Newsroom
In The News
The librarian of our genes: Ottawa researcher Miguel Andrade is writing 'genetic dictionaries,' organizing a labyrinth of genetic information found in stem cells so that other researchers can scan through millions of pieces of DNA.
By Tom SpearsThe Ottawa Citizen
Saturday, June 18, 2005
On his way to becoming a geneticist, Miguel Andrade grew up in Spain, studied and worked in Cambridge and Heidelberg, and married a German woman. The couple had two young children. It was time to pick a country.
Maybe somewhere English-speaking, they felt.
United States? The political and social climate didn't attract them.
What about Canada?
They looked into Toronto and Montreal, and suddenly a job opening popped up in Ottawa. It seemed like a smaller city, and Canada seemed welcoming -- a good family atmosphere, friendlier than Germany, "not as friendly as Spain, but better organized."
Today Mr. Andrade speaks to his three young daughters in Spanish; his wife, Carola Kuhne, speaks to them in German; the kids speak both among themselves, plus some English picked up from their Ottawa playmates; he and his wife both speak each other's languages pretty well though he can't write German ("Actually I can, but people would laugh at it") and the couple often lapses into English as a linguistic meeting place.
But his most valuable language is the language of genes.
Now, at 38, he is the master librarian of all our genes even though he won't ever discover a cancer gene or repair a heart attack gene on his own.
He eagerly comes to work each day to help other scientists scan through vast piles of genetic data, using the power of computing to sort out genetic material in a nearly new science called bioinformatics.
He says this about his chosen field: "There won't be a Nobel in bioinformatics. And if we help cure some disease, it won't be directly. It will be people using our databases."
So why does one of Canada's most famous gene discoverers say Miguel Andrade "is the vanguard of a generation" in exploring the genes that make us tick?
That view comes from no less than Michael Rudnicki, discoverer of the Pax7 gene that turns stem cells into muscle cells, and scientific director of the Canadian Stem Cell Network -- scientists learning to help sick and injured bodies repair themselves.
While the other gene-seekers are doing their work, Miguel Andrade is writing genetic dictionaries and building a reference library that allow each scientist to embark on his or her project at full speed.
The young Spanish scientist works at the Ottawa Hospital Research Institute, specialising in the crucial use of computers and databases in molecular biology -- the machinery of individual cells. He covers many facets of gene work, but his own creation is StemBase, which relates to the genetics of stem cells -- cells the body stores in an unfinished form to use as construction material later. A major push in modern medicine is to learn whether doctors can direct these cells to repair tissues that can't be repaired today, such as injured brain tissue and heart muscle.
"We can do discoveries," he says. "By looking at these stem cell data for example you can say: Oh, look at these two genes! When one is expressed (i.e. working actively), the other is absent. But these two genes are very similar. So, wow! Why do they hate each other?
"And this can happen in 30 seconds. You just notice.
"And then of course someone has to check that. Checking may be impossible, or may take a month, but sometimes..." He leaves the possibilities dangling.
"We use data that is massive -- it could be complete genomes, all the genes in an organism (or) the structures of the proteins," he says.
Assembling all this fast-spreading knowledge is genes is "like writing many heterogeneous dictionaries, many dictionaries of different views of the objects of microbiology."
Once upon a time -- say, until the early 1990s -- gene work was slow and tedious. Individual scientists slaved over bits of genetic material, a single glass dish at a time. When someone teased out the role of a single gene among the tens of thousands in our bodies, it was a great discovery in the science world.
Today the need for creativity is still there, but automation and computers allow a researcher to scan through millions of "bases," or the smallest pieces of DNA coding material, at a time. To do this job without computers would be like asking Canada Customs and Revenue Agency to check everyone's income tax returns with a pencil and paper.
"We need that computing power. It's just far too much information for a human mind to understand," says Michael Rudnicki. "When we did our very first experiments we would grab our printouts and lists of genes and try to read. It's like trying to read tea leaves. You need a higher order of sophisticated analysis to make sense of it all, because the human mind just doesn't work that way."
There are some 300 fully deciphered genomes of bacteria, and about 30 more of organisms with more complex cells -- humans, mice, rats, yeast, cattle, and the mosquito that carries malaria. All have at least millions, and likely billions, of bits of DNA to catalogue in a way that lets scientists search quickly through it.
The trick is that a gene in a human and in a mouse may be almost identical. For instance, even the humble yeast cell has a gene that governs calcium, and it's remarkably similar to the human cystic fibrosis gene. Comparing from one organism to another is one major foundation of gene studies today: Figure out what role a gene in one species plays, and it's an enormous clue to the genes of another animal.
