With the increasing ubiquity of smartphones, the growing popularity of tablets, and the burgeoning market for third-party applications, the public appetite for touchscreen technology is virtually impossible to ignore. It shouldn't, then, surprise anyone that bioinformatics software developers would look for ways to tap into the multi-touch market.
But it's not just about making sure that download tallies are going up. In fact, some life science application developers say they dipped into the app space out of sheer curiosity more than anything else.
"For us, it was an experiment," says Andrew Su, an associate professor at the Scripps Research Institute. His laboratory developed an application called BioGPS, a mobile version of his group's web-based gene annotation portal, which runs on Apple's iPhone.
"We imagined that somebody in a meeting or a seminar or at a conference who needed quick access to some gene-centric information would use the BioGPS app," he says.
Benjamin Gross, senior software engineer at Memorial Sloan Kettering Cancer Center, echoes Su's comments. Gross developed the BioGene app for the iPhone to provide researchers access to gene function content from the National Center for Biotechnology Information's Entrez websites.
An ideal use case for his app, Gross says, would be "a scientist at a conference listening to a presentation … about the BRAF gene" who doesn't know what that gene does can find out "through BioGene what the BRAF gene does, so then the rest of the presentation you are listening to makes a lot more sense."
For Sean Ekins, a senior consultant for Collaborations in Chemistry, a division of Collaborations in Communications, a mobile app was a logical step given the proliferation of mobile compute technologies.
"We are going to rely on mobile computing the same way we've relied on desktop computing for the last decade or so," Ekins says.
He and his colleagues developed the Open Drug Discovery Teams app for Apple's devices. The app collates information from the web related primarily to research into cures for rare and neglected diseases, including malaria and tuberculosis.
"We can do so much more on the mobile devices than we probably could have done a couple of years ago," he says.
'Not rocket science'
For Apple's iOS, at least, the app-development process is simple, says William Van Etten, a founding partner of the BioTeam informatics consultancy.
For $99, "you sign up for the Apple development program, you download the Xcode integrated development environment through the app store — most of developing the graphical user interface is kind of drag and drop — then you have to write your own code to provide the functionality behind the user interface, then you compile it into an application and submit it to the app store through their developer web portal system and usually within a couple of weeks it gets approved and into the store," he says.
With its program, Apple has made it very difficult to put together a poor app, Van Etten says.
MSKCC's Gross adds that "there is a learning curve," but it is "not rocket science." Still, Gross notes that it is wise to remember that sometimes programs don't translate well from the development environment into reality even with the best of toolkits.
"When you are developing in a tool that simulates an iPhone … sometimes that's quite different from your code or application running on a real phone," he says.
While most bioinformatics developers have developed iOS-based apps, some say they have considered trying out Android's Linux-based operating system and others do try for compatibility with the BlackBerry.
To DIY or not
Deciding whether to develop an app internally or to partner with a third party really depends on how much time and effort can be expended poring over developer documentation and learning new compute speak.
"Web development is different enough from app development that we thought it would be more efficient just to contract it out to a third party," Scripps' Su says. Although his team developed most of the functionality for the BioGPS website, they tapped Optra Systems to develop the iPhone app version of the resource.
That decision allowed them to shift the burden of translating the Python-based BioGPS website into Apple's Objective-C programming language to "somebody who had already gone through that learning curve," he says.
That's one of the benefits of working with firms like Optra, company CEO Gholap says. Another advantage is that the customer doesn't have to drudge through the grunt work of ensuring that apps connect securely to their main information source — a process that is different for each mobile provider.
Optra already offers infrastructure for that purpose, so "we don't need to write everything from scratch," he says. "We can use components out of our … infrastructure" and then fill in the gaps between "what we have versus what they require" with custom code.
But outsourcing isn't for everyone. There are some folks who like to do it themselves. MSKCC's Gross, for example, developed the BioGene app himself — a process that, he says, took about two months.
Most of that time, according to Gross, was spent writing a backend server that sat between the iPhone and NCBI's EntrezGene website that would filter information that was coming to the phone and by extension speed up the response time to queries.
The largest chunk of the process was "intelligently figuring out how to display the information and then dealing with the performance tuning and figuring out the right way to represent the information in a database," he says. "The actual coding wasn't that much of an issue."