Compugen Ltd (CGEN) 2010 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by. Welcome to the Compugen Ltd. quarter 2010 financial results conference call. All participants at present in listen-only mode. Following management's formal presentation, instructions will be given for the question-and-answer session. (Operator Instructions) As a reminder, this conference is being recorded July 27, 2010.

With us on the line today are Mr. Martin Gerstel, Chairman of the Board; Ms. Anat Cohen-Dayag, President and CEO; and Ms. Dikla Czaczkes Axselbrad, CFO.

I would like to remind everyone that the Safe Harbor language contained in today's press release also pertains to all content of this conference call. If you have not received a copy of today's release and would like to do so, please contact Dikla Czaczkes Axselbrad at telephone number 9723-765-8595.

Mr. Gerstel, would you like to begin, sir?

Yes, thank you. First, on behalf of my associates and all the employees of Compugen, welcome to our second-quarter 2010 webcast and conference call. Also, I am very pleased to note that as of now have many more participants on this call than we have ever had on any prior calls, and we thank you for that. Welcome to the new participants.

Following some introductory remarks by me and Dikla's highlighting of a few key financial items, the majority of today's prepared remarks will be devoted to a slide presentation by Anat providing additional information about our discovery on demand capabilities, with a specific focus on and using as an example the discovery of CGEN-15001.

15001 is an extremely attractive product candidate for autoimmune disease, transplant rejection, and oncology; and therefore it is not surprising that this molecule, which was the subject of a press release by the Company earlier this month, already has resulted in a very substantial level of interest by the industry, with many companies contacting us for additional information. Perhaps even more importantly, for those that continued to have doubts, this discovery should provide very clear and solid and convincing evidence of the uniqueness and power of our predictive discovery capabilities.

I think this will become very evident to you as you listen to Anat's slide presentation. If you do not already have access to the presentation on your computer, I encourage you to go to our website home page, www.Compugen -- I'm sorry, dot-CGEN, cgen.com, and follow the instructions under Events; or follow the alternative instructions in today's press release.

In prior quarterly calls, I have attempted to provide an overview of what we were trying to do, where we were, how we got there, and why what we were trying to accomplish should result in substantial medical and commercial success. Today will be different, since Anat will be focusing solely on how we are now in a position to capitalize on these past investments and accomplishments.

From a strategic standpoint, it is important to note that Compugen today is clearly in the best position in its history. This is based on our substantial and very unique infrastructure in terms of scientists, technologies, and candidate discovery platforms, resulting from our long-term focused pursuit of the underlying sciences and which is continuing to be improved and expanded; on the increasing recognition by leading companies in the industry of these capabilities; on our growing in-house pipeline of attractive early-stage product candidates; and on our truly unique capability to undertake discovery on demand collaborations in a very wide range of diagnostic and therapeutic areas of high interest.

And of course our current financial situation. With cash on hand, including the current market value of our remaining Evogene shares, that would be sufficient to provide funding at our current level of operations through year-end 2013 even in the absence of any of our anticipated research revenues, or milestone payments, or any other revenue or cash source.

The combination of a well-established strategy and the required elements in place to accomplish its goals, and with available capital that should provide us the time and flexibility required to achieve these goals, is definitely a first for Compugen.

I will now turn the call over to Dikla, our CFO, to briefly discuss some key facts regarding our financial results and status, which will be followed by the main event of this call, Anat's slide presentation as previously described. Following Anat's presentation, all of us will be available to answer any questions you might have. Dikla?

Thank you, Martin. In the second quarter of 2010, Compugen revenues increased to $800,000 compared with $225,000 for the second quarter of 2009. For the first six months of this year, revenues increased to $925,000 compared with $225,000 for the same period last year. The substantial increases for 2010 compared with last year results from the recognition of initial fees and research revenues from certain collaborations. Also, it should be noted that reported revenues for all periods do not include amounts received from governmental and other grants, which are accounted for as reduction of research and development expenses and not as revenues.

