Posts by TestPilot
|
1)
Message boards :
Cafe Rosetta :
Moderator contact thread archive
(Message 25827)
Posted 1 Sep 2006 by TestPilot Post: Mod.Tymbrimi And because he was informed after posting, you claim he shouldn't have had that impression when he made his post? That guy should not insult others by posting personal attack, irrelavantely if he was informed or not and his "impressions". [an opinion stated as fact] Here come example of opinion stated as a fact: Mod.Tymbrimi If WCG using an older version of Rosetta requires 1000 results be tested to find a potential cure; while WCG using a newer version of Rosetta two years from now would require 20 results be tested to find the same cure. You came to thread where problem gone, brought old stuff up, said " my sincerely apology" to guy who did nothing but posted unmotivated attack. You knew you starting flame. Now you selectively deleted posts, and starting to talk about "opinions as facts" and someones "impressions". I'll try(the best I can) to post no futher on this subject cuz this is a flame discassion by nature. |
|
2)
Message boards :
Number crunching :
Bye bye Rosetta
(Message 25817)
Posted 1 Sep 2006 by TestPilot Post: TestPilot, did you crunch over at Find-a-Drug? Yes! I never posted on FAD forum, but I definitely remember you form just reading it!:) An open source science community are using some of the results, Malaria mainly (you can follow all the work, since their 'logs' are publiched to the web). Wow! That is nice to hear. I remember it was somewhat disappointing when Keiht said there is a little interest from industry in results. And it really cool that ppl working on them. And I do remember it was some lab testing made in Europe based on academic grants - during course of project or closer to the end. Plus, hopefully, there is way more activity/progress that is not made publically available. But they also have some of the HIV (if not all the 'anti' active ones).I think with all the commercial investments in HIV research open source almost have no chance to compete, but I'm wishing them luck! |
|
3)
Message boards :
Cafe Rosetta :
Moderator contact thread archive
(Message 25806)
Posted 1 Sep 2006 by TestPilot Post: Mod.Tymbrimi I'll go clean things up a bit. Oh - very impressive way to handle situation. Deleting Johns offence and my plus another moderator responce to it. But leaving in tact all the garbage directed against me posted by Sangamon Taylor and Phoenix. Like they were not explained that it was not my choice and not my intention to post under Administrator name. John seemed under the impression that you chose the name "Administrator" for ulterior motives to present your opinions of the other projects. But you know that this impression is plain wrong - it was explained to you, John, his team members and anyone else dozen times. And not only by me but by other participants and Mod.DE You posted your opinions about other projects as "facts". Oh, I state my opinion and always tried back it with what I know. If someone disagree - he can post his opinion and explain why he think so. That what forums are about. Those that have run those projects may see your opinions stated as "facts" as offensive. This sounds really confusing to me. You use that expression second time. What do you mean by saying opinions stated as "facts"? I'm expressing my opinion the best way I can, if someone disagree he always welcome to critique me and back his point. I never said something "whatever I wrote here is final facts and not to be discussed further". Or what do you mean? You don't have to insult someone directly when insulting the project they run. It is and never was my intention to insult anyone. Being sceptical about worthwhileness of some another projects is my right, and yes, I back my point whenever possible - not trying just "insult projects" for the sake of insulting. Read for example my last post there. |
|
4)
Message boards :
Number crunching :
Bye bye Rosetta
(Message 25801)
Posted 1 Sep 2006 by TestPilot Post: Ok. Let me clarify a bit my point as to FightAIDSathome. Here is the link to the most resent description of the project progress: http://fightaidsathome.scripps.edu/news/vol1.html What they are basically saying there, is that they compute interaction of 1900 drug like molecules against 270 variation of one of AIDS protein. So they, lets open Excel, are trying to predict outcome of reaction of total 513,000 interaction. Not that many – but they mention some scary sounding quadrillions. I’ll address those a bit later. There was another DC project Find-A-Drug. It actually over now – but lets take a look at it statistics. The project was running for ~3 years. Here is the link : http://www.find-a-drug.org but you have to navigate stat page yourself. 