DR. YOUNGMAN: Thank you for your comments.

If we can now consider question four. What are the top three elements of a well-coordinated collaboration for the molecular characterization of isolates from all three arms so that we can demonstrate or refute a continuum from animal, food and human origins for specific pathogens and/or resistant phenotypes? Do you want to start with that, Lyle?

DR. VOGEL: I will try. Take what I say with a large grain of salt. I am not a microbiologist, and I am a long way from being a molecular biologist. But I will try.

What are the top three elements? I would say a linkage to epidemiologic information, for number one. Linkage to epidemiologic information for number two. And the same for number three. So I think it is extremely important that we gather epidemiologic information in order to make these -- either to demonstrate or refute the continuum. And somehow, we need to get back to the on-farm sampling as well.

DR. YOUNGMAN: Just so I am clear, when you say linkage to the epidemiological information, are you saying that things that link it to human disease?

DR. VOGEL: Yes. Was their human disease involved for the animal isolates. You know, especially, on-farm. Was their antibiotic use on those farms, et cetera. That type of information.

DR. YOUNGMAN: Thank you for your comments. Awa.

DR. AIDARA-KANE: --- standardization of methodologies and --- of numbering of the PFGE patterns, and timely information sharing. And I also put that linking of typing to other epidemiologic data is important. An important issue to track outbreaks. So epidemiologic also.

DR. YOUNGMAN: Thank you for your comments. Scott.

DR. MCEWEN: I think as Tim and Tom outlined yesterday, there was a lot of discussion about this area in the CDC review, so I would just make a note of that. I think it is relevant to the other arms as well; at least, some of those comments and recommendations. So just to make a couple of suggestions further to that.

Again, I think it is important to be strategic in the use of the resources available for characterization and concentrate on those efforts where NARMS is in the best position to make an impact. And then try to rely on collaborators in academia, or elsewhere, for some specific things. And, again, that was mentioned in the CDC review.

I think it is important to link the efforts. The efforts have to be relevant to human health outcomes, and I think we should try to be broad in all those sort of things that contribute to the burden of illness; not just resistance and treatment failure, but explore some of the other possibilities, including links between resistance and virulence of pathogens.

And, especially, the thing that I think needs attention is better understanding of the role of commensal populations as reservoirs of determinants and explore opportunities for trying to estimate the amount and type of transfer there is of genetic determinants among species of bacteria to human pathogens.

And not just enteric pathogens, but possibly other pathogens. I think that is an issue that kind of simmers in the background and doesn’t get enough opportunity. I think NARMS, with its very rich database of isolates. And, prudentially, epidemiological information, as Lyle has said, is in an excellent position to contribute to that. If not, be the sole investigator at least provide those opportunities for others.

I am not, sort of, ranking mine in threes because I am not, again, a microbiologist. I did have some questions about the extent of efforts in PFGE. But, I think, in answer to the questions yesterday, the group showed that that information is potentially useful, so let’s see how that goes.

DR. YOUNGMAN: Thank you very much. Marissa.

DR. MILLER: My understanding of the molecular characterization program currently in NARMS is utilization of pulse-field gel electrophoresis, and linking through PulseNet. So that is the strategy, as I understand it. And I think that was the opportunity presented to the government, and it has a lot of advantages. It has some disadvantages.

I think that Paula pointed out, and I really do agree with her, that there is going to be an evolution over time. And with the hope in my mind that there is movement towards a micro-array system, ultimately. And as long as everyone is sort of thinking about the future as you deal with the present, I think you will be prepared.

So, in just then thinking about the PulseNet system, I think the number one element that I would put forward is all the isolates from all three arms need to go in in a timely way and be available to everyone for interpretation. That is critical. My understanding is you are moving in that direction and, ultimately, that will happen. And, probably, in the near future.

I think that the second element is that the interpretation needs to be done in concert with the three entities. And that is important that there is discussion, and it is not just another generation of data that is out there. But that there is secondary hypotheses that are formulated. So there is a thought process about what is going on that leads to either field studies, or other collaborations, or other data analysis.

