Happy New Year to all! 2007 should be a very exciting year at rosetta@home!

Continuing the theme of my last post, I just met with Barbara Wakimoto, who leads a group at the UW
which trains science teachers (see http://monera.biology.washington.edu/hhmi/), and I just submitted a proposal to HHMI to develop a curriculum unit arounds rosetta@home for middle school and high school and later with direct input and feedback from teachers. There are a lot of excellent materials already available on the internet, so what we will mainly need to do is to collect the most appropriate information and assemble it. If you have ideas, perhaps a thread could be started on these boards with useful links. Of course, once these materials are asssembled coherently, it should be an ideal way for newcomers to learn about the science underlying rosetta@home.

In another news, I just put the final touches on the paper describing the first round of results from the rosetta WCG project led by my former graduate student Richard Bonneau who is now a professor at NYU. This paper is going to be published in a very successful relatively new journal called Public Library of Science, which is based on the premise that all research results should be freely available to everybody, and hence there are no subscription fees; anybody can freely look at all the papers they publish. I'll put up the link to the paper as soon as it appears so you can see how this works.

---

I met today with Dr. Wayne C. Koff, the Senior Vice President for Vaccine Research at the
International AIDS Vaccine Initiative (IAVI). You can read about IAVI at http://www.iavi.org/.
We have recently joined the team of researchers they are supporting to collaboratively work towards a
vaccine, and we discussed many different strategies and the ways in which we can work most
productively with other IAVI vaccine developers.

On a different front, I gave the opening keynote talk at a chemical engineering meeting on creating new biological molecules and systems earlier this week in San Diego. Our computational approaches are very complementary to the experimental approaches developed by chemical engineers, and in discussions over the next two days I established a number of very exciting collaborations on creating new drugs and enzymes for many applications. You can read about the meeting at http://www.aiche.org/sbe/events/ICBE/program.aspx. (you will see that the keynote speaker the second night was Craig Venter, who led the private efforts to sequence the human genome--he gave a quite futuristic talk on creating brand new organisms).

---

We are just finalizing two manuscripts which describe some of the most exciting results obtained thus far by rosetta@home. We have identified the rosetta@home contributors who found the rather spectacular low energy structures reported in the first paper, and would like to acknowledge them in the paper. Our question is whether to acknowledge these contributions through usernames or real names, and we certainly don't want to use the latter without permission. Here are the user names of the four participants for the first paper, I'll post the participants we would like to acknowledge in the second paper when we have identified them.

78884 aotama
66651 DJ N-4ceR
77194 DOJ F@T Elvis Clan
49481 S-A-M (corrected fromoriginal post)

It would be great if the above participants could let us know how they would like to be acknowledged.

---

Yesterday I acknowledged the participants who found the low energy structures being reported in Rhiju's paper. Today I have a list of the participants who found the low energy structures for the proteins reported in Bin's paper (the list is longer because Bin's paper describes a number of different proteins, we are citing the four participants who found the lowest energy structures for each of them). Again, please let us know how you would like to be acknowledged.

33706 gibsonrr
22840 Jynxedu
93320 John Michael Hess
59655 Luke

66275 Marko
757 Ian_D
82740 pxee
42892 csbyrosetta

96970 Allan Staib
97036 TJSwan
80380 Yunomi
7028 Etienne Guyot

31134 GeorgeF
68333 Jarom Smith
9221 UBT - Menace
62362 Boštjan Rudolf

17199 [SETI.USA] D. Brown
90304 bs98909
86711 papagoof
76056 Ackiss

86743 Noel E. Shea
58129 Bjarke.Orbeck
2446 SerVal
55755 Erik

33464 [DPC]FOKschaap~devzero
44026 A's Team
13172 jeidler
9030 Rebel Alliance

---

Here is the list of team names where available for Bin's paper acknowledgements. Please contact your teammates--I have only heard back from a few of the people we would like to acknowledge. Interestingly, the people who found the lowest energy structures for Rhiju's paper are not members of any team.