31,320,743,182 total interaction calculated (I was using total from targets page). Kinda different number. Number of proteins they were running against drug like molecules was ~280. Very similar. Number of drug like molecules checked against protein vary from 40,000,000 up to 500,000,000 with an average of ~100,000,000, a far cry from 2,000 of FightAIDS. Lets try to estimate efficiency of software used by projects. Ok, Find-A-Drug was running for 3 years, it had more members, and that press release from FightAIDS most likely referring to very recent developments of FightAIDS project. Plus WCG running other projects, not solely FightAIDS. So I think it would be fair to reduce total number of Find-A-Drug calculation by 10 for above mentioned reasons in sake of comparation. In this case we are getting that FightAIDS roughly 6,000 times slower when compared to Find-a-drug. Well within expected scope. That mean that for every hour your computer spent doing work for FAD the very same computer need to be ran for more then 8 months nonstop to calculate similar amount of interactions in FightAIDS project. If that is not impressive enough – think of in-house farm with 200 CPU that beat distributed project with 1,200,000 active CPU crunching nonstop. Ok, let me explain how come the difference is so huge. The answer is fundamental difference between software used by above mentioned projects. When AutoDock try to predict interaction between target protein and a ligand, it “wrap up” protein with net, where in every point of net crossing AutoDock performs energy evaluation. The bigger the protein – the bigger net you need. And this way more energy calculations. And HIV protease used by FightAIDS relatively big one. The tighter the net – the higher chances to catch strong reaction. And net should contain at least dozens of thousands net squares, for most proteins. On the other hand to run THINK (software behind Find-A-Drug project) you have to know the active site. It is place on the surface of protein which is responsible for function that is carried out by protein. This way THINK eliminating need for the net, thus dramatically reducing amount of computation needed to be performed. The obvious advantage of the AutoDock, is that you can feed it with just a 3dementional structure of a protein, without knowing anything else about protein you want to suppress. The disadvantages is huge. It is not only speed. A lot of molecules that will show strong interaction using computational simulation, and will confirm reaction in vitro, would have no effect or very weak effect in vivo – simply because even when they were bidding protein, they fail to actually block it function! Other advantage of having running protein against A LOT of compounds is that you can choose not only those ones that have strong predicted reaction, but select those ones that would have way higher chances to react ONLY with protein of interest. Which is very important in terms of limiting side effects. As to active site. AIDS is very well studied and relatively good understood virus. And for most, if not all of it active sites are known. You can check FAD web site – they were running interactions against 30 AIDS targets, what sounds for me as FAD covered pretty much all viral pathway, and in every case they knew active sites. So this should not be a problem at all. As to quadrillion number. Most of it comes from “net” calculations, performed by AutoDock. Plus it looks like they count all energies calculations against all atoms in HIV protease, some kind of “marketing” way of counting reactions for the sake of getting impressive number. Plus they included redundancy 5 time redundancy in that calculation:) FluffyChicken Anyone know what target's they are going for? (from the PDB). Most likely they use different submission of HIV protease in PDB database made by different groups on different strains of HIV. Plus some sort of possible mutations. But that is my guess. |
|
5)
Message boards :
Number crunching :
Bye bye Rosetta
(Message 25793)
Posted 1 Sep 2006 by TestPilot Post: Mod.Tymbrimi So, could you please explain what I did wrong? How could you say "Pass on my apology to John" referring to the guy who posted anmotivated attacking post with the whole point of bringing someone down? And it is clear that "Administrator" nickname were used by him just as a point of attack. I was not even talking to him priot to his post! You feel like it is Ok for someone to post stuff like that? Let me publish my reply under my name so ppl do not read the whole post wondering who posted it untill the very end of the post where is signature, sure with your edit Mod.Tymbrimi I noticed someone requesting Administrator use another account and several were helping him recover his normal name. Let me clarify this. I was the one who requested help to get my original credentials. The only reason why I did that is because I did wanted to change project settings on my home computer. No one at that point never ever requested me to change my name - and couple forum moderators replied to my posts having no problem with Administrator name. When that didn't help, I contacted the staff. Perhaps others did, too. It got changed. In fact the changes came after I wrote mail to Ethan (Forum moderator/Project administrator/Project developer). With his help I got credentials of TestPilot back. The reasons I switched my forum nickname was for my own convenience and because I do find nickname Administrator somewhat silly. Pass on my apology to John for not choosing the correct path sooner, as it may have kept this from developing as far as it did. The very first post that that John posted in this thread was personal assault against me. He tried to start flame by making personal attacks on the ground of my nickname, my status and mine imaginable behavior. I think such posts should not be tolerated on any forum. When the guy got pointed out on that - he decided to leave. If his major problem wasn't Administrator's name, then I'll mention that a local resident of where I live passed away. He lived extremely frugally and cut corners by not having running water. He and his wife