Building Enterprise-Grade Bioinformatics Pipelines using Cloud-Scale Kubernetes

Workflow tools like Nextflow allow us to standardize and automate our bioinformatics pipelines. Coupled with Docker and Kubernetes, our pipelines then become portable and scalable. In this session, we will take a deep dive into the world of cloud-based Kubernetes and showcase the utility of containerized workflows for enabling enterprise-grade bioinformatics workflows in the cloud. Plus, we’ll demonstrate how cloud data lake are perfect for storing the immense amount of -omics data that life science orgs need to manage.

Model. Package. Deploy. Repeat: A DevOps Approach to Biomolecular Scalability

Computational tools that allow us to model biomolecular processes are at the forefront of scientific discovery with new models seeming to be announced every week. However, given that many of these models come from academic labs or independent developers, there can be a disconnect between running a model locally on a single example versus at-scale for thousands of experiments.

In this talk, we will dive into the best practices surrounding the lifecycle of biomolecular model development. From model training and tuning, to containerization and deployment, the talk will emphasize scalable strategies across high-performance computing (HPC) clusters and cloud-native platforms (e.g., Kubernetes and serverless technologies).

Attendees will gain practical DevOps insights into building flexible, scalable systems that accelerate discovery in biomolecular modeling while maintaining portability and reproducibility across computing environments.

Boston Protein Design and Modeling Club - June 2024