33706 gibsonrr
22840 Jynxedu from TechIMO
93320 John Michael Hess
59655 Luke

66275 Marko from Serbia - The Wild Bunch
757 Ian_D from Team Downham (UK)
82740 pxee from Poland Null-Zero Team
42892 csbyrosetta from SETI.Germany

96970 Allan Staib
97036 TJSwan from Arizona State University
80380 Yunomi
7028 Etienne Guyot

31134 GeorgeF
68333 Jarom Smith
9221 UBT - Menace from UK BOINC Team
62362 Boštjan Rudolf

17199 [SETI.USA] D. Brown from SETI.USA
90304 bs98909
86711 papagoof
76056 Ackiss

86743 Noel E. Shea
58129 Bjarke.Orbeck from danish team
2446 SerVal from Russia Team
55755 Erik from Dutch Power Cows

33464 [DPC]FOKschaap~devzero from Dutch Power Cows
44026 A's Team from Team Art Bell
13172 jeidler from BOINC@Austria
9030 Rebel Alliance from TeAm AnandTech

---

I only ended up hearing from 3 contributors, who are acknowledged by their real names in the papers we just submitted to Science; the others are acknowledged through their user names. There is still a month or so during which we can substitute in real names if we hear from you.

In other developments, inspired by a discussion with my 12 year old son Benjamin, we have started to look hard at how to improve on the carbon sequestration activity of the plant enzyme rubisco. there are many exciting applications in this area, ranging from biofuels to attempting to counteract global warming by reducing C02 levels in the atmosphere. I hope to have experiments up and running on rubisco within a month or two, and to start generating and testing computational designs of improved rubiscos (which all of you will be able to work on) not too long after that. The HIV vaccine design project is running full speed, with many potential vaccines being tested as I type, so we have some time now to get up to speed on the next big challenge!

---

One of the key advances that underlies our HIV vaccine design efforts has recently been published in the scientific journal Nature by the research group of Dr. Peter Kwong, with whom we are collaborating closely. Dr. Kwong's group has determined how an antibody can neutralize the virus by binding to a critical region on its surface required for binding to and entering our cells. We have been using the newly published structure, which Dr. Kwong made available to us many months ago, to design small protein mimics of the virus that we hope will elicit similar antibodies when people are vaccinated with them. You can read about Dr. Kwong's discoveries by picking up Nature at a newstand (his work was featured on the cover!) or from the press releases; I'm pasting the first part of one of these below:

Scientists Unveil Piece of HIV Protein that May Be Key to AIDS Vaccine Development
In a finding that could have profound implications for AIDS vaccine design, researchers led by a team at the National Institute of Allergy and Infectious Diseases (NIAID), part of the National Institutes of Health (NIH), have generated an atomic-level picture of a key portion of an HIV surface protein as it looks when bound to an infection-fighting antibody. Unlike much of the constantly mutating virus, this protein component is stable and — more importantly, say the researchers — appears vulnerable to attack from this specific antibody, known as b12, that can broadly neutralize HIV.

“Creating an HIV vaccine is one of the great scientific challenges of our time,” says NIH Director Elias A. Zerhouni, M.D. “NIH researchers and their colleagues have revealed a gap in HIV’s armor and have thereby opened a new avenue to meeting that challenge.”

The research team was led by Peter Kwong, Ph.D., of NIAID’s Vaccine Research Center (VRC). His collaborators included other scientists from NIAID and the National Cancer Institute, NIH, as well as investigators from the Dana-Farber Cancer Institute, Boston, and The Scripps Research Institute in La Jolla, CA. Their paper appears in the February 15 issue of Nature and is now available online.

“This elegant work by Dr. Kwong and his colleagues provides us with a long-sought picture of the precise interaction between the HIV gp120 surface protein and this neutralizing antibody,” says NIAID Director Anthony S. Fauci, M.D. “This finding could help in the development of an HIV vaccine capable of eliciting a robust antibody response.”

---

We have spent the last several days devising simplified biochemical cycles that would convert carbon dioxide into simple sugars using enzymes we would computationally engineer with your help on rosetta@home. Graduate student Justin Siegal and postdoc Eric Althoff have come up with a very clever new reaction cycle using new enzymes we would collectively engineer that in total carries out the following reaction:

2C02 + 2e- + H20 -> C2O3H2 + O2

the product is a simple sugar that could be used in a variety of ways, and the removal of C02 from the atmosphere would be great for countering global warming. A nice thing about this compared to current ideas of forming inorganic carbonate compounds is that it requires no other inputs. However, it does require electrons, and hence a source of energy. We are currently assessing the energy requirements of this process and comparing them to those of other proposed carbon sequestration mechanisms.