With respect to our current cash status, our short-term target remains to be cash flow breakeven by year-end 2011. We ended the second quarter of 2010 with $21.1 million in cash and cash related (technical difficulty) representing a decrease of some $1.9 million compared with the end of the first quarter of 2010.

For the full 2010 calendar year, we expect that our net cash usage will not be more than $5 million, which could be further reduced in the event of additional research revenues, initial fees, or milestone payments during the remainder of the year. All of these amounts for cash and cash-related accounts do not include the market value of the some 1.1 million Evogene shares held by Compugen.

Compugen continues to have no long-term debt.

Before concluding, I wish to remind you again that the slide presentation that accompanies Anat's remarks can be accessed on our website. In order to access these, go to our homepage at cgen.com and then to Events, and click on the presentation link.

Please note that in the window which opens after entering your personal information, you should select the bottom option, Web Participants Application. This option will enable you to contact without installing software on your computer. With this, I will turn the call over to Anat.

Thank you, Dikla. As Martin mentioned in his introductory remarks, I would like to provide you with some insight into our discovery on demand capabilities. In doing so, I will be using as an example one of our most recent discoveries, CGEN-15001, which was the subject of a press release by the Company earlier this month.

First, a very brief introduction to the B7/CD28 protein family. In general, all proteins either remain within the cell where they were expressed or are secreted into the bloodstream, such as insulin, or are attached to the cell surface. The B7/CD28 family of proteins belongs to this latter type. They are attached to the surface of cells, primarily cells of the immune system.

B7/CD28 proteins are known to play a critical role in the defense of the body against disease. Essentially they modulate the immune system in order to protect the body's self cells and to destroy non-self cells such as tissue cells introduced by organ transplant. In addition, some of these proteins are found on cancer cells and apparently are used by the tumor to block the immune system from attacking it.

Due to this involvement in immune response, these proteins play a role in many pathological conditions including autoimmune diseases where the immune system turns against self cells, such as in multiple sclerosis, diabetes, and cancer. This is the reason why this family of proteins is of very substantial interest to the pharmaceutical industry and led to our decision to utilize our discovery on demand capabilities to attempt to discover new B7/CD28 protein family members.

For our initial efforts on this program, we decided to focus on that subgroup of B7/CD28 proteins that are expressed specifically on cancer cells, since such proteins would have potential therapeutic usage not only in autoimmune diseases and transplant rejections, but also in cancer.

Existing approaches for discovery of B7/CD28 proteins in general involve traditional bioinformatics combined with experimental approaches. Family members discovered in the last few years were largely found by sequence homology analysis and experimental verification. However, current attempts at discovery of additional novel members are much more challenging, since as with all traditional discovery approaches, the easier ones -- the so-called low-hanging fruit -- are found first and then it gets more and more difficult.

Those family members that are still believed to be left to discover probably have very low sequence homology. Otherwise, they would have already been discovered by the traditional methods that have been employed by researchers in the field. Also, available public computational tools are largely insensitive to key known B7 characteristics, and therefore are unlikely to aid in the identification of those B7 members that have low sequence homology.

So, how did Compugen discover new family members? And specifically, how were we able to identify a novel B7/CD28 family member?

First, we employed our LEADS infrastructure platform, which in addition to other capabilities predicts the human proteome. Even though the LEADS infrastructure platform is one of Compugen's earliest platforms, it is continuously updated with improved algorithms and new data and remains a core competitive advantage for our protein and certain other discovery capabilities.

Use of the LEADS platform provided us with a starting point for our program, an in silico proteome consisting of what we believe to be all the possible human proteins that could exist. Since as previously mentioned all B7/CD28 family members are membrane proteins, we next applied tools based upon machine learning and additional algorithms to identify from all the predicted human proteome those proteins that are predicted to be membrane proteins.