And good examples are the ones that were provided, I think, by CDC of working with David Hooper and Fred Tannover to drill down to that next level and try to figure out what is really going on. And why that is so critical is that PulseField really is a quick look-see that is very useful for outbreak investigations, but beyond that, it doesn’t tell you much. So you have to take it that next step.

DR. YOUNGMAN: Thank you very much. Sue.

DR. KOTARSKI: I don’t know if my background qualifies me as an expert to comment on this, but I will try as best as I can. I have a question about the question. And that is, it says, so that we can -- what are the top three elements of well-coordinated collaboration for the molecular characterization isolates from all three arms?

So that we can demonstrate or refute a continuum from food and human origins for specific pathogens and/or resistance phenotypes. And if I look at the question I can say, well, is that the objective is to refute a continuum or demonstrate a continuum. And I got kind of confused. Well, is this another objective we would like to have as a focus.

So I just offer the question, I was a little confused by that. Maybe you can help me a little on that.

DR. YOUNGMAN: I wasn’t one of the ones that helped to formulate the question, but to my take on this question is, what are the aspects of our molecular characterization research programs that are the most helpful in looking at foodborne disease? What are the things that work the best, in your opinion?

DR. KOTARSKI: Okay. Well, thank you. Again, from my limited understanding, I am very impressed with the fact that you do have the PulseNet system and the opportunity to forge you to quickly go through your different isolates and identify matches, if you will. I think that is a strength and is something that should be supported.

I think another element of that though is an integrated sampling strategy. We have an integrated sampling strategy from human illness, the FoodNet system, and the retail sampling strategy. And then we have another integrated system of sampling, which is in the CAFSHE.

And then we have the NAHMS Program, which is a different sampling strategy than the CAHFSE or the FoodNet and the retail. So they are not completely integrated, but you can take advantage to which each system has its own integrations. You have it with the VetNet and the PulseNet.

The fact that you have good standardization among the different programs is also a top element of that. It allows you to identify isolates that are identical and allows you a platform then to identify whether or not there is an epidemiological link.

But to use that as a major objective, I am not sure that is really the nature of the program, or if, in fact, that -- I am getting a little confused as to -- like I say, I come back to my original thing, what is the point of the question.

But I think that to have the integrated sampling would be very useful to enhance the PulseNet system that you have. And the standardization that you already have, or are putting into place.

DR. YOUNGMAN: Thank you very much for your comments. Sean.

DR. ALTEKRUSE: Well, I would say that the system should be transparent. That is something that Marissa said. I will get back -- transparent and that it should be useful by a variety of laboratories for comparisons. And then third, it should evolve with technology. But I would like to go into what I mean by those things.

In terms of transparency, queries should be possible across all three parts of it, and those will allow exploration of similar patterns that can be useful for deciding about field studies, or case control studies, or on-farm studies, whatever the case might be. And then going one step further for confirming those results in terms of finding a pathogenic bacteria in the environment where the studies would suggest you might find it.

So there needs to be an openness across the three systems to allow comparisons. And then the second thing is that not all laboratories are going to be as sophisticated as the three arms of NARMS are. Sometimes we will want to do comparisons with a laboratory in Thailand, or something like that. So it is not always possible to go with the very best technology.

Because as Shaohua pointed out yesterday, a laboratory in another location may not have a sequencer available for the purpose of comparisons. So for the time being, I do think that PFGE is something of the standard. Each of the presentations mentioned micro-array platforms that they were developing.

And I would encourage the three groups to consult on that so that they have a really well-defined set of resistance genes, pathogenicity, island genes, maybe somatic and flagella antigens. Because these micro-array platforms allow tens of thousands of oligo- nucleotides to be screened simultaneously.

So that is much more powerful than a banding technique for comparisons. And once the platform is established, it allows for high throughput and at very low cost. The cost of these chips is probably in the range of $5.00 to $10.00. But there is a research and development, and consensus building process that has to go on. That is sort of the evolution of new technology.