---

Our HIV vaccine design efforts are now going full speed ahead after months of building up after our Gates foundation award began in August. There are now seven full time exceptionally talented scientists working on the project, and we are exploring a broad range of strategies. Because computational protein design has never been applied to vaccine design before, our efforts have been
very enthusiastically recieved by the community of scientists working on HIV vaccine design problem, and we are now collaborating with many of the most active research groups in this area and recently joined the Neutralizing Antibody Consortium (http://www.iavireport.org/Issues/0602/Neuts.asp). We are very much hoping our new computational methodology can contribute to a breakthrough in this critical area!

---

I would like to apologize for the recent problems many participants had which were due to the filling of the rosetta@home queue with the memory intensive "HINGE" workunits. With these work units we were building models for the current CAPRI protein-protein structure prediction challenge http://capri.ebi.ac.uk/, for which predictions must be submitted by Sunday, and so the jobs were targeted as high priority. The problem was that while we specified a minimum of 512MB of memory for these jobs, this was only sufficient for many of your machines if only a single CPU was being used. Once again, we apologize for the problems which the combination of high memory and high priority caused, and will definitely avoid this in the future.
[I'm posting this here for people who didn't see the thread where the problem was discussed]

---

For those who didn't see it, I'm reprinting below Rhiju's post describing the science behind the new RNA work units. Also--I hope you are enjoying the new graphics! It is now much easier to see the sidechain packing problem that your calculations are trying to solve. Watch carefully and you will be ready to win at the interactive game version in which you will be able to guide the protein into a better packed lower energy structure. A super computer scientist, Adrien Treuille, is now working on this and we hope to have a prototype out for you to play with in several months.

Here is Rhiju's post:

Rosetta@home has been used to predict protein structure and design new
proteins. Improving protein enzymes that process carbon dioxide,
designing a protein vaccine for HIV, and understanding the basic
physical principles underlying protein folding are all being carried
out in the workunits that you are currently running or will be running
in the near future. Why study this other biopolymer, RNA?

Indeed, for a long time, RNA was mainly thought of as a passive
messenger in the cell, sort of a temporary copy of DNA (the storehouse
of information in our cells) from which proteins were translated. That
view changed dramatically in the 1980s with the discovery that RNA can
do more than carry information -- it can fold up into intricate
stuctures and carry out chemical reactions, just like proteins can. In
fact, its ability carry out the function of both DNA and proteins
provide a convenient answer to the chicken-and-egg problem of how
proteins and DNA evolved ... RNA came first! For more on the RNA
world, see
Wikipedia,
for example.

In addition to being important in early life, the known roles of RNA
in our present cells are currently exploding. From small RNAs
("Breakthrough
of the Year" in Science in 2002, and the subject of last year's
Nobel
prize winning work) to the catalytic heart of the ribosome, RNAs
play fundamental roles in biology that go beyond merely passing
messages. From a medical standpoint, most antibiotics target RNA in
the ribosome; the genomes of HIV, SARS, and other retroviruses are
composed of RNA; and RNA "aptamers" are beginning to be used as easily
evolvable drugs.

With Rosetta, we are making a serious stab at solving one of the most
fundamental puzzles for this interesting molecule... how RNA
folds. We're starting small, though some of the molecules you'll see
on our screen are important pieces of HIV. The hope is that we'll be
able to soon predict the folds of bigger RNAs -- a challenge that is
all the more important because RNA structures are difficult to solve
experimentaly. In the further future, using these structure prediction
tools to design or enhance ribozymes (that is, RNA enzymes) for
medical therapy is not that far-fetched ... stay tuned!

---

Lots of exciting things happening now!

The entire HIV vaccine design team (close to ten people now!) went to a meeting near Vancouver last week and were delighted at the enthusiasm their application of computational methods to vaccine design is generating in the field. Now, in collaboration with the many research groups we are working with, we just have to come through with vaccine candidates that elicit a neutralizaing response to HIV!