Next, we applied algorithms developed by our scientists to identify and rank characteristics of known B7/CD28 family members based on the protein family's genomic information and protein domain. These algorithms, based on extensive research by Compugen during earlier years, are a key component of our protein family discovery platform, a platform which has not been formally announced since it is still under development, but has been used very successfully in this activity.

So at this stage, we applied the B7/CD28 characteristics algorithms that we developed to our prediction of all human membrane proteins. And the result was our prediction of probable B7/CD28 family member proteins.

Next we screened probable B7/CD28 family member proteins for their expression on cancer cells. This was done through our second infrastructure computational platform that allows the analysis and prediction of expression of various predicted proteins in healthy and diseased conditions. The result of these tests was our prediction of putative targets for cancer treatment.

These in silico predicted family members were then prioritized for experimental validation and analysis, based on their relative scores to be membrane proteins, to be members of the B7/CD28 family, and to have potential to serve as cancer targets. This process resulted in the selection of CGEN-15001T for experimental validation.

On this slide in the diagram on the right, the orange bars represent the cell membrane and the figure resembling a twisted rope represents a membrane protein, such as CGEN-15001T, attached to the membrane. Also as shown, a piece of CGEN-15001T is outside the cell; and this is termed the extracellular domain.

It is this extracellular domain that can serve as a target for an antibody totally specific to that extracellular domain, that will attach to it and then by various mechanisms attempt to kill the cancer cell.

With respect to serving as a drug target for oncology, on this slide we see how the transcript for CGEN-15001T is very strongly differentially expressed in small-cell lung cancer. Also, as we predicted, the respective recombinant protein was found to localize to the membrane of cells expressing it.

These results are very encouraging. However, they are very preliminary, and therefore additional studies are now underway. These additional validation studies for use of CGEN-15001T as a drug target for monoclonal antibody therapeutics in cancer are now ongoing.

Although CGEN-15001T is very promising as a drug target for cancer, as of now the most exciting aspect of this discovery program is that we have created an additional novel therapeutic entity based on it, CGEN-15001, for potential use in autoimmune diseases and transplant rejection and which was the subject of a press release earlier this month.

Before providing additional information regarding CGEN-15001, a few words about autoimmune diseases and transplant rejection. Sometimes the immune system mistakenly attacks its own self molecules, resulting in autoimmune diseases such as multiple sclerosis, rheumatoid arthritis, diabetes mellitus type 1, or others. In other cases, such as transplantation, there is a need to shut down normal immune response to non-self molecules introduced by the transplant.

The majority of approved drugs for both autoimmune disease and transplant rejection act by suppressing the entire immune system. A more targeted shutdown approach such as offered by CGEN-15001 would be more specific and therefore clearly advantageous over existing medication. For example, it is likely to cause fewer side effects. Therefore, such a targeted shutdown approach is a major area of therapeutic need and a key focus for many pharma and biotech companies.

The new Compugen molecule is based on the extracellular portion of CGEN-15001T as shown on this slide. It is a soluble protein that can act as the therapeutic agent itself. CGEN-15001 was predicted by Compugen to interfere with certain immune cell processes, such as those occurring in autoimmune diseases as well as in transplant rejection.

After producing CGEN-15001 we began the experimental validation of its therapeutic potential. We first tested the molecule in vitro. Our goal was to demonstrate that this molecule can inhibit the immune system response that occurs in autoimmune diseases, such as multiple sclerosis and diabetes, when the body attacks its own self molecules.

For this purpose we measured typical T-cell responses -- T-cell proliferation, differentiation, and cytokine secretion, processes known to occur in T-cell modulation. Here are examples of the results.

In the upper left graph, we measured the T-cell proliferation. The y-axis relates to an increased level of proliferation and the x-axis relates to an increase in concentration of the molecule tested. As expected, adding a negative control molecule -- represented by the red line -- had no effect on the multiplication of T-cells.

As a positive control, we used a known B7 family member, represented by the green line, which resulted as expected in the reduction of cell proliferation as a function of its concentration. The blue line represents the results obtained for CGEN-15001, which shows -- as we predicted -- a similar reduction in cell proliferation.

