Vol 8. Issue 36 / November 24, 2008
Touch and Go
By Eric Sauter
It was Valerie Cavett and Jennifer Busby of Scripps Florida's Proteomics Core who first came up with the whole crazy idea, according to Bruce Pascal, Scripps Florida's senior software engineer.
"They wanted to know if we could apply a multi-touch screen approach to proteomics analysis," said Pascal. "I said, 'You're crazy,' but once I explored existing technology I came back and told them it was possible."
Defining the proteome—the range of proteins expressed by cells, tissues, and the genome—can be accomplished in a number of ways but mass spectrometry remains the most important, as an established method that allows scientists to precisely identify the composition of a protein.
Multi-touch technology, in one form or the other, has been around since the early 1970s but the possibilities, the real WOW factor, arrived more recently when a bespectacled engineer named Jefferson Han made his touch screen presentation in 2006 at the Technology Entertainment Design (TED) conference, an invitation-only event. This was the basis of the crazy idea that Bruce Pascal thought he might be able to develop for Scripps Florida scientists.
"In mass spectrometry there are two main camps," Pascal said. "One believes in the statistical results to validate spectral matches; the other group wants to confirm spectra visually. What you end up doing is staring at a lot of pictures of spectral data, therefore the idea was to use multi-touch technology to review the pictures more rapidly. Once you can review the pictures, there are many other ways you can manipulate the data, for example sorting based on spectral quality or protein ID."
Mass spectrometry is a way to measure the mass or weight of molecules. Mass spectrometry determines molecular weight by measuring the ratio of mass-to-charge of a molecule that has been converted into an ion. Basically, it comes down to adding a charge to the molecule and placing it in a mass analyzer where ions are then sorted according to their mass ratios. Mass spectrometry can be used to identify unknown molecules as well as revealing their structural and chemical properties.
The results—mass spectra—are presented as vertical bar graphs; each bar represents an ion with a specific mass-to-charge ratio or signature, while the length of the bar indicates the relative amount of the ion. As a result, these ratios or signatures resemble flat line drawings of mountain ranges, with peaks usually highlighted in color; mass spec data can also be represented in three dimensions, making them look like real, albeit candy topped, ranges.
"There is a lot of labor involved in conventional proteomics," Pascal said. "We want to apply this new technology to increase productivity."
Less Tedious, More Efficient
In one demonstration of the potential of the multi-touch screen technology to make reading these signatures less tedious and more efficient, Pascal showed off what he calls the spectra wall, a seemingly endless line of graphs depicting the data for a seemingly endless number of compounds. With one hand, he can brush across the surface of the screen to move the wall back and forth rapidly. Or, with a simple touch, pluck out a specific graph from the wall and pull it into focus to examine more closely. It is deceptively simple and, for anyone who saw the film Minority Report and its larger but similar wall of evidence, remarkably compelling.
From the beginning, the trouble has been that there isn't an abundance of multi-touch hardware—or software—available right off the shelf. As it turned out, there were only four hardware systems available, all very expensive and some without the functional response time that Pascal needed. Also, several systems were problematic because they are essentially flat tables and using them to read spectra would not be ergonomic, Pascal said. Eventually, Pascal and Busby bought a 47-inch flat screen television monitor, with a mounted multi-touch frame equipped with infrared optical sensors at the corners on top of that to capture touch movements on the screen. In reality, touching the screen is unnecessary. Passing your fingers within a quarter of an inch of the screen is all it takes to make things happen.
To make the system work in a laboratory setting, they took the flat screen and mounted it on a motorized drafting table, so they could dial up any viewing angle.
Pascal and colleagues have also experimented with using the Wii remote in a number of interesting ways. The Wii, a popular video game console by Nintendo, has a handheld remote that can be used as either a pointing device or a three-dimensional motion detector. That capability allows players to bowl, for example, and have their arm movements replayed on the screen. The team has used the Wii remote as a kind of air mouse, where it is hooked up to control the cursor, and has programmed the command buttons to scroll through files. In one of the team's more creative moments, it set up Wii remotes as infrared light sensors on a wall, which allowed infrared light pointers to turn the wall into a kind of multi-touch whiteboard.
Even with the hardware capable of handing multi-point touch, Pascal had his work cut out for him. The zoom effect so popular on Apple's iPhone—using two fingers to expand the size of a picture—may seem simple enough, but in terms of software, it's quite involved.
The team's next step is to develop software applications that can leverage the ease of the touch screen system to increase productivity in the lab.
"Our ultimate goal is to be the first to develop a multi-touch system for proteomics," Pascal said. "My job is to find what's out there in terms of software and, if it's missing this or that piece, I'll build it from scratch."
Pascal has been involved in bioinformatics for almost ten years, the last four at Scripps Florida where he was one of the institute's original hires. His group is playfully called Omics Informatics, which basically means he handles informatics needs for any group whose name ends in omics.
Born in Montreal, he has a degree in programming and studied bioinformatics at Harvard. He also plays the piano and electric bass, skills he uses in the rock band he plays with on the side—named, appropriately enough, Sideshow. In that capacity, he's played clubs in Abacoa and is scheduled to play at the Scripps Florida holiday party in December.
"It's tremendously interesting to work in biology," he said, "and I'm honored and very fortunate to be the go-to guy in proteomics because proteomics is emerging now as the important technology. For example, Jennifer asked me to write a program to automate the extraction of spectral peaks from a list of known peptides and associated modifications. I was surprised that a tool like this didn't exist for their specific workflow. This means I can write things that can have a real impact in the laboratory."
As one example, Pascal recently developed new software for Pat Griffin, the head of Scripps Florida's Translational Research Institute, who was using a combination of mass spectrometry and solution phase amide hydrogen/deuterium exchange (H/D exchange), which is an effective method for characterizing protein dynamics, and protein-protein or protein-ligand interactions. But despite improvements in instrumentation and automation, data analysis and display remained tedious.
So Pascal created an integrated software system for the automated analysis and representation of H/D exchange data, calling it HD Desktop. Using the various components of HD Desktop, scientists are now able to analyze and present mass spectrometry data from various instruments. The system has been used successfully for more than a year and is freely available as a web tool.
Pascal is clearly in his element, ready to play his part in the next big thing.
Send comments to: mikaono[at]scripps.edu