Paul Murphy, a graduate student in the group, is working on a project I haven't described previously which may interest many of you. As you might know, many human diseases caused by mutations in specific genes can potentially be cured by introducing stem cells that have the mutation fixed. The problem has been that in some cases the stem cells can divide too much which can lead to tumors. So what is necessary is a way to kill the introduced cells if things go wrong with the therapy. Paul's approach is to take advantage of an anti fungal drug -- 5 fluoro cytosine -- which is not itself toxic
but is converted by an enzyme present in fungi, but not in human cells, to a toxic compound. Paul is working to redesign a human enzyme to catalyze this reaction, if he succeeds, this enzyme could be included in the introduced cells so that they can be specifically killed by the drug if necessary. (simply putting the fungal enzyme into the cells is not good because it will be recognized as foreign by the immune system).

Finally, welcome to the ESL team which has given us a much needed surge in computing power this past week!

---

As I have described previously, we are working toward developing agents for gene
therapy by redesigning homing endonucleases (DNA cleavage enzymes) to
cut within genes containing mutations responsible for various
diseases. It has been shown that cleavage near a disease-causing
mutation will induce cellular recombination pathways leading to
correction of that mutation. Our focus is on diseases amenable to
gene therapy, where the healthy cells have an advantage over the
diseased cells, and correction of the genomes of even a fraction of
the cells can cure the disease. The diseases on our list include (but
not limited too), fanconi anemia, cystic fibrosis, haemophila A,
XSCID, ADASCID, and tyrosinemia I. Recently, we have produced
successful endonuclease designs toward partial target sequences near
mutations in genes responsible for fanconi anemia, haemophilia A, and
tyrosinemia I, bringing us steps closer to our goal of developing proteins
that can help cure these diseases.

We are all excited about the increase in computational throughput in the last month which
is letting us test many more ideas more quicly. Thanks to all of you!

---

Chu Wang is finalizing his manuscript on protein-protien docking which many of you contributed to. He would like to specifically acknowledge in the paper for finding low energy structures:

2357 Silver Streak 117 SETI.USA
109346 Libor B 46 Czech National Team
22119 zoaken
37078 highflyr
16605 Blackfly 19 AMD Users


if any of you would like to be acknowledged by your real name rather (or in addition to) your user name please let us know.

---

Rhiju is just finalizing a paper to be submitted to the proceedings of the national academy of sciences describing the RNA structure prediction calculations you have all done over the past several months. He would like to acknowledge the following users for finding very low energy structures. Ethan Owens has offered to contact the users to be acknowledged by email to see if they would like to be refered to by their real names, so if you are on this list expect an email soon!

RNA
User name
Team id (where relevant)

157d
hsugano

1a4d
Winkle

1csl
Drengy
Team MacNN

1dqf
Mark_F_Williams
TeAm AnandTech

1esy
Jaco J. Otto

1i9x
vatavale
TSC! Russia

1j6s
Chad

1kd5
Stevea
PC Perspective Killer Frogs

1kka
L3vi

1l2x
wcy5002

1mhk
Lennart
Electronic Sports League (ESL)

1q9a
TCU Computer Science
TCU

1qwa
Fuzion
Intuitive Legion

1xjr
ToniVR
BOINC.BE

1zih
von Greg
BOINC@Poland

255d
Nethack
Electronic Sports League (ESL)

283d
Mika Immonen
AncientGods Distributed Computing

28sp
uNion

2a43
PhilPlacier

2f88
Italian in Los Angeles
Los Angeles Prostate Cancer - Rosetta

---

Welcome to all of our new participants! The recent increase in computing power is very exciting and we are making good use of it!

We had a long meeting today with the three other research groups in Seattle we are collaborating with to create DNA cutting enzymes to correct mutations which cause diseases as I described a week or two ago. Our collaborators have tested a DNA cutting enzyme we have designed that targets mutations responsible for severe combined immunodeficiency disorder (SCID) in human cells and the preliminary results are that it targets the SCID site just as we designed it to! the next steps our collaborators will take with this and other disease targeting enzymes we are designing is to deliver them along with the normal (non disease) version of the gene into mice with the disease and see if the mutations are corrected and the disease cured. my students and I are impatient to see if we can cure human diseases with designed DNA cutting enzymes but our collaborators are emphasizing that a lot of testing has to come first.