In the lower right graph, we measured the effect of CGEN-15001 on cytokine secretion from T-cells. The y-axis relates to the level of cytokine secretion and the x-axis relates to the various cytokines that were tested.

As you can see, all of the red bars representing the negative controls show high cytokine secretion, which is typically associated with T-cell activation, while the blue and the green lines -- the positive control and our CGEN-15001 soluble protein, respectively -- as predicted inhibited cytokine secretion from the activated cells. These promising results and additional results from studies not shown in this presentation clearly indicate that the protein inhibits an immune T-cell response as we predicted.

Our next step was to confirm the potential of this protein to serve as a therapeutic drug for various autoimmune disorders. We selected multiple sclerosis as the first of these models to evaluate this potential.

The animal model we used represented relapsing-remitting multiple sclerosis disease. MS stands for multiple sclerosis. This disease is the most common form of MS; and moreover, the clinical and pathological symptoms present in this well-recognized animal model are similar to those in human disease.

Our motivation to perform such a study was to determine whether our molecule could prevent relapses in MS, which is a major goal in the development of treatment. This is because multiple sclerosis relapses result in recurring attacks of clinical symptoms and lead to aggravation of existing symptoms and to the appearance of new ones.

This graph demonstrates the results we observed with CGEN-15001 when it was administered before disease onset in this animal model. The y-axis relates to disease severity and the x-axis relates to the number of days after disease induction. The correlates of the graph are similar to what you have seen before, only now the positive control -- the green line -- is representing a different known active molecule.

As you can see, the negative control -- the red line -- has no effect on the disease, which starts at day 10 and peaks at about day 22. It is then remitted; but as you can see, additional disease relapses occur with time. Under these disease conditions, the animals of both the positive control and those that were administered with CGEN-15001 demonstrated a delay of disease onset from day 10 to around day 15, as you can see, and a significant decrease in disease severity.

Additional exciting results have been achieved to date as shown on this slide. In this case, CGEN-15001 was administered in the presence of the disease at day 20, what is called a therapeutic mode of administration as opposed to the preventive mode that was shown in the previous slide. The y-axis relates to disease severity and the x-axis relates to the number of days after disease induction.

As you can see, the negative control -- the red line -- as expected has no effect on the disease, showing spontaneous relapses and remissions. However, the administration of CGEN-15001 completely abolished spontaneous relapses essentially at day 50 and onward.

These animals are still under evaluation, and I am very pleased to report that even now, about two months after administration of CGEN-15001, the animals are still in full remission. These experiments provide strong evidence of the therapeutic effect of CGEN-15001 in multiple sclerosis diseased animal models and as such indicate its likely relevance for use in human therapy.

It is worth repeating what Professor Miller from Northwestern University, a leading expert in the field who supervised the studies, said about our molecules. The effect of this molecule "is quite dramatic." It "may prevent disease progression as efficiently as immune tolerance induction." And, "These findings, together with those demonstrated in our earlier studies, are unique among the molecules targeting the B7 family of co-stimulatory molecules that have been published to date."

It is therefore not surprising that we're now seeing a very high level of interest in this molecule by numerous pharmaceutical companies.

As mentioned earlier, the majority of FDA-approved therapies in this general area focus primarily on global inhibition of immune inflammatory activity. In comparison, CGEN-15001 is expected to be devoid of general immunosuppressant effects, and may therefore have a significantly decreased risk of opportunistic infections occurring while treating the disease condition.

Also, based on additional results from studies not shown in this presentation, it appears that CGEN-15001 has the unique ability to inhibit the production of undesired immune responses while promoting preferred immune responses. This is a unique feature among molecules targeting B7/CD28 molecules and, if further confirmed, would represent a major medical and commercial advantage.

