Genomics in the Cloud

Chapter 12. Interactive Analysis in Jupyter Notebook

With this chapter, we round out the experience of working in the cloud. We started out in Chapter 4 running individual commands in a shell, and built up proficiency with GATK tools in Chapter 5 through Chapter 7. Then, in Chapter 8, you learned about scripted workflows and discovered progressively better ways to run them in subsequent chapters.

Yet we’re now coming back to the inescapable fact that not everything in genomics can (or should) be done as a scripted workflow. Sometimes, you just want to interact directly with the data, maybe generate a couple of plots and determine what your next step should be based on what the plots look like. You might be in an early exploratory phase of your project, stuck midway through with some failed samples to troubleshoot, or moving on to digging into the genetics of a group of people. In any case, you need to be able to try out ideas quickly, keep track of what each attempt produces, and share your work with others.

In this chapter, we show you how to use Jupyter Notebook in Terra to achieve these objectives. We kick it off with a brief introduction to Jupyter, in case you’re not already familiar with the concept and tooling. We spend a bit more time on describing how Jupyter works in Terra, focusing on the capabilities and behaviors that are more specific to Terra and the cloud environment. Then, in the hands-on portion of this chapter, we guide you through an example notebook that demonstrates ...