Interviewed by Marilynn Larkin

Nobel Prize winner Sydney Brenner has proposed several new techniques for obtaining sequence information from thousands of genomes simultaneously; this is in contrast to current technologies, which can analyze only one genome at a time. These techniques, which are being studied and validated by a new company, Population Genetics Technologies, are expected to reduce significantly the cost of studying large populations of genomes. The Wellcome Trust has granted the company a £1.1m Programme Related Investment to enable it to begin development of the technology pending receipt of additional venture capital investments.

SBS News asked Sam Eletr, acting Chief Executive Officer of Population Genetics Technologies and keynote speaker at the upcoming SBS 2006 Conference & Exhibition in Seattle, Washington, to provide an overview of the new technologies, as well as insights into his success as an entrepreneur.

Why the rush to study populations of genomes?
SE: Scientists have tried to discover genetic associations with disease by studying the linkage of a specific disease to genetic markers such as satellite repeats or single nucleotide polymorphisms. Variant sequences typically are discovered by sequencing genomes from a few individuals. The fact that variation is found at abundances of 10% to 50% indicates that the discovered variants are ancient and, therefore, are not the ones causing disease.

For many diseases, this approach has not been very successful; diseases may be due to several different loci, some acting in concert. Moreover, the resolution of such studies is quite low. Three million markers must be used to attain a resolution of 1kb on the genome. In addition, these measurements can be done only on one individual at a time or on small pools of individuals at a time.

Thus, what is needed is high resolution genetic analysis—i.e., sequencing done on a large number (as many as several thousand) of individuals. We chose 4k, or 4096, as the upper limit for the initial implementation of the technology invented by Dr. Brenner, which can be applied to study in detail the candidate genes suspected of involvement in a particular disease. It looks for enrichment of alleles of given genes in an affected population; if such enrichment is not found for a specific gene, then that gene can be definitively excluded.

Unlike other technologies, our method does not have to look at the majority of normal sequences in genomes; rather, it concentrates on selecting and extracting from the population of genomes those regions that contain variant sites that can subsequently be sequenced by either the technology itself or by other means, if more cost-effective.

An important point is that unlike other means of genetic analysis, the cost of using our technology does not depend on the number of genomes being studied; it will cost about the same whether we do a few, 100, or 4096 genomes.

How does your approach differ?
SE: Dr. Brenner has come up with a number of ideas that we’re still in the process of attempting to validate. We haven’t validated all of them, by any means, but we’ve made enough progress to be optimistic. These techniques would allow us to work with as few as 30 copies of each genome of several thousand individuals for an unlimited number of investigations. We would fragment each genomic sample individually and tag the fragments using a combinatorial technique, in a manner such that all the fragments from the whole population end up in one sample, with each fragment labeled with a sequence identifying the individual genome from which it originated.

The resulting few micrograms of DNA would be an immortalized repository of that particular population that our technology would be able to interrogate for specific sequence information, in different sets of genes of interest, over and over again. The archiving advantages of the technology alone are worth the decision to develop it. The main advantage, of course, is that the technology would enable one to study thousands of samples simultaneously, in one experiment, instead of doing as many experiments as there are individuals in a population.

Unlike current pooling techniques that allow simultaneous analysis of multiple genomes, but that provide only population-wide characteristics such as the frequency of gene variation, our technology provides and retains individual-specific information. In addition, we expect our technology to allow handling of much larger numbers of genomes than pooling does, and to have the further advantage of protecting the identities of individuals involved in any population study by allocating them a code that may be kept confidential. We expect it will also be applicable to any collection of DNA molecules and genomes, whether from plants, animals, micro-organisms or humans.

I am not aware of any other work attempting a similar approach. Most efforts seem to be expended on developing cheaper and faster means of studying individual genomes. We follow them with interest, as our unique pooling techniques may some day enable the simultaneous application of such sequencing approaches to whole populations, if they are compatible with our technology and more efficient with respect to particular sequence determinations.

Our technology would also enable the construction of a database and associated computer programs that will allow the selection (and subsequent extraction from the tagged, immortal, population pool) of candidate genes to study, based on the best information that is available. And, of course, we will learn more as we study many genomes.

Ultimately, we hope to provide the instrumentation and tools (alone or together with a partner) that will allow individual research groups to study their own populations. Although our technology deals with a need in human genetic studies, there is no reason why it could not be used to study genetic variation in other animals, microbes, and plants.

As Dr. Brenner has noted, the technology could also be used for a broad range of complex biological problems requiring many parallel analyses. Examples include elucidating genetic changes in expressed genes in many samples of cancer, or understanding the different responses that people have to drug treatment so as to better adapt medications to the needs of individual patients. The technology might, for example, enable the discovery of mutations that are rare in a clinical trial population, but that are responsible for serious deleterious side effects that are discovered only when the drug is very broadly prescribed. Patients that are potentially subjected to such side effects could be screened if these mutations are identified.

This sounds like yet another exciting entrepreneurial venture for you. What factors are key to your success?
SE: I was recently part of a panel of entrepreneurs who were asked to address this very issue. And when we met beforehand to fine tune our presentations so that they would complement each other, we decided that one of us would talk about how important it is to have the right scientific advisors; that another would talk about how important it is to have the right investors; that yet another would talk about how important it is to have the right directors; and, finally, that one would talk about the crucial importance of hiring the right employees.

I was the last to speak. I said ‘yes, all this is very important, of course. But even if you have all the right ingredients, it’s still not enough to insure success.’ I remember that when I began forming in 1980 the company that became Applied Biosystems, there were at least four or five other companies that had people at least as good as those whom I had hired or proposed to hire. Some had access to similar (or the same) technologies as we had; all had similar, savvy venture capitalists behind them; and some were led by people who were much more competent in the applicable sciences and technologies than I was. Yet, we succeeded and they failed, and it was, and still is, very difficult for me to say why. Probably, as a group, we managed in the Darwinian sense to better confront and parry the challenges presented by the outside world, though I couldn’t really tell you why.

What did we have that other teams didn’t? I have thought a lot about this. I think that for a company to really take off, all the people in that company have to have something in common that enables them to work together in a cohesive way to address the fluctuating realities of the market place. Perhaps the manner in which they are brought together is responsible. Perhaps it is because they have been chosen by a founding CEO according to a common set of criteria that has insured that such a group ends up with a common goal, with enough affinity for one another, and with a way to work together cohesively. That’s probably the most important ingredient contributed by the founding CEO, in addition of course to the quality of the people that he or she has assembled.

Another key is to be realistic about what the market expects and what the customer expects and to avoid hype. It’s good to have a vision, but not to hype it.