UW Information Technology

UW researchers use cloud computing to understand how to turn microbes at the bottom of Lake Washington into allies for a cleaner world

Microbial samples go through several steps in the lab before genetic data can be analyzed in the cloud.

A single scoop of sediment collected from the bottom of Lake Washington may not look like much to the untrained eye, but for UW researchers, the complex microbial communities living in the mud might hold the key to a cleaner world.

David Beck

David Beck
Research Associate Professor of Chemical Engineering
Director of Research, eScience Institute

David Beck, a professor in the Department of Chemical Engineering and director of research at eScience Institute, is one of several UW researchers who are seeking to understand how microbes collaborate and compete in freshwater lake sediment. This information someday could be used to tweak these microbial communities for our benefit, including to create a cleaner environment, cleaner fuels and even more efficient wastewater treatment facilities.

“We are trying to predict if microbes are competing or collaborating,” said Beck, who is doing the bulk of his research using cloud computing. “Does A like being around B? What are they doing? And how are their genes turning on their RNA (which acts as a messenger to carry instructions from DNA to proteins)?

“And in turn,” Beck said, “how can we manipulate these communities so they do what we want them to do?”

By learning how microbes affect one another in a tight environment, for example, researchers can manipulate their numbers for a specific outcome, such as getting more methane production that can be turned into a fuel.

In particular, Beck is studying methanotrophs, single-cell organisms that metabolize methane as their only source of carbon and energy. How they go about their business may be affected by their environment and their neighbors, thousands of other unique organisms that share the same space at the bottom of the lake.

Cloud computing can help UW scientists accelerate and expand their research computing capabilities.

To crunch the vast amounts of microbial genetic data involved in understanding these dynamics, Beck moved his work to the cloud with assistance from UW Information Technology’s (UW-IT) Cloud and Data Solutions team. Cloud computing, also known as the public cloud or commercial cloud, is a massive computing resource that researchers can access via the internet. UW-IT works closely with UW researchers as they consider cost, security, usability and scale.

The cloud helps Beck simplify his job

“To do our work,” Beck said, “we have collected a large amount of metagenome (all the genetic material present in an environmental sample) and metatranscriptome (complete set of RNA molecules for an entire group of interacting organisms or species) sequencing data.”

That’s a lot of data — about 9 terabytes in all, roughly the equivalent of 153,000 hours of music or nearly 3 million digital photos.

“I chose the cloud because we can move a lot of data very quickly as we need to, from one place to another,” Beck said. “What could have taken weeks to transfer, now we can do it overnight, and often in even less time. With the cloud, it is not that hard of a problem to move nine terabytes of data.”

With the cloud, it is not that hard of a problem to move nine terabytes of data.
David Beck

Moving the data in and out of the cloud also has become more economical, including renting virtual machines to handle his work.

“I can rent a 2-terabyte RAM machine for $6 an hour, and we can lease what we need when we need it,” Beck said.

The cloud also allows him to easily compartmentalize his work.

“We need to process all that information by assembling, binning and mapping software to produce data that can be fed to social network machine learning and analysis algorithms,” Beck said.

At the UW, the eScience Institute and UW-IT’s Cloud and Data Solutions work together to guide researchers toward the best cloud computing solution and other innovative computing tools.

“Being in the cloud has given us an unbounded playground to experiment with different software tools and their parameter settings to produce the highest quality data for our machine learning efforts,” Beck said.

Erik Lundberg, assistant vice president of UW-IT’s Research Computing, Collaboration & Strategy, said Beck’s work exemplifies the goals of his division.

“Providing researchers with a broad range of options is the next big thing we’re trying to establish for our research computing services,” Lundberg said. “Besides high performance computing (a shared cluster supercomputer on campus known as Hyak) and multiple cloud options, we are offering consulting to help a researcher make the best choice of which path to follow.”

From mud to zeroes and ones

Mary Lidstrom
Vice Provost for Research, Professor of Chemical Engineering and Microbiology

On this project, Beck is collaborating with Mary Lidstrom, vice provost for research and professor of chemical engineering and microbiology. She is conducting several studies on microbes, including methanotrophs.

Lidstrom’s microbiology lab isolates the DNA, RNA and proteins of all the microbes present in the slice of mud. It is Beck’s job to turn that data into a computational problem that is helping them understand how microbes affect one another.

For an example, Beck cites wastewater treatment facilities, which use biological processes to purify the water.

“From what we learn, we could optimize the microbial communities at these facilities so they remain healthy and do a better job of cleaning the water, he said.”

Or researchers could tweak a particular microbial community to more effectively produce or consume more methane — one way giving us cleaner air and the other a cleaner fuel.

“By crunching the data in the cloud, we can learn, predict, test, and ultimately make a better tool to design better microbial communities,” Beck said.

Learn more:
Cloud computing for research is helping UW researchers to successfully do their work.