Human genome is open source, too

At a gathering of open source software developers, two leading researchers involved with the sequencing of the human genome delivered presentations citing the benefits of open source both in the development of computer systems and in science.

Ewan Birney, team leader for genomic annotation at the European Bioinformatics Institute (EBI), and Jim Kent, a research scientist at the University of California, Santa Cruz, spoke Thursday at the O'Reilly Open Source Conference here.

"I don't think you can have science without open source," said Kent, who is credited with helping to produce the first assembly of the human genome and creating the human genome browser, an application that allows users to navigate databases of genomic data on the Internet.

In genetic research, both speakers noted that openness is essential in order to allow scientists to come up with applications for the human genome that benefit humans.

"We have 3G bytes of information (about 99 percent of the entire human genome), and everything we understand about humans must somehow be encoded in that 3G bytes," Birney said. "Everything we see in the human biology is represented in that genome."

Because of its wide-reaching implications, nonprofit researchers and private industry, which raced to compile the human genome, have agreed to make that 3G bytes of data freely available. A similar philosophy is advocated in the software industry, especially here at the O'Reilly conference, where developers have gathered this week to discuss the philosophy that software code should be open to developers for review and modification.

In both science and computing, openness helps ensure the quality of products: research and software, respectively.

"You can't do science without having reproducible results," Kent said, explaining that scientists employ a system under which they conduct peer reviews on scientific discoveries before accepting them as sound. "People can't do that unless they can see your source."

Some open source developers argue that the same standards that are used for testing scientific experiments should also be used for testing software. Having access to code gives programmers the ability to dissect a software program and identify bugs, many attendees here say.

"That is definitely a very good idea," said Paul Wilt, senior principal software engineer with ProQuest Co., whose work with open source programming tools brought him to the O'Reilly conference. "Having a culture that is encouraging software developers to test products by examining the underpinnings is important."

Conversely, software makers that control the intellectual property of their software code and don't make it openly available run a greater risk of sending out buggy software, Wilt said.

There are differences between the openness of the human genome and a piece of software such as an operating system that could differentiate their need for being open source, said Alex Lewin, an open source developer who attended the presentation here.

"There is a fundamental difference in that the human genome is preexisting and software is a human creation," he said.

However, protocols such as TCP/IP (Transmission Control Protocol/Internet Protocol) have one similarity to the genome: they are essential for communication over the Internet in the same way the genome is necessary for human life, Lewin added. "Those need to be open," he said.

The speakers don't just give open source computing lip service, they also use it. A project Birney leads at EBI called Ensembl has worked to make available on the Internet the genomic sequences of a human, a mouse, a zebrafish and a mosquito. The group compiles all this data in databases built entirely using the open source database MySQL. It also runs two data centers that have a total of 800 Linux blade servers from RLX Technologies Inc., he said.

Join the newsletter!

Error: Please check your email address.

More about European Bioinformatics InstituteMySQLReillyRLXRLX Technologies

Show Comments

Market Place