In summary, I hope I have demonstrated by this specific application Compugen's discovery on demand approach, which is capable of discoveries in fields of potential industry interest, specifically those where traditional discovery approaches are not succeeding. In this specific case, our efforts to date have resulted in the entirely in silico prediction of two previously unknown molecular entities from the same core discovery -- one being the entire molecule that was discovered, and the other based on a portion of such molecule, with applications in many important therapeutic areas including oncology, autoimmune disorders, and transplant rejection.

Looking to the future, the protein family discovery platform used in this program, which as previously mentioned is still in development, may be applicable to additional protein families of significant clinical and industry interest and thus adding to our capability in future discovery on demand collaborations.

Furthermore, with our powerful infrastructure platforms such as LEADS and MED and an additional 12 announced validated discovery platforms, we are confident that Compugen has the capability to uniquely address discovery challenges across an extremely broad range of unmet needs in diagnosis and therapy.

With that, I thank you for your attention and we will now be glad to answer any questions that you might have.

(Operator Instructions) Brett Reiss, Janney Montgomery Scott.

Good afternoon and thank you for taking my question. I will leave for other callers the questions on the slides, which I appreciate.

I got a number of questions. The $800,000 in revenue this quarter, where did that come from? Is that from the Pfizer collaboration, or from other collaborations?

No, as we have stated in the past we are not going to be announcing specifically the receipts or identifying minor amounts such as these. Obviously, when and if very significant numbers come in, it will be required to be announced.

But as we are in the situation that we are right now, as we mentioned in the past, we anticipate -- I think we said last year that we anticipated during the second half of this year to begin seeing some early milestones and other fees. And I think at that time we stated that we were not going to -- it is not in our best interest or our clients' best interest to be specifically identifying the agreements that they go with.

Okay, fair enough.

I also might mention that unlike most companies in this situation, we did not put out a press release saying that our revenues quadrupled, or went up 400%, because it must be taken into perspective. But we are very pleased to see the beginning of these kinds of revenues.

And for a company of our size these are meaningful amounts; but relative to the ballgame that we are playing in these are not meaningful.

So from one standpoint, as I said, happy to see them. They are meaningful for us financially. But they are not meaningful from the standpoint of the strategy that we are pursuing.

If a couple years from now we say, wow, $800,000 really mattered, it means that we were unsuccessful as a Company.

(Operator Instructions) Ronald Urvater, Ormed Capital.

Thanks and congratulations. It looks like this recent review is really stunning in terms of the ability, I guess as I understand it, to much more specifically be able to modulate the immune system. My question really goes to some of the prior announcements though, that you surfaced in the past.

Can you comment generally speaking? There has been over the last 18 to 24 months many significant announcements regarding retinopathy and IBD and some of the other clinical applications you have talked about. Can you -- it's a two-part question.

Number one, are all those relationships ongoing? I presume so.

Secondly, in terms of the collaborative process with your partners, I tried to get this last time. Do you take into account -- do you get information from your partners through the dialogue which then allows you to refine or enhance your predictive models based on the feedback they give you? Or is it -- or alternatively if it doesn't work out, you simply move on and are not necessarily able to ascertain why something may not have worked out?

I'm just really curious what the feedback mechanism is between your partners and the Company in terms of enhancing and refining the predictive models.

Okay, so I'll relate to your second question which is -- I think one of the big assets of Compugen is the fact that we can learn from failure and from success. Actually even more from failures. And we can improve our models to have less failures in there next time we use them.

So in general, we do have feedback loops that include feedback that we get here internally by our -- by the validation studies that we are doing. But also as you mention from the evaluations that are done by the partners.

Of course, in all the collaborations we are aware of the results. We share the work plans that are done to evaluate our molecules. And we do have periodically phone calls and reports that -- so we see everything.

So in general, we take all these pieces of information from all the partners and all the internal work that we are doing here and incorporate that to our models in order to make them better.

Just to refine that a little bit more, you mentioned when things don't work out -- and I also believe sometimes failure can be very educational. In terms of the feedback that you get, does it relate more to the physiology that you're not necessarily being able to ascertain properly? Or is it more the computational process?

