Hi all.
I run both Rosetta and Predictor AT HOMe and I happy so.
But I would like to understand well the differences between them.
In particular, I would like to find some results about gastric cancer and it's peculiar and well effective self defence system against anti-blastic therapy.
I've a particul interest in this desease because ot it killed my mother few months ago.
Any explanations or suggestions from the project's creator?

---

just noticed no one has replied to your thread yet so im bumping it and adding my 2 cents

Well Rosetta is not curently working directly to find a cure they are optimizing the processes as i can see it to make it so all the science programs that use rosetta run more optimaly and therby find the cure quicker.

I am never very good with the science detales and i know your answer has been posted on other messages on this board. I hope you get your answer and sorry for you loss

---

Hi Konstantino,

You can find out more about the difference between Predictor and Rosetta on this thread

As for your question regarding gastric cancer, let me try to explain what we are doing and how it relates in the big picture.

Rosetta@home is about developing computational methods that use only the knowledge of a gene sequence to determine the structure of the protein it encodes. This is important because the structure of a protein is essential for its proper functioning in the cell, and mutations that affect the strucuture of a protein can cause disease. In the past, drug development was often either a matter of chance, or due to screening large numbers of drugs to see if one would have an effect on a disease. In the future, we hope to use rational, knowledge based methods to deliberately develop drugs targeted at specific proteins involved in specific diseases. To do this, we need to know the structures of all the proteins involved in the proper functioning of our cells, and we need to know what structural changes occur when we get a disease, whether that disease is due to a virus, bacteria, or our own cells going haywire (as in cancer).

Now the real question is whether you believe this idea of structure-based drug development will ever really come to fruition, will it ever be a practical application and if so, will this happen in my lifetime. This is a matter of debate, but let me tell you that I personally believe that this will happen and it will start to happen within the next decade. Why am I optimistic? Partly it is because I work with a group of incredibly smart people, and the advances they have already made are really amazing. Secondly, there has been an explosion in the amount of biological information available in the last 15 years, and that information along with our approach will guide new efforts at drug discovery.

I don't want to mislead anyone into thinking that we are actually going to cure a disease in the next few months or so. In these days of modern molecular biology, there are always many years of incremental discovery that finally lead up to that eureka moment. But those eureka moments do happen. Think about it. 5 years ago, deriving human stem cells and the hope of therapeutic cloning seemed like a distant prospect; now stem cells are routine, and we're on the verge of breakthroughs in theraprutic cloning. 15 years ago, cloning a sheep seemed like science fiction - and then it actually happened. 30 years ago, even cloning a single gene trying to determine its function was a laborious effort; now this happens routinely. So 5 or 10 years from now, understanding the functions of all our proteins using computational biology and fixing any "broken" proteins involved in disease may not just be a pipe dream - there is a good chance it will be reality.

Ok, that was really long-winded but if you are still reading, then let me finish with this. I'm a post-doctoral researcher with David Baker, and some of my work has involved p53 and Mdm2, two proteins that are involved in many human cancers, including gastric cancer. I am trying to inhibit Mdm2 with the hopes of developing protein drugs that may, in the future, actually be used in disease treatment. So far, my results have been very modest, nothing that could actually be useful in a clinical setting. But they have been encouraging enough to make me want to continue to work on this, I do have hope that this research will lead somewhere productive. While my jobs don't run on Rosetta@home (yet ;-) the information that we gain from the work your CPUs are doing is still valuable to me, because prediction and design are 2 sides of the same coin, and whatever we learn from prediction helps us make better protein designs.

I hope that answers your questions somewhat, and if not I encourage you to keep asking questions until you are satisfied with the answers.

PS I have deliberately oversimplified things here, so don't take this as the final word on structure prediction or drug design - it's just an overview.

PPS I am sorry about your mother. I lost my grandfather to prostate cancer recently, and I'm also working on another project involving proteins that are misregulated in prostate cancer. It does help to feel like I'm contributing something to the fight against cancer, and I hope you feel that way too, because you are also contributing by getting involved with both our project and predictor.

---

Thank you,Vanita. Your insights are very interesting and enlightening. We learn something new on this project every day ! Your willingness and patience to share your knowledge with us is amazing and really appreciated. It will attract and maintain interest in the R@H project more then you know. I am also sorry to hear of both your losses as well....Rog.

---

This is what's great about this project, people actually take time to explain the science! Good luck in your research Vanita.

---