Most lay people reading the results of a mass spectrometry peptide analysis, wouldn’t make head nor tails of it, but with Randy Whittal it’s a different story.
Used to poring over such data for hours and manually putting together the puzzle pieces that when completed, for Whittal the analysis would reveal a brand new biological structure.
Peptides, or small proteins, are a collection of amino acids, strung together to create life. Without these basic strings of biological matter that emerged from the Earth’s primordial soup, humanity – and all life on Earth – simply wouldn’t exist. It’s no wonder that modern-day chemists are looking for the best tools available to help them understand these building blocks of life.
For Whittal, the director of mass spectrometry in the department of chemistry at the University of Alberta, a pencil and paper powered by his own brain was the best tool to do that. Until he found PEAKS software, powered by Waterloo, Ont.-based Bioinformatics Solutions Inc.
Done manually it’s quite a time consuming process.
“You basically sit down with a calculator and a mass spectrum sitting in front of you,” he says. “It could take anywhere between 15 minutes and an hour for a spectrum. Of course, PEAKS comes up with the answer for you in a few seconds.”
The software that Whittal describes as the best he’s ever used for peptide sequencing was invented by Bin Ma, founder of Bioinformatics Solutions and a professor at the University of Waterloo. Since its development, PEAKS has earned a place in the labs of pharmaceutical companies, government research facilities, and academic institutions around the world. It is known for its unparalleled accuracy in de novo sequencing – the process of identifying a completely new peptide that hasn’t been encountered before.
Scientists typically rely on large databases of known peptides when trying to identify one they are studying. The peptides found in the human body, for example, are fairly well known. But the method can lead to errors when a new type of organism is being examined. A database stands no chance of identifying a brand new discovery – and neither does a human.
“Mass spectrometry generates a lot of data about the protein. No human can actually go through it all,” Ma says. “Using de novo sequencing is like having an infinitely large universal database with every possible sequence in it. We’ve made an algorithm to do this efficiently.”
Though Whittal did cross-check many of the results from PEAKS with his own calculations, he admits he couldn’t have done a recent project without some help from the software.
“There’s no way I could have got this manually, but PEAKS was able to interpret the spectrum,” he says.
Whittal and his associates at the University of Alberta were able to identify a new peptide found in pigs that could be put to use as a safe food preserver or in certain anti-biotic therapies for humans. Known as Carnocyclin A, the peptide has anti-microbial properties that help to prevent harmful pathogenic bacteria from forming.
It’s just one example of the usefulness of PEAKS. The software has also been used by a research team at the Oregon Health and Science University in Portland to determine the role a key marker of down syndrome. The research could be used in the future to determine if an unborn baby has Downs syndrome by sampling the mother’s blood.
Ma founded Bioinformatics Solutions in 2006. The 35-year-old was awarded as the best young innovator at the Premier’s Innovation awards hosted by the Ontario Ministry of Research and Innovation in March. The award provides $200,000 to the company for product development and marketing.
Related story: Read about another Premier’s Innovations Awards winner – Former IT employee turns hobby into award-winning business.
“It’s really important, not only financially, but it gives us the reputation,” Ma says. “We entered this market much later than some competitors, so we need a bigger marketing effort.”
PEAKS owes its success to its ability to accurately employ de novo sequencing to identify peptides. De novo sequencing is done examining the mass spectrometry scan data of a peptide. After dividing a peptide into different fragments and outputting its scan spectrum, the peak in the spectrum is the mass for that fragment of the peptide. By comparing the different masses from many different fragments of the peptide, it is possible to derive the entire sequence, in order.
But PEAKS (the software) also includes a database search component. That provides an option to scientists who can’t get high-quality mass spectrometer readings, Ma explains.
“In this market, there are a lot of users who use databases to do their research,” he says. “If you have a long peptide, as long as you can get several high-intensity peaks that match the sequence of the peptide, the database method works. For de novo, you’d need almost all of the peaks.”
The chemistry lab at the University of Alberta is now using PEAKS regularly for its database identification capability, Whittal says. This is the most often used feature of the software for the researchers.
When de novo sequencing is required again, Whittal is confident the software will perform well.
Bioinformatics Solutions plans to add new features to its software in the future, Ma says. That includes protein quantification – a feature that will tell researchers what protein they’re dealing with, but whether the concentration of that protein is different across many samples.
“If you can find that, then you know that probably has something to do with the disease,” he says.
It’s also useful for drug research because researchers can hone their drug to target a specific protein.
“You want to kill the cancer cells, the over-expressed proteins,” Ma says. “Once you can identify them, you can try to kill them.”
A preliminary function of this feature was recently released for PEAKS.