In other words if something doesn't pan out the way you expect it to be, is that because of there is additional physiological understanding that is required? Or is it more the way in which you go about predicting it from a computational point of view?

As you rightfully pointed, in some cases it may be also that physiology that is there and cannot be predicted. But I think that in most of the cases, we can get some better understanding and incorporate this information to our model.

So, yes; sometimes it is internal failures here that we didn't model correctly, and we try to catch all these before we announce something. So usually when we announce a molecule we are pretty sure that it does what we are doing, just because we have already evaluated it.

But in general, we can incorporate the message almost from any feedback that we get into our system.

Then finally, just on my first question, so these past collaborations, are they all pretty much ongoing at this point? The ones, the retinopathy, etc.?

As I mentioned, we don't specifically state. We are in a somewhat awkward situation here from the standpoint that for most small biotech companies, life science companies, that is probably the most important question that you can ask. Because they've got a few discoveries, they've got a certain way of discovering things. And if they fail it may well be the end of the company.

Our strategy here is entirely different. And that is one of, based on what we have done in the past, we are expanding and deepening our discovery capabilities across new areas. And in the areas we are working, we are getting better and better all the time.

We are -- our platforms are not based on technology. Our platforms are based on an understanding of the science that we just continue to move forward, the predictive aspect of it. So, that is on one front.

On the other front, we don't have any interest at the present time in further developing and marketing our own products. That means that it's critical for us that -- the most important thing for us is to have a group of pharmaceutical companies, a growing group of pharmaceutical companies, that appreciate and values what we can do.

Because we know that our capabilities are going to improve over time. We know that we are going to have more and more product candidates coming out. If we have companies that are learning more and more about us, that to some degree -- on some of these early arrangements with our companies, the issue as to whether the specific product that they licensed in failed or worked is almost irrelevant. The question is -- how do they look at Compugen after they have given us that try? After they have put their -- because again we are looking at this on a long-term basis.

And I can say without a doubt -- and Anat, please correct me if I am wrong -- I cannot think of a single company that we have dealt with that was not impressed with our capabilities and enthusiastic about trying to find additional things that we could do together, other than those that got merged or disappeared.

But those that are still around right now -- I don't want to comment as I said, and I don't think it really matters to comment on these, because it sends the wrong impression. Personally -- and I hesitate to say this, because I don't want people to get the wrong impression that I am trying to say that it has ended. Here I -- already getting into trouble here.

To me, there has been much too much focus on the Bayer -- on the Pfizer agreement from the standpoint of what is its status, what is its status? All right?

The issue there is a company like Pfizer, after kicking the tires here, said -- you know what? This is a company that we would like to see how their predictive capabilities work. All right?

And that program is -- I guess I can say that program is ongoing. We are not going to give statements as to what the stage is and when things are happening or whatever.

And I can say that Pfizer, the people that we're interacting with at Pfizer, are giving us very, very positive feedback about the caliber of our predictive capabilities, about the caliber of our people.

This is the key issue. Because the ones -- the things that you're seeing now are the -- it's -- well, let me stop there.

But it's -- I think the focus on individual current agreements and whatever, I am convinced without any doubt that the current agreements that we have in place that are active more than justify the current market value of our Company. All right? Now I'm not going to say individually what is happening or whatever; but they more than justify the current value of our Company.

But that isn't what the true value is. The true value is what kind of arrangements and what kind of collaborations are we going to have in the future as our capabilities continue to expand and the recognition of the industry continues to move in a positive direction.

And the last comment that I will make on this is to say that clearly when the book is written on Compugen, the discovery of CGEN-15001 will be seen as perhaps one of the true inflection points -- whether it makes it or not.

This is another thing that is hard for people to understand, because of the industry we are in. For most small companies the absolute question that you live or die on is -- will their molecule make it? All right? So that is what everybody is kind of -- their head is just so focused on.

That is not the case. Obviously we have to have successful molecules; but we have a capability to continue to put the -- to [plan].

