| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Retired Forums
Forum: Member-to-Member Support [Read Only]
Thread: How many protein molecules are contained in each work unit? |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 5
|
|
| Author |
|
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges: 
|
The general consensus seems to be that there's just 1 protein molecule per work unit and that is why the work units vary so greatly in length as each protein is unique.
----------------------------------------I can't find any information to dispute this but..... I've been crunching Cancer work units for first ud.com and then grid.com since 2001 and each work unit used to consist of a number of so called "Ligands" The number of Ligands in a work unit was changed each time a new target protein arrived so that the average time to run a Work Unit would always be around 24 hours on a 1Ghz machine. Some work units contained only 20 Ligands, other work units 450 ligands, but the time to complete each work unit remained about the same. I suspect that this is actually possible for our HPF tasks as Admin claims in the Community Maintained FAQ's "We are tuning work units so that they take an average of about a week of wall clock time to complete" So they can be tuned..... See my posting in suggestions/feedback in the "Smaller Work Units" thread. Personally I think one week is too long for the reasons detailed there. Could WCG confirm the number of proteins contained in each work unit we are running at the moment? Keep Crunching Dave p.s. I've edited this post numerous times because my spelling was terrible  [Edit 5 times, last edit by David Autumns at Nov 25, 2004 11:17:54 PM] |
|||
|
||||
|
True54Blue
Advanced Cruncher Joined: Nov 17, 2004 Post Count: 97 Status: Offline Project Badges: 
|
An estimate of the total number of results would be nice. According to the Protein Data Bank there are 28278 structures. (http://www.rcsb.org/pdb/) I was under the impression that we downloaded a protein structure, folded the stuffing out of it and then returned it. We are now at 70,000 results so obviously this can't be the case.
----------------------------------------If we aren't examining a particular protein what are we doing? Part of one? What's up with the lovely protein picture?   |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
In reply to the two above posts...
----------------------------------------All contain valid speculation however i have a few comments. Firstly pdb.org or rcsb.or/pdb - one in the same - contain proteins structure that have been verified mainly by X-ray crystallography and other means. However they represent a minimal number of proteins as a majority of proteins only hold their functional shape in water - and cannot be crystallised for analysis. The total number of proteins at this site is also not representative of what we are analysing. The site contains crystal structures for proteins from all different species - most of them bacterial (not human). However, known human-derived protein structures may be compared to results obtained from this study to verify folding patterns. One protein per work unit i believe - Alther has suggested the packaging of multiple units or downloading multiple units at one time which i think is a great idea. The word proteome means the proteins expressed in a cell at any one time... it might be the case we are not analysing all possible human derived proteins in this project unless they are directly derived from the human genome itself (ncbi.nih.gov) - this would have to include alternative splicing - ie: more than one protein variant from the same gene. Could Alther or Admin clear up where the initial protein data has come from??? Seeing the results are to be public domain I dont see why the public should not be informed... (unless ive missed it in the faq). Thanks to all who read my ramblings... To those who do not - I dont blame you...  [Edit 1 times, last edit by Former Member at Nov 26, 2004 10:49:11 AM] |
|||
|
||||
|
USA
Advanced Cruncher Joined: Nov 22, 2004 Post Count: 107 Status: Offline |
What They Said |
|||
|
||||
|
David Autumns
Ace Cruncher UK Joined: Nov 16, 2004 Post Count: 11062 Status: Offline Project Badges:
|
Apologies but this post was languishing on page 4 and still all the attention this question has received is a deafening silence from WCG.
----------------------------------------This is a post to get this post back in the number 1 slot so that it remains in full public view so maybe one day we might get a response to this question. Sorry for abusing the operation of the forum. I won't do this again if no-one answers the question I promise to let this post disappear off into oblivion Please forgive me - don't flame me I have the best interests of the WCG project at heart. Regards Dave  [Edit 1 times, last edit by David Autumns at Dec 2, 2004 3:51:34 PM] |
||
|
|
|