Paper
Benchtop genome sequencers such as the Illumina MiSeq or MiniSeq [1], [2] are revolutionizing genomics research for smaller, independent laboratories, by enabling access to low-cost Next Generation Sequencing (NGS) technology in-house. These benchtop genome sequencing instruments require only standard laboratory equipment, in addition to minimal time for sample preparation. However, post-sequencing bioinformatics data analysis still presents a significant bottleneck, for research laboratories lacking specialized software and technical data analysis skills on their teams. While bioinformatics computes clouds providing solutions following a Software as a Service (SaaS) are available ([3]-[6], review in [7]), currently, there are only a few options which are user-friendly for non-experts while at the same time are also low-cost or free. One primary example is Illumina BaseSpace [8] that is very easy to access by non-experts, and also offers an integrated solution where data are streamed directly from the MiSeq sequencing instrument to the cloud. Once the data is on the BaseSpace cloud, users can access a range of bioinformatics applications with pre-installed algorithms through an intuitive web interface. Nonetheless, BaseSpace can be a costly solution as a yearly subscription depending on whether the user is associated with an academic or private institution, ranges in price from $999 - $4,999. Additional \"iCredits\" [9] might need to be purchased for frequent users that exhaust the base credit allowance as part of the subscription. Considering the reduction of computer hardware cost in recent years, a multi-core Intel Xeon server with 64 GigaByte (GB) of memory and multiple TeraByte (TB) of storage is priced less than the yearly subscription to Basespace [10], and similarly when compared to renting compute cycles from providers such as Amazon Web Services (AWS) [11]. Furthermore, the current generation of laptops usually come with 6-10 GigaBytes (GB) of memory and 1 TeraByte (TB) of storage, providing enough computational capacity to analyze data from small NGS experiments [12] that include only a few samples.