And the reason that I say that CGEN-15001 is an inflection point -- whether or not it actually ever gets to the marketplace -- is because this is exactly the type of molecule that enormous amounts of monies have been spent, are being spent, in trying to find. And no one else has been able to come up with something like this.

And the molecule that we came up with was completely novel. It is not a minor modification of something that is out there. It is a completely novel molecule in this most important pathway.

Right, I understand, Marty. So essentially what you are saying is the Company has been successfully able to engineer an ongoing dialogue with the industry, with feedback and progress in terms of the dialogue itself. And that is really the -- that is the real value on a long-term basis here (multiple speakers).

Exactly, yes.

Right. Okay. Thank you.

[Kayna Kay], [Shawdon] Capital.

Thank you for taking my questions. My first question is for Anat. In your slides you showed the immune knockdown. I am wondering if you can tell us what animal model that was achieved in.

The animal model that was used is EAE, experimental autoimmune encephalomyelitis, which is -- in this specific model we used relapsed-remitting animals. For these models, sorry.

Okay. My second question is, you presented the case study for that molecule within your discovery on demand platform. So I am just wondering what motivated you guys to look for that specific molecule.

This was our understanding, that this area is of significant interest to -- both in the research and also in the pharma industry. And we knew that if we will be able to discover a novel molecule for this family member it would have a lot of interest in the outside there, in the pharma industry.

So this is something that we knew will have an impact. Also, the fact that we could come up with a discovery that is actually a discovery of two molecular entities that are drug targets for oncology and therapeutic protein for autoimmune diseases and transplant rejection was also compelling, because we knew that we will have a much higher impact with these molecules.

Okay. A follow-up question for Marty. Just as a follow-up to your response to the previous question, and understanding the uniqueness of your model.

But is there a level of materiality, either in a positive decision by a collaborator to go forward or a negative one to walk away, that prompts you to announce that?

Well, I mean if it is a -- I think we have to look at each one individually. If there -- let's say first if they were going ahead and there was some type of a significant payment that was going to be made, I think we would have to or would want to or whatever -- that would be announced.

I think if -- clearly if all of our clients all disappeared I think we owe it to our shareholders to let them know that there is something wrong out there.

On individual agreements or -- and also with respect to some of these agreements it is very hard -- when you say is it moving forward or not? I mean the question in -- because we have a few situations where the initial molecule that sort of started the whole thing off is no longer being considered. However, that has led to other things that are being considered.

So is that -- what is that? Is that a success? A failure? What is it?

But as I said that is part of our strategy of building these hopefully long-term relationships with the industry, and starting them with one of our platforms, and hopefully over time having them get more of an understanding of the complete capabilities of the Company.

Okay, thanks.

Geoff Gilbert, Peak Investment.

Congratulations on the recent successes. We appreciate the walk-through of 15001. Much appreciated.

Question about the discovery process. Is it right to assume that additional in silica candidates targeting the B7 family were discovered but not prioritized for further evaluation?

Let me answer, since I am not a scientist. If you understand, as you evidently do from your question, the nature of our discovery efforts, I think it is reasonable to assume that that may be the case.

Correct.

In every situation, not just in this current one. Because keep in mind what we -- it's almost a mistake for us to use the word discover. We use it only because that is what people are used to in our industry. You discover things.

But we really discover nothing. Or -- from the standpoint that I say that Intel doesn't discover chips; IBM doesn't discover computers. All right?

Now we -- our whole process, if you remember what Anat went through, was you could summarize it all in prediction, prediction, prediction, selection, validation. That is what we do.

There is no discovery in there. All right? The end of the day is we have something new that people haven't seen before, so I guess you can say -- I guess you can use the term discovery.

But since we are doing prediction and we predict essentially as she described -- the initial prediction step in one sense is a prediction of every possible molecule based on our understanding of how life works, how are genes expressed. Transcripts. How transcripts become proteins; how proteins become peptides.

