Name: Marco Marra
Title: Director, Genome Sciences Center, British Columbia Cancer Agency, Vancouver
Experience and Education:
Postdoc with Bob Waterston, Washington University School of Medicine Genome Sequencing Center, 1994-1999
PhD, Genetics, Simon Fraser University, 1994
BSc, Molecular and Cell Biology, Simon Fraser University, 1989
As a postdoc in Bob Waterston’s lab at Washington University’s Genome Sequencing Center, Marco Marra helped sequence, identify, and map genes from numerous organisms. As director of the Genome Sciences Center at the British Columbia Cancer Agency in Vancouver, he now applies genomics and bioinformatics tools and technologies to study cancer and other diseases.
As an early-access customer for Solexa (now part of Illumina), the Genome Sciences Center installed one of the first commercially available Genetic Analyzers late last year. Since then, the group has added two additional instruments.
In Sequence spoke with Marra last month to learn about his experiences with the platform.
What has your experience with Illumina’s sequencer been so far?
We were part of an early-access program with Solexa. That collaboration started in early November of last year. Since that time, we have been working to establish a variety of different protocols on the device, and have realized our first publication, directly as a result of using the device [an article in Nature Methods
, see In Sequence 6/12/2007
We have been working to get together a suite of applications on the device [that] encompasses various RNA-type methods as well as DNA methods.
What are some of the projects you have used the platform for so far?
We have been looking at microRNA profiling, and the focus there has been on various types of stem cells. Another area has been whole-transcriptome shotgun sequencing. So there are efforts underway to try and understand how well that performs. We have implemented serial analysis of gene expression on the platform [as well], which is focused only on a portion of the transcript, usually very near the 3’ end.
How has the instrument performed in your hands, and how has its performance improved over the time that you have had it?
The performance has improved dramatically. Recognizing that we were part of an early-access program, we really didn’t know that to expect from the instrument when we first got it in-house, and it has pleasantly surprised us. The machine performs at least exactly as advertised, and there clearly is additional potential that we look forward to realizing, in terms of increased read length and things of that sort. I would classify the machine as being very robust. We have been pleased enough with it, in fact, [that] we have acquired two more.
Why did you decide to become one of the first customers, given that this often involves some development work?
One of the areas of emphasis around here has been technology development, assessment, and implementation. We view these as credible scientific projects. So being at the forefront of the development of this hugely important technology was something that we actually embraced. We were familiar enough with the concept of the device, and impressed enough by the opportunities that it might afford, that it was a gamble that we felt was well worth taking.
What has it enabled you to do that you were not able to do before?
Generate many, many, many billions of bases of DNA sequence data. Fundamentally, that’s it, isn’t it? So the instrument is a huge advance in terms of cost-effectiveness. Of course there are limitations with the technology, [as] there are with any technology. We are not expecting it to be perfect, especially at this point in time. We are expecting to have to work a little bit to be able to apply it efficiently in certain areas, and we recognize that for certain studies, that it’s probably premature for application.
What studies are those?
Well, it’s not the $1,000 genome yet. But it is, in my view, anyway, a pretty credible first step. So for certain of the types of applications that we view as important to our current research efforts, it’s great. There is transcription factor profiling, for example, looking at the sites of association between proteins and DNA in the genome. It seems [it is] very well suited for that, as well as these various other tag-sequencing based approaches.
Where do you see potential in the future for this? What would you like to use it for in the future?
I want to sequence cancer genomes. We are at the British Columbia Cancer Agency and have a strong desire to apply this directly to the analysis of entire cancer genomes. That’s what we are working towards, and what we hope to do, if not with this device, then [with] one that builds upon it.
Does that mean another device altogether, or a new version of this one?
[It means] in terms of the family of devices, next-generation sequencing technologies. There is something that’s going to come after this, as a consequence of the impact of devices like this on the field. I have been quite happy with the Solexa device, but I am confident that it’s not the end of the development cycle for Illumina, or for anybody else. So I look forward to competing devices that further push the envelope in the coming years.
There are a lot of academic and commercial efforts going on. Which ones do you think have the greatest potential?
I can’t say. I hope that more than one succeeds. As a user of these devices, it’s good to have a choice, and so I encourage them all to go for it.
Have you had a chance to compare this instrument to other high-throughput sequencing platforms?
Not directly. We have based our comparisons essentially not on head-to-head comparisons with data so much, but simply looking at what we know to be true about the performance of this platform versus what we calculate might be true with others. So I can’t give you a sophisticated scientific comparison.
In the future, are you also planning to adopt any of the other next-generation sequencing platforms?
Our concept is that as these become available, and if they are affordable, we would like to be able to evaluate them all. And we would like to be able to acquire those that offer substantial benefits in terms of cost and throughput and ease of use. These are all relevant issues, I think. So if a device came out that tomorrow offered an order of magnitude increase in throughput for the same cost, of course we would have to have a serious look at such a thing.
When you say ’if they are affordable,’ what do you mean?
It’s complex. Affordable means how much the thing costs, and then it also means, if you have the issue of setting up a second type of technology, how much you lose by taking people off something that is working and putting them on something that may not be working, at least not immediately. So there is a cost associated with that, which is an opportunity cost, and also a real cost, in the sense that people are not maybe being as productive as they might have been had you left them alone.
So those are all aspects of calculating the costs. It’s not simply the cost of the machine — that seems to me only one of the costs.