would carry water to the house from a public water faucet. The guy passed away with more than a million in the bank. The next week, his wife had water service installed.. I leave this parable on your conscience. Your don't know me, to draw such a primitive and offensive analogies. The whole reason about this paragraph is to show your personal preferences. You think you found indirect and “witty” way of doing that, but reasons behind it remains pretty much straightforward. The rest of your post is set of sounding good banalities, in an effort to make your post looks like it was written in good faith. As to nickname Administrator. It is not in any way offensive nickname. It not violates any rules. As to point that John or someone else might assume that person behind that nickname have something to do with staff/moderators – is complete nonsense. Do not think ppl are stupid – they see forum status. And John’s post clearly illustrate it – no one so far assumed me being somehow involved with the project other than crunching for it. I did create that account on my buddy’s computer just to show BOINC and Rosetta screensaver to illustrate what DC is about. I never had intentions to make forum post under that nickname or use it for crunching on the first place. But it happens to be that that was only nickname I can use to log in to the system and there is no way I can change or create other record without loosing my TestPilot status in BOINC client on home computer. And, yes, I am an administrator in real life. I still fill like using TestPilot for other posts on this forum – but since I started to post in this tread under Administrator – I think nothing prevents me from keep doing that. On my side I was doing all I can not to go into pointless and unnecessarily flame, that John’s team members tried to start over and over in this tread. But you post make it sound like it is official point of view that I did something wrong and I’m should be blamed of John’s unmotivated anger about my nickname/status/etc. I did nothing I’m ashamed of, I did absolutely nothing wrong in this tread or in the project and I could not be blamed for anything bad. Furthermore I was trying to keep my posts as informative as I can, backing my points with facts and real arguments not switching to personalities, except for one time when I replied to John’s assault. But yes, there is one “fault” on my side. I am trying to inspire people to think about projects and critically evaluate them, not just mindlessly go for good slogans and good declarative goals. Sure critics draw some resistance and negative reactions. If I be here to gain popularity I would make point of my posts something like “You guys doing such a grate work! Keep up doing it!” But that was my choice and I do realize consequences. |
|
6)
Message boards :
Rosetta@home Science :
Solid answer needed, for who do we do this?
(Message 25269)
Posted 28 Aug 2006 by TestPilot Post: I'm pretty sure the application refining process is coming to an end. The method that would allow accurate and reliable determination of shape of protein based only on DNA sequence I'm afraid dozens year away from now. Hopefully there would be major breakthrough in the field, and that will happens faster - but chances are very low for that. So Rosetta most likely would be constantly improving for years to come. Don't get me wrong - I think this is very worthwhile cutting eage research, made by one of the best team of scientist in the protein structure prediction field. This research have a good chance to help develop better drugs /cure in the future. So I'm happily crunching - but please do not say that it free for everyone! |
|
7)
Message boards :
Number crunching :
Bye bye Rosetta
(Message 25266)
Posted 28 Aug 2006 by TestPilot Post: But be careful - most of other projects - like world community grid or fightAIDSathome is completely useless. Oh, for example WCG was running project that determine 3-d shapes of protein of humans and dozens of other genomes. They were doing that for couple years. Last time I checked - they did not implemented publicly available database with results of the project. What D. Baker says : "These efforts used the low resolution version of rosetta (which is all we had several years ago when the HPF project started);" And even if they create that database they were going to - do you think researchers would use it or they run newer version of Rosetta for proteins they interested in? What was thouse thouthands years of CPU time for? Do you own reserch for FightAIDS... |
|
8)
Message boards :
Rosetta@home Science :
Solid answer needed, for who do we do this?
(Message 25263)
Posted 28 Aug 2006 by TestPilot Post: The purpose of the project is to improve methods behind the Rosetta software. Rosetta is closed source package with very restrictive license. So Any findings Rosetta@Home makes are made freely available to everyone is a lie. As a claim: This is a not-for-profit project, so the results won't be sold. Yes, this is also not true - you can see from pdf, Rosetta is for sale - and for a price. I'm claiming nothing more the what stated above. The results of what we crunch are protein structures. Those are made available at no charge to the scientific community. The aim/purpose of the project not generating structures - but helping team to improve software that predict 3d-shape of proteins. |
|
9)
Message boards :
Rosetta@home Science :
Solid answer needed, for who do we do this?