Based on our predictive models of those processes of life at the molecular level, our first step is to predict every possible molecule that we believe could exist that could potentially be a candidate for what you are looking for. Then it is sort of a whittling down in different ways as mentioned.

First it was, let's look at all proteins. Okay. Now let's look at the subset of membrane proteins. Okay. Now let's go and look at the membrane -- let's look at all cancer targets. Now let's put those two things together and see where cancer targets and membrane proteins overlap.

So it's, in that process at the end of the day, highly unlikely that you are going to have one candidate. You are going to end up with a series of candidates with -- and then we have to use -- and then the scientists and the experts have to decide how to prioritize, and which one, on what basis.

And sometimes there is good evidence as to which one might be the best candidate. Other times, to be honest, it is really a shot in the dark when you have a series of candidates. You know that they are all good candidates; but you don't know which is the best one to take forward at that stage.

So sorry, I am not going to -- I won't -- I can't specifically answer your question. But I can say that our discovery process, in general it would be very unlikely, almost impossible, for us at the end of a discovery run to say -- this is the candidate. It will be -- here are a series of candidates that need to be further evaluated.

Fair enough. Maybe as a follow-up could you give us an update or comment on how you are directing or prioritizing or focusing present efforts, given the pretty large amount of possibilities on your horizon these days as opposed to maybe two years ago?

Someone commented that maybe it's Compugen's perspective of the market demand; or is it also client driven; or a combination of both? Just an update.

It's -- I think we are reaching a point now where -- let me say that initially essentially everything here was science driven. To be honest, we really -- in some cases we almost purposely stayed away from the idea of -- what is going to be the product? The idea, and this was during the early years here, it was much more -- let's understand how life works at this level in a predictive way.

And given that understanding, we believe that we will have product opportunities in the future. I think that then, that led to, like in 2006, 2007, led to saying -- okay, now the time has time for us to take all of these, this science, and see if we can package it into ways where we can make some money and do some good for the world of medicine.

That is what led to our discovery platforms. It's interesting to note that the majority of our initial discovery platforms were in diagnostics, biomarkers, because that is kind of where the science sort of led us. From the standpoint of the way we do our research, those were kind of the easiest things to do.

As we interacted more with the marketplace, we realized that for a small Company we are going to need to focus much more. And the real interest seems to be on the therapeutic side. So that -- if you follow our development you will see that then most of our platforms then became therapeutic platforms.

Now as -- when we visit with clients and potential clients and discuss our platforms, out of that discussion very often will come a concept, an idea. And also our people here have become quite knowledgeable in certain fields, specifically -- and, Anat, maybe from what I can tell, I would say in oncology and areas of inflammation, immune.

Right.

If you want, in other areas, you talk about CNS or GI or whatever, we don't have much information. Now we can discover stuff in that area, but that would pretty much have to be client driven, not based on our expertise.

But in areas now like oncology and immunology, other areas of inflammation, we have developed an internal capability where we have a good sense of what the world needs. I think CGEN, this current discovery or --

Announcement.

I used the word discover -- yes, it shows that we have got -- we know what we're doing now.

Sounds good. Thank you.

There are no further questions at this time. Before I ask Mr. Gerstel to go ahead with his closing statement, I would like to remind participants that a replay of this call is scheduled to begin in two hours for a period of 72 hours. In the US, please call 1-888-782-4291. In Israel, please call 03-925-5929. Internationally please call 9723-925-5929.

Mr. Gerstel, would you like to make a concluding statement?

Yes, just again to thank all of you for your continuing interest in our Company. I see a number of our long-term shareholders on; I hope you are as pleased as we are, as we watch the pieces beginning to clearly come together.

Just at this point I can tell you without exception all of us in the Company are extremely optimistic and enthusiastic about the future at this point in time. So thank you for hanging in there with us.

Thank you. This concludes the Compugen Ltd. second-quarter 2010 financial results conference call. Thank you for your participation. You may go ahead and disconnect.