Databricks Runtime for Genomics

Databricks Runtime for Genomics (Databricks Runtime Genomics) is a version of Databricks Runtime optimized for working with genomic and biomedical data. It is a component of the Databricks Unified Analytics Platform for Genomics. For more information on developing genomics applications, see Genomics guide.

What’s in Databricks Runtime for Genomics?

• An optimized version of the Databricks-Regeneron open-source library Glow with all its functionalities as well as:
  • Spark SQL support for reading and writing variant data
  • Functions for common workflow elements
  • Optimizations for common query patterns
• Turn-key pipelines parallelized with Apache Spark:
  • DNASeq
  • RNASeq
  • Tumor Normal Sequencing (MutSeq)
  • Joint Genotyping
  • SnpeEff variant annotation
• Hail 0.2 integration
• Popular open source libraries, optimized for performance and reliability:
  • ADAM
  • GATK
  • Hadoop-bam
• Popular command line tools:
  • samtools
• Reference data (grch37 or 38, known SNP sites)

See the Databricks Runtime for Genomics release notes for a complete list of included libraries and versions.

Requirements

Your Databricks workspace must have Databricks Runtime for Genomics enabled.

Create a cluster using Databricks Runtime for Genomics

When you create a cluster, select a Databricks Runtime for Genomics version from the Databricks Runtime Version drop-down.