I promised an explanaation of my "Search_Pairings" work units. as many of you know, many proteins contain beta strands which pair to form beta sheets. for many proteins, the key to predicting the structure correctly is getting the correct pairing between a critical pair of strands. the problem is that
in some cases the critical pairing forms extremely rarely. in the Search_Pairings work units I solve this problem by making a list of all possible pairings, and at the beginning of each of your calculationsn one of these pairings is randomly selected and forced to be present throughout the calculation of that structure. this procedure ensures that all pairings are sampled at a reasonable frequency so no critical pairings are missed. if you get one of these work units, see if you can pick out the pairing that is being kept fixed!

I am testing this approach out on proteins of known structure. In parallel, graduate student Rob Vernon is using the same approach to build models for the fibers formed by the alpa-synucein protein which are implicated in Parkinson's disease.

---

A warm welcome again to all new rosetta@home participants. I hope you find participating in the project interesting and fun! We greatly appreciate your contributions!

As I have described previously, we are actively working on designing HIV vaccines that mimic different portions of the viral coat protein. This is a big effort with seven graduate students and postdoctoral fellows here at the UW working full time with Rosetta to design vaccine candidates, and many collaborating groups in Seattle, the NIH, and elsewhere testing to see if the potential vaccines elicit a neutralizing response to the virus. These very large scale projects are funded by the Gates foundation and the International Aids Vaccine Initiative. I will of course keep all of you posted on advances on this important problem! I should also emphasize that the computational design techniques we are developing for HIV should be applicable to other viruses, and we are thinking about starting up a project on Influenza within the next year.

I haven't described previously a complementary approach in which we design proteins not to mimic the virus, but to bind tightly to and inactivate the virus. We just completed the computational design of the first such protein, and have just started up the experimental testing. Our next step in this project is to design inhibitors that bind to and inactivate other human pathogens, we are currently deciding on the next target to go after.

---

We are just finalizing a manuscript describing predictions of protein-protein interactions made using rosetta@home during the recent CAPRI international test of protein-protein docking methods. As with our previous papers, we would like to acknowledge participants who produced models which contributed to our final predictions. If you are on the following list, and would like to be acknowledged by your real name rather than your user name, please let us know.

USERID USERNAME SENDEMAIL TEAMID TEAMNAME
5590 SteveK 1 7 BOINC Synergy
112024 Administrator 1
125433 QuickBeam 1 4232 H4xx0rz
3377 devzero 1 5 Free-DC
64398 raptur 1 219 Ohio State University
36366 cyclistgb 1 887 BOINC@Canada
79579 borekv 0 1964 boinc.cz
78679 yopjpeg 1
111834 orion2598 1
147099 MK_I 1 46 Czech National Team

---

A lot of things have happened since I last posted!

For example, with your help, we now have a number of brand new designed enzymes that speed up the chemical reactions they were designed to catalyze by more than 50000 fold! This is a particularly exciting time because with the feedback from the experimental characterization of the designs computed with rosetta we are learning more and more about how to design new enzymes, and with this new knowledge we are constantly improving over our previous best designs. For example, for one of the reactions, the enzymes we had generated up until recently only worked once and then stopped working, but new designs we tested in the last couple of days just keep going. This is definitely a property we want to have in the enzymes we are hoping to design for carbon dioxide fixation and renewable fuel applications.

---

I am just putting the final touches on four manuscripts that feature the work that all of you have done (and acknowledge participants who made key contributions) that have been accepted for publication; we will put pdf's of these papers and the others which have involved contributions from rosetta@home participants on the home page soon.

Ian Davis will be training high school teachers on how to use the first version of the "Rosetta Game" in the classroom in two weeks. We have come up with a really simple version that should be easy and hopefully fun to play that focusses on packing sidechains like puzzle pieces in protein cores. I am hoping to make it available to all of you in not too long so you can see what we are up to (and play the game yourself!).

---