(Message 25259)
Posted 28 Aug 2006 by TestPilot Post: Ahh, I understand what you mean now. Any findings Rosetta@Home makes are made freely available to everyone. That is a lie! Rosetta is a closed source project! No one can use software except he/she got licence, which strictly forbid redistribution. This is a not-for-profit project, so the results won't be sold. That is total bullshit! Won't be sold? We are helping improve Roseta package - and it officially on sale - read prices at the end of the file: http://depts.washington.edu/ventures/UW_Technology/Express_Licenses/Rosetta/Rosetta++EX.pdf $50 000 + $20 000 yearly fee! How ppl could lie so blatantly? |
|
10)
Message boards :
Number crunching :
Bye bye Rosetta
(Message 25218)
Posted 28 Aug 2006 by TestPilot Post: Keith Akins If you read the last two DB jounal entries, I think that you will see that the project is moving beyond just perfecting the program. Actual research appears to be comming. I think even if the project will work for years to come only on improving folding methods - that would be fine with me, it is by itself very worthwile task. Soriak The idea to keep an open ear for possible extraterrestial communication isn't all that bad. Even if the odds are incredibly low, it doesn't take much and the possible consequences are massive. I'm SETI fun. But my opinion - technology is not ready yet. And yes, government should spend billions on SETI. But not now. May be like in 150 years from now. 15 years ago there was no cell phones and no internet. And now I browse world wide web through PDA wirelessly connected to the internet by bluetooth/gprs. I do think technology will improve big time in next couple centuries, and current SETI attempts would look simply silly and naive... |
|
11)
Message boards :
Number crunching :
New Forum Software
(Message 25215)
Posted 28 Aug 2006 by TestPilot Post: This isn't a social community like MySpace. It's the support forum for a serious scientific project. Yes exactly - it is a forum. And very crappy one. Inconvenient and with almost no functionality. And why ppl forced to use this one? Just because project developers decided to save few hours and decided to go with built in installed by default option. There dozens of open source/GPL options for forum software, and all of them way better and way more convenient. The only reason things are the way they are is because admins simply do not care... |
|
12)
Message boards :
Number crunching :
New Forum Software
(Message 25185)
Posted 27 Aug 2006 by TestPilot Post: This forum is by far ugliest and most inconvinient one I use at the moment. phpBB, vBulleten, Simple Machines and 3-4 inhouse developed - all are better. And yes - phpBB is open source, and way more advanced, and there is no possible way open source project could close source. Someone think this forum serve the need of project? If there is convinient, neat and fast way, and inconvinient plus ugly one, wich would serve project needs better? Plus SIMAP switched phpBB already. And, phpBB is not only choise. |
|
13)
Message boards :
Number crunching :
Old/New Credit system comparisons
(Message 25126)
Posted 27 Aug 2006 by TestPilot Post: I am a Mac guy and if the Macs can not contribute then you don't get my PC either. I do think they use same source code for all platforms. And most likely nothing can be done with old G3/G4/G5 mashines. |
|
14)
Message boards :
Number crunching :
How to fake out the new credit system
(Message 25121)
Posted 27 Aug 2006 by TestPilot Post:
Yes, but to effectively profit from this you need: 1) Overclaim points only for some type of specially selected WU(which is relatively easy to implement) 2) Get thouse WU you are overclaiming on average more often than non cheating computer gets. The problem is 2). Hm... Yes, you can count only WU of that type you are overclaiming! Just return all other types with errors, and wait until server give you thouse WU you really want! So, solution to the topic question found, am I right? |
|
15)
Message boards :
Number crunching :
New Forum Software
(Message 25118)
Posted 27 Aug 2006 by TestPilot Post: I certainly think a lot of people would like to see improvements in the forum GUI. Yeah! I totally agree. I would definitely love to see that happened. |
|
16)
Message boards :
Number crunching :
How to fake out the new credit system
(Message 25113)
Posted 27 Aug 2006 by TestPilot Post: All PowerMACs with the old IBM processors get less than 50%. For example here: Yes! Exactly! So new system grant fair amount of points to computers, irrelevantly of claimed amount of points and benchmark results! Furthermore. At the top of the thread it was suggested that if big enough team over claim points - they will eventually get more points. That is true, but ironically that would not help cheating team. Simply because by over claiming they increasing amount of points not just for themselves, but also for there rivals/competitors! The only effects would be that the project as whole, Rosetta at home will claim slightly more boinc points compare to Einstein@Home / ClimatePrediction.Net... |
|
17)
Message boards :
Number crunching :
Please HELP! Read FAQ - no answer!
(Message 24946)
Posted 26 Aug 2006 by TestPilot Post: When you press the button on the Boinc client, it opens up IE with the TestPilot info. How about changing your default browser to FireFox and pressing on the button in the Boinc client to open up the TestPilot page in FF. Does it let you change preferences now? Make no difference - FF behave exactly same as IE. When you choose Login - it gives you the option of logging in with an account key. On the the left, it says, "if you don't know your account key.. click here." If you click there, it asks for your email, and will send your account key(s) there. If it sends you two, then you're in luck. If it only sends you Administrator's account key, then perhaps others have run across this situation before and can offer a solution that worked in the past. Yeah, it sends just "Administrator" account key. Furthermore, if I change email of "Administrator" - it do not send anything to "Administrator"'s previous email, it say No account found with email address ****@****.com. Looks like all my emails(I got 3 total I use) ether not in the system, or associated with Administrators account. |
|
18)
Message boards :
Number crunching :
Please HELP! Read FAQ - no answer!
