An Empirical Study of Genomic Sequencing Workflow in the Cloud

Cloud computing is a well-known and long lasting paradigm that has transformed how large and small businesses manage their computational resources. The cloud and network’s global footprint match the global nature of science, with security and access controls that allow genomics researchers to manage data sharing. It’s also changing the way genomicists and other researchers collaborate and work with massive archive datasets. Researchers must understand the cloud and the new modes of analysis and collaboration it enables as it makes inroads into genomics. This work discusses genomic sequencing workflow procedures and software, as well as discuss how to setup a genomic sequencing workflow in the cloud with appropriate services, using the AWS Cloud as an example. We performed three genomic sequencing workflow execution Case Studies that use appropriate software with different hardware in the cloud. We ran a small-scale genomic sequencing workload with 290 MB of input data (FAST5) and a large-scale genomic sequencing workload with 232 GB of input data (FAST5). We provide a comprehensive performance evaluation of the genomic sequencing workflow with our execution Case Studies and comparisons to bare metal servers. The final output of the workflow will be used for filtration, validation, and clinical report generation as per clinical guidelines.