"For example you have the human genome, you have all of those genes -- they are tools for a microbiologist to make careful experiments.
"You have to direct your efforts. You don't want to make experiments on a gene that is very similar to one that has been already studied," he says.
"Or you want to say: This one looks like it is responsible for this genetic disease." Databases can tell where this gene sits in our chromosomes, and what proteins it produces to carry out its work through the machinery in our cells.
Gene research, he agrees, is like a series of explorers setting out across a very large and unfamiliar country: Each can use the database to see where others have travelled and what they have seen.
"Some people call it a vast ocean of information and say, 'This is terrible. It's too much.' It's the opposite: I think the more information we have, the better. If you have a library, the bigger the library is, the better... The molecular biologists are becoming more and more aware of the need for tapping into that amount of information."
"But you need a better indexing mechanism and you have to be more clever to organize information."
The key to his job is to figure out a useful pathway through all that mass of stored information.
"You have to think a lot. Once you develop a method, then everything seems to be very easy. So you go to the biologist and you say, 'Look, I found this interesting gene for you!' And for him the perception is that we did that in 10 minutes, but that's not true. There was hard work before, you see."
The Andrade lab's main business is in stem cells.
About 150 scientists who work on stem cells have sent samples of those cells to the Ottawa Hospital Research Institute for analysis. Stem cells come in many shapes and sizes: From mice and men, from adults and embryos, from blood, brain tissue, bone and muscle.
All these share two abilities: They can divide and make copies of themselves, or they can mature from their current state -- a blob that doesn't serve any immediate purpose -- into a finished cell of the kidney, or blood, or eye, and so on. But as the stem cell waits to be assigned a job, it's alive and some of its genes are working. But these aren't the same genes that will begin working later, when it begins to take on its mature form, or later again, when its final form is attained.
The many scientists have sent in their cell samples with the same question: Which genes are "expressed," or working, in each of them? And just as important, how hard are these working? And when in the cell's life do they go on and off? It's like asking which lights are turned on in a house, but also how many of those are set on dimmer switches and timers.
Pearl Campbell's lab at OHRI does the analysis work on these samples.
"Our whole project is to find out what makes a stem cell a stem cell," she says. She analyses gene expression with a microarray -- a chip with DNA dotted all over it. An RNA extract from a stem cell reacts with the dots of genetic material on the chip. This shows which genes in the stem cell are "switched on" and which are inactive.
She's interested on what genes control this cell's unique ability to shift form. "But also, we're interested in what triggers them to make the choice" to change their form today instead of tomorrow.
And a key to this appears to be not just what's switched on, but how switched on it is -- the level of "expression."
Her current project is to look for transcription factors -- bits of protein that lock onto a specific part of a gene and tell it when to switch on, and how far to go. This is called "regulation," and mapping all these factors and their targets is the task of the International Regulome Consortium.
Mr. Andrade's group take over and builds the answers into a set of information that can be stored efficiently, "and we try to offer ways to query the data, so that the researchers can, for example, see which genes are particularly expressed (active) in the stem cell," and not in the finished produced of that stem cell. That knowledge would be key to figuring out the crucial question that's still unknown: What makes a stem cell turn into its finished version? Answer that, they hope, and doctors may some day be able to direct out own bodies to rebuild broken and scarred tissue with our own cells.
The Campbell-Andrade partnership is called the Ontario Genomics Innovation Centre.
"We do analysis as well as storing the data," Mr. Andrade says. For instance "if two genes are present always together, then you can presume that they are doing something together."
He doesn't develop the software. His job is be the idea guy, picking the directions, "thinking what is reasonable or what isn't interesting."
Besides StemBase, his "library" keeps other gene-related databases, such as those created by the U.S. National Center for Biotechnology Information. These other centres are happy to have him share their load, under license. In a good day at the office, he enjoys the people, and the chance to happen suddenly upon a useful chunk of information hidden in all those billions of morsels of DNA.
"Another exciting moment is when you discuss with a geneticist, and you say something and they say suddenly Wow! That's it!
"We know everything and we know nothing at the same time, because we work with all the genes and all the proteins. But then we have to go and talk to the guy who's the expert in this protein. You have to be humble because the other guy has been working 20 years on this one protein.
"Then there is a moment where we say, 'We saw this in the data,' and the guy says, 'Yes, that's interesting. I'll have to check that.'
"That's a very nice moment, very exciting. And even better when it works."
Note: Reprinted with permission of the Ottawa Citizen