(Message 24939)
Posted 26 Aug 2006 by TestPilot Post: I made 2 screenshots with screens I'm getting. One when I pressing Your Account button, and one where I'm pressing Your Preferences button in boinc client. ftp://70.30.188.177:2121/ip/YourAccountButton.JPG ftp://70.30.188.177:2121/ip/YourPreferencesButton.JPG On the second screenshot non secret email. I steel wanna keep my main one secret. Feet1st Here we go! I did not change anything, yet. <project> <master_url>http://boinc.bakerlab.org/rosetta/</master_url> <project_name>rosetta@home</project_name> <user_name>TestPilot</user_name> <team_name>FaDBeens</team_name> <email_hash>cb5537da664b925c2ba3691a9e5efd4a</email_hash> <cross_project_id>1385e343efd0c253a269aff431e9b9fd</cross_project_id> <user_total_credit>18311.070406</user_total_credit> <user_expavg_credit>23.131097</user_expavg_credit> <user_create_time>1127457095.000000</user_create_time> <rpc_seqno>17822</rpc_seqno> <hostid>1821</hostid> <host_total_credit>18311.070406</host_total_credit> <host_expavg_credit>23.131097</host_expavg_credit> <host_create_time>1127457199.000000</host_create_time> <nrpc_failures>0</nrpc_failures> <master_fetch_failures>2</master_fetch_failures> <min_rpc_time>1156559260.722859</min_rpc_time> <short_term_debt>-940.361052</short_term_debt> <long_term_debt>-49551.156896</long_term_debt> <resource_share>100.000000</resource_share> <duration_correction_factor>0.382841</duration_correction_factor> <scheduler_url>http://boinc.bakerlab.org/rosetta_cgi/cgi</scheduler_url> <code_sign_key> |
|
19)
Message boards :
Number crunching :
Please HELP! Read FAQ - no answer!
(Message 24937)
Posted 26 Aug 2006 by TestPilot Post: BennyRop When you log in to the TestPilot account, will it let you make any of the other changes allowed on the "your account" page? No I could not change anything. The only way I see TestPilot http://boinc.bakerlab.org/ web page is when I pressed "Your Account" button in boinc client. perhaps the cookie is causing a conflict with logging in as TestPilot. Try a different browser than normal.. or try logging in on computer that doesn't have Boinc installed and see if that helps. I tryed to use Opera /Firefox. Both do not automatically log me in - both times when I used credentials of TestPilot(the one that work for QMC/SIMAP) - i got logged into Administrator account. How about changing the email account on Administrator to a new hotmail account? And see if that helps. I did it - in fact I changed it to ****@lycos.com account i already had. Now when i'm using my real email to login rosetta says: No account found with email address ****@gmail.com . Gmail is my real email, using that email I still can log in to QMC/SIMAP using Opera, Internet Explorer keeps me there logged in there automatically. Feet1st Detach Rosetta. ...now attach to Rosetta again, and set it up with your "existing" TestPilot account's EMail address and password. If i do that I will end up having Administrator account! If u attach to the project it will ask you your email/password. And rosetta tend to log me in when I eneter email/password as Administrator. But in client I'm TestPilot in Rosetta... I do want to keep it that way. I'm sure the project team will be glad to see posts from "administrator" coming in as "TestPilot" too! :) Yeah I agree - kinda stupid nickname. Keith Akins If you go to your browser's tools or network options and delete cookies I'm very afraid cookies have nothing to do with this problem. I could not log in as TestPilot from Opera, and last time I used Opera was like couple years ago. |
|
20)
Message boards :
Number crunching :
Please HELP! Read FAQ - no answer!
(Message 24929)
Posted 26 Aug 2006 by TestPilot Post: Oh, I don't know how to login using Account name! If I use email+password associated with TestPilot account - I'm getting into Administrator account. And yes - I do know my email+password for TestPilot, it work fine for QMC and SIMAP projects, I could login there as TestPilot no problem!!! I pretty much sure that on 13 June I have created new account with different name("Administrator") for Rosetta with same email - that is what causing problems. But I do want to use testPilot, I don't want to be an Administrator in Rosetta@Home! |
Next 20
©2025 University of Washington
https://www.bakerlab.org