Page 3 of 8

Re: Protein Modeling C [NY Trial]

Posted: March 16th, 2010, 1:13 pm
by sciolykid101
Did you guys use a 10 ft minitoober or some other structure to base your model on (for the pre-build portion)?

Re: Protein Modeling C [NY Trial]

Posted: March 16th, 2010, 5:38 pm
by Phenylethylamine
sciolykid101 wrote:Did you guys use a 10 ft minitoober or some other structure to base your model on (for the pre-build portion)?
I was given a mini-toober of the exact right length (320 cm, somewhat more than ten feet) by my coach, who received it from MSOE (the organization sponsoring the event). I don't know what he had to do to get them to send it to him, but he didn't buy it- part of MSOE's sponsorship is that they spot everyone the mini-toobers.

Re: Protein Modeling C [NY Trial]

Posted: March 19th, 2010, 11:20 pm
by kwijiborjt
I ran this event at the RCS invitational... i think it's a great event! It's a huge, but i think it's very much in the creative spirit of science olympiad. How have your experiences been?

Re: Protein Modeling C [NY Trial]

Posted: March 20th, 2010, 12:56 pm
by johngrant
Is there a site showing how to use JMOL, for example, how to select and view a particular chain? This does not appear to work with the GUI within the JMOL tool.
A presenter showing the tool said "restrict" when she was keying in data on a scioly JMOL tutorial, but there wasn't any depth to this. Outlining the syntax and basic command structure would be helpful.An old manual guidance led me to believe Restrict[*b] would work but the JMOL app rejected this command.
Thanks!

Re: Protein Modeling C [NY Trial]

Posted: March 21st, 2010, 9:23 am
by Phenylethylamine
johngrant wrote:Is there a site showing how to use JMOL, for example, how to select and view a particular chain? This does not appear to work with the GUI within the JMOL tool.
A presenter showing the tool said "restrict" when she was keying in data on a scioly JMOL tutorial, but there wasn't any depth to this. Outlining the syntax and basic command structure would be helpful.An old manual guidance led me to believe Restrict[*b] would work but the JMOL app rejected this command.
Thanks!
If you go to http://cbm.msoe.edu/stupro/so/index.html and open the "2010 Pre-build Competition Environment" in the sidebar on the left, right above the screen showing the protein should be a link to a "Jmol Quick Reference Sheet", or something to that effect. It's a PDF that outlines the basic commands used in Jmol. That might be what you're looking for.

The "restrict" command is in fact what you want if you're trying to select and view only a particular subset of the protein, but the brackets are unnecessary. Just "restrict*b" or "restrict *b" (the space has no effect) will show only chain B.
The asterisk refers to a chain. To restrict your view even further, you would use, for example, "restrict*b and 1-160", where "1-160" refers to the range of numbers of amino acid residues within chain B you want to show.
If you wanted to show multiple chains or multiple ranges of residues, you would use "or", as in "restrict*b or *d".
If you're using an "and" and an "or" in the same command, you might need parentheses (not brackets) to specify which goes with what; e.g., "restrict(*b or *d) and 1-160" would show you residues 1-160 of chain B and residues 1-160 of chain D, while "restrict*b or (*d and 1-160)" would show you all of chain B and residues 1-160 of chain D.

Another tip: after restricting your view to any subset of the protein, I recommend using the "center" command to avoid wrestling with the small peripheral part of the PDB file you're viewing.
The default is for the center of rotation of the file to be the center of mass of the whole thing, but if you're only looking at a part of it on one edge, you don't really want it rotating around that point, because then in order to see the other side of your little part, you have to swing it all the way around that center. Some angles of viewing aren't even possible that way, because your part would be off-screen when you've got it sufficiently zoomed in to see the details.
Therefore, it can be very useful to add "center*b" if you're viewing chain B, or "center*b and 1-160" if you just restricted it to chain B and 1-160 (i.e., whatever you just put in the "restrict" command, use the same thing in the "center" command). This will save you a lot of annoyance.

Re: Protein Modeling C [NY Trial]

Posted: March 21st, 2010, 2:53 pm
by organic chemist
Along with sidechains, what are other important things to include on the protein? Are properties of the amino acids, like hydrophillic and hydrophobic, good to include too?

Re: Protein Modeling C [NY Trial]

Posted: March 22nd, 2010, 6:02 am
by Phenylethylamine
organic chemist wrote:Along with sidechains, what are other important things to include on the protein? Are properties of the amino acids, like hydrophillic and hydrophobic, good to include too?
The point of the model is to explain the "molecular story" of the protein. In other words, you want to show its function. Any aspect of the protein that elucidates its function would be good to include.

Re: Protein Modeling C [NY Trial]

Posted: March 22nd, 2010, 5:37 pm
by LinusPauling
Remember to balance creative additions with accuracy - too many pre-builds emphasized creative additions and lacked appropriate secondary/tertiary protein structure.

Re: Protein Modeling C [NY Trial]

Posted: March 22nd, 2010, 5:39 pm
by LinusPauling
kwijiborjt wrote:I ran this event at the RCS invitational... i think it's a great event! It's a huge, but i think it's very much in the creative spirit of science olympiad. How have your experiences been?
My experiences have been great!

Re: Protein Modeling C [NY Trial]

Posted: March 23rd, 2010, 1:18 pm
by Phenylethylamine
LinusPauling wrote:Remember to balance creative additions with accuracy - too many pre-builds emphasized creative additions and lacked appropriate secondary/tertiary protein structure.
This is certainly true. Accurate secondary structure is essential, as well as roughly accurate tertiary structure. However, showing complete primary structure is unnecessary; there were several teams that showed every sidechain, and that really doesn't add anything to your model. Only show the important sidechains.