Google wants to help university labs and hospitals store their clients’ genomes in the cloud, and they’re calling the effort "Google Genomics."

Google Genomics was actually launched last March, but a number of other announcements caused it to go largely unnoticed, reports MIT Technology Review. The goal of the system, Google says, is to "explore genetic variation interactively." This means allowing researchers to access millions of genomes and run batch analyses, easily.

Right now, decoding a single human genome can take a few hours or more, and the size of the resulting data amounts to about 100 gigabytes of raw data. Storing it with google costs $25 for a year, and doing computations on that data costs extra. But once the raw data’s cleaned up, a human genome can amount to less than a gigabyte of data. At that point, storing that information in Google’s cloud computing system cost $0.25 cents a year.

David Glazer, the Google software engineer leading the effort, told MIT that making the act of comparing genomes as easy as possible is becoming increasingly important, because the way researchers study genetics has shifted. People used to study a single genome, but now they’re looking at many at once. This, he says, is something Google can help with. With Google Genomics, researchers will get a chance to compare millions of genomes — and multiply their discoveries in the process, he says.

Already, the National Cancer Institute has signed on. The organisation said it would pay $19 million to upload copies of its 2.6 petabyte (1 petabyte = 1000 terabytes) Cancer Genome Atlas to Google Genomics and Amazon’s data center. And at least 3,500 genomes from public projects can already be found on Google Genomics. The prices for Amazon and Google’s storage centers are reasonable, Stanford physicist Somalee Datta told MIT. They cost about the same thing as keeping them in her own data center, and she thinks "they will keep dropping."