies to its platform. The partners of Second Genome can now discover, develop, and test drugs much faster and on an industrial scale, without having to worry about the security of their research. With its new platform, the company expects to quickly build new collaborations across industry, government, and academia, to advance its biomarker and drug discovery programs.

Moving Forward

1. Learn more about the Provectus NextGen Data Platform and Machine Learning Infrastructure solutions
2. Watch the webinar on MLOps and reproducible ML on AWS
3. Apply for NextGen Data Platform Acceleration Program or Machine Learning Infrastructure Acceleration Program, to get started