You are here

Re: Using a database of loop conformations together with de novo folding protocol

56 posts / 0 new
Last post

Dear Sir,

Thank you for your reply...I guess I am not able to explain my queries clearly, sorry for that ..

My sole objective is to find the direction of the twist(right/left) of the beta-strands. I have already calculated the direction of twist(whether right /left) for the turns. Now, I want to check whether the twist direction of beta-hairpin's strands match with that of it's beta-turn.

Now, as per the paper, it says that "The chirality of β-hairpins is determined by the length of the loop between the two strands ". What is the meaning of the word chirality here ?? Does it mean that the word "chirality" refers to right/left handedness of beta-hairpin strands alone. Also, what is the meaning of right/left handed beta strands then, I guess the answer to this should clarify my doubt (ref ??). Just by looking at the strands can we say whether they are right or left hand twisted ??

In the supplementary materials its mentioned that , "the chirality of a ββ-unit was considered using the vector ! along the axis of
the first strand, and the vector ! perpendicular to ! between the centers of the two strands. Since twisted strands lead to inaccurate assignments of the chirality, however, we used atom coordinates close to the loop between the strands for the definition of vec u and vec v . u is a vector from the N (backbone amide nitrogen) to C (backbone carbonyl carbon) atoms of the strand residue preceding the connecting loop and v is a vector from the Cα atom of the strand residue preceding the loop to the Cα atom of the strand residue following the loop", this is similar to the explanation that you provided me, when I asked for the calculation of handedness of beta-hairpins.

In order to confirm my calculation, I calculated the twist handedness of the beta-hairpins having two different types of turns - type I'(right-handed) and type II(left-handed)??. For both I got the handedness of hairpin (calculated using the above mentioned method)
as right-handed.

For type-I' eg I used pdb-id 1a1x (pdb residues 94-97) and for type-I eg I used pdb-id 1dqg (pdb-residues 20-23). They both give the angle positive, which means that they are the strands in both cases are right hand twisted.

Also, further check my calculation , I did the same calculation for on of the NMR models from the same paper. I used the pdb id 2LN3 (residues 64-67). In this case also, the handedness comes in right hand direction ?? Since the pdb contained NMR models , I therefore used the first model details for calculation, Here's the result for that calculation :-

In [10]: vecN = p.residue(64).xyz("N")

In [11]: vecC = p.residue(64).xyz("C")

In [12]: vecNC = vecN-vecC

In [13]: vecCA = p.residue(64).xyz("CA")

In [14]: vecCB = p.residue(64).xyz("CB")

In [15]: vecAB = vecCA-vecCB

In [16]: vecP = p.residue(64).xyz("CA")

In [17]: vecF = p.residue(67).xyz("CA")

In [18]: vecPF = vecP-vecF

In [19]: print vecAB.cross(vecNC).dot(vecPF)
19.063025436

.. Is the calculation correct or not ??. As the paper mentioned that all in all the simulations with the 2 residue turns the chirality was L ?? Pls find the attached figure containing all the beta-hairpins used in the calculation.

Hope I have made my points clear this time ... looking forward for your reply

Regards
-----------
BHARAT

Tue, 2013-05-28 18:21
bharat_46010

For Koga et al. when they refer to the chirality of the beta hairpins, they're talking about which direction the loop of the hairpin is turning with respect to the orientation of the residue immediately prior to the loop of the hairpin - or equivalently, which side of the N-terminal beta strand the C-terminal beta strand is on.

To my understanding, the difference between type I' and type II' beta hairpins is the orientation of the peptide bond (the orientation of the carbonyl) between the two residues of the two-residue turn (see http://swissmodel.expasy.org/course/text/chapter2.htm for a diagram). This sense of handedness is different from the handedness that Koga et al. talk about. For example, both of the hairpins in the linked figure have the same Koga-handedness, but one is type I' and one is type II'. -- This goes back to the point I was trying to make before. There are multiple senses of handedness at play here, and they don't all measure the same thing, so they're not going to agree for all structures. You need to carefully consider what it is you actually want to use the results of the chirality calculation for, in order to pick the correct calculation and get interpretable results.

Tue, 2013-05-28 19:32
rmoretti

Yes Sir, I understand the difference between the type I'&II' turns. I know this sense of handedness is different from the one described in the paper.

What I want to know that you just looked at the structure at told whether it's chirality is L/R ??

Sorry for repeatedly asking the same thing , but I am not able to understand that why my calculation's result doesn't match with that of the paper. What about my calculation for the first hairpin from the paper, is it correct ?? Also, I don't understand the meaning of this sentence "which side of the N-terminal beta strand the C-terminal beta strand is on" ??

You also said that "For example, both of the hairpins in the linked figure have the same Koga-handedness, but one is type I' and one is type II'. " It means that my calculation doesnot seem to be right. So, here's the calculation for the second hairpin :-

In [13]: vecN = p.residue(92).xyz("N")

In [14]: vecC = p.residue(92).xyz("C")

In [15]: vecNC = vecN-vecC

In [16]: vecCA = p.residue(92).xyz("CA")

In [17]: vecCB = p.residue(92).xyz("CB")

In [18]: vecAB = vecCA-vecCB

In [19]: vecP = p.residue(95).xyz("CA")

In [20]: vecP = p.residue(92).xyz("CA")

In [21]: vecF = p.residue(95).xyz("CA")

In [22]: vecPF = vecP-vecF

In [23]: print vecAB.cross(vecNC).dot(vecPF)
18.693263828

In response to your query "You need to carefully consider what it is you actually want to use the results of the chirality calculation for, in order to pick the correct calculation and get interpretable results." , I am rephrasing my question again . Can I use the chirality to tell/calculate whether a hairpin is right/left hand twisted ??.. Is there any other reference that you can provide (if possible) in this regard ??

Tue, 2013-05-28 23:48
bharat_46010

Regarding "which side of the N-terminal beta strand the C-terminal beta strand is on": Imagine lying down on the N-terminal beta strand of the hairpin, with your head at the c-terminal end, your feet toward the N-terminal end, and looking up at the Cbeta atom of the last residue. The loop of the hairpin comes up off of the top of your head, curves around and twists down to lie by your side. Which side is it? The left or the right? That's what the chirality from the Kogas' paper is measuring.

I'm at a bit of a loss to provide further help, as I'm not clear as to what you mean by "whether a hairpin is right/left hand twisted". Twisted how? In what sense? What's the geometry of the situation? In some sense a chirality is simply a rotation and a direction. What's being rotated with respect to what? - You say you're comparing chiralities, but what do you hope to achieve by this comparison? What insight do you intend to obtain? Perhaps by being clear on the big-picture goals, the smaller details will also become clearer.

Wed, 2013-05-29 11:31
rmoretti

Dear Sir,

I am really thankful to you for helping me clearing my doubts. After such a long discussion on this topic, I am now totally clear about what I need and what actually chirality means. Your last post cleared all my doubts. I once again thank you for lending your time in clearing my concepts.

Regards
-----------
BHARAT

Fri, 2013-05-31 01:26
bharat_46010

Pages