Abstract
BackgroundTechnical improvements have decreased sequencing costs and, as a result, the size and number of genomic datasets have increased rapidly. Because of the lower cost, large amounts of sequence data are now being produced by small to midsize research groups. Crossbow is a software tool that can detect single nucleotide polymorphisms (SNPs) in whole-genome sequencing (WGS) data from a single subject; however, Crossbow has a number of limitations when applied to multiple subjects from large-scale WGS projects. The data storage and CPU resources that are required for large-scale whole genome sequencing data analyses are too large for many core facilities and individual laboratories to provide. To help meet these challenges, we have developed Rainbow, a cloud-based software package that can assist in the automation of large-scale WGS data analyses.ResultsHere, we evaluated the performance of Rainbow by analyzing 44 different whole-genome-sequenced subjects. Rainbow has the capacity to process genomic data from more than 500 subjects in two weeks using cloud computing provided by the Amazon Web Service. The time includes the import and export of the data using Amazon Import/Export service. The average cost of processing a single sample in the cloud was less than 120 US dollars. Compared with Crossbow, the main improvements incorporated into Rainbow include the ability: (1) to handle BAM as well as FASTQ input files; (2) to split large sequence files for better load balance downstream; (3) to log the running metrics in data processing and monitoring multiple Amazon Elastic Compute Cloud (EC2) instances; and (4) to merge SOAPsnp outputs for multiple individuals into a single file to facilitate downstream genome-wide association studies.ConclusionsRainbow is a scalable, cost-effective, and open-source tool for large-scale WGS data analysis. For human WGS data sequenced by either the Illumina HiSeq 2000 or HiSeq 2500 platforms, Rainbow can be used straight out of the box. Rainbow is available for third-party implementation and use, and can be downloaded from http://s3.amazonaws.com/jnj_rainbow/index.html.
Highlights
Technical improvements have decreased sequencing costs and, as a result, the size and number of genomic datasets have increased rapidly
We developed a Perl script that can aggregate all the single nucleotide polymorphism (SNP) from multiple samples and merge them into chromosome-based genotype files, thereby allowing the files to be used as inputs to other genome-wide association studies (GWAS) tools
After the binary version of SAM (BAM) or FASTQ files have been uploaded to Simple Storage Service (S3), they can be processed in parallel by launching multiple Elastic Compute Cloud (EC2) instances or clusters in the cloud
Summary
Technical improvements have decreased sequencing costs and, as a result, the size and number of genomic datasets have increased rapidly. The data storage and CPU resources that are required for large-scale whole genome sequencing data analyses are too large for many core facilities and individual laboratories to provide. To help meet these challenges, we have developed Rainbow, a cloud-based software package that can assist in the automation of large-scale WGS data analyses. Cloud computing offers network access to computational resources where CPU, memory, and storage are accessible in the form of virtual machines (VMs) Using these VMs eliminates the need to build or administer local infrastructure while addressing the challenges involved in the rapid deployment of computing environments for bioinformatics. Cloud-based bioinformatics software include CloudBurst [11], Crossbow [12,13], Myrna [14], and CloVR [15]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
