Messages

1426

General discussions (Q&As) / Re: script for extracting data from 3DNA output file

« on: January 04, 2011, 09:47:06 pm »

Welcome back. It is certainly clearer than before. However, it would be far more helpful if you could be even more specific, i.e., by providing an example. For example, I am not sure what the "100 snapshots (PDBs) from the simulation trajectories" look like. Are the 100 snapshots stored in 100 separate PDB files, or all in one? If the later, how are the snapshots separated? By MODEL/ENDMDL as in NMR structure? What would be an appropriate output format for the extracted parameters? In addition to the mean values of some parameters, e.g., Twist, how about their standard deviations and other related simple statistics? All such details need to be considered to come up with a script that is more generally applicable.

Thus, to help others help you more effectively, try to come up with a (minimum) concrete example, including all necessary input data files and your expected results (in numbers). Moreover, if you have already written some scripts, attach them with your post.

Alternatively, as mentioned in my blog post "Curves+ vs 3DNA", Curves+ has built in support for the analysis of MD simulation trajectories, and it may well serve your need.

HTH,

Xiang-Jun

1427

General discussions (Q&As) / Re: script for extracting data from 3DNA output file

« on: December 25, 2010, 10:44:26 pm »

Hi Aneesh,

Over the years, I have written a few posts related to the topic of applying 3DNA to the analysis of molecular dynamics (MD) simulations, including:

• 3DNA for the analysis of molecular dynamics simulations
• 3DNA in molecular dynamics simulations
• Curves+ vs 3DNA

Also, I contacted a couple of practitioners in the MD field, trying to seek a possible collaborator to make using 3DNA more straightforward for this increasing user community. For various reasons, nothing significant has come out from this effort. I am hoping users who have successfully applied 3DNA in MD analysis would contribute their scripts so others can benefit from and build upon. In the meantime, if you could post your MD analysis procedure and the problems you faced, others (myself included) may be able to help you more concretely.

Xiang-Jun

1428

General discussions (Q&As) / Re: window 7

« on: December 15, 2010, 09:09:23 pm »

Thanks for using 3DNA in Windows!

Which version of 3DNA are you referring to? If you are still using v1.5, it is definitely time to upgrade to v2.0. There are two 3DNA v2.0 binaries for Windows:

• Cygwin on Windows (see Cygwin homepage for details)
• Native Windows binaries with MinGW (see MinGW homepage for details; you may also want to have MSYS installed)

I am not a regular Windows user, but I did not meet much problem to compile 3DNA on a Windows XP machine in the two environments. I guess they should run on Windows 7, if you install either Cygwin or MinGW/MSYS -- you are the first to report such a problem of installing 3DNA on Windows. Please have a try and report back how it goes.

Any comments from other Windows users?

HTH,

Xiang-Jun

1429

General discussions (Q&As) / Re: Single strand DNA

« on: December 15, 2010, 08:52:10 pm »

I am not sure I understand the question. 3DNA does not perform any structure validation; it simply calculates some geometrical parameters given a structure (including single stranded DNA) in PDB format. As far as (NMR) structure validation goes, have a look of CING, or MolProbity.

HTH,

Xiang-Jun

1430

General discussions (Q&As) / Re: Missing Groove Measurement

« on: December 08, 2010, 09:07:26 pm »

Hi, Sean:

Hopefully the blog post you referred to, "3DNA for the analysis of molecular dynamics simulations", helps solve your problem. If any thing is still unclear, please provide a concrete example case.

Xiang-Jun

1431

General discussions (Q&As) / Re: O1P_O2P program

« on: November 15, 2010, 07:42:00 pm »

Hi Pascal,

Thanks for reading my recent blog post on O1P/O2P labeling, which was inspired by your above question!

given the recent PDB changes, is the O1P_O2P still interesting to use? Yet, after checking some of them, I noted that these changes occur mostly for terminal residues and I am not sure that everything is OK at this level (you could eventually like to check file 10MH).

The mislabeling issue should not have happened in PDB/NDB in the first place. The very fact that such error did happen, as shown for [mono:3dv13ht8]adh026[/mono:3dv13ht8], prompted the creation of the [mono:3dv13ht8]o1p_o2p[/mono:3dv13ht8] utility program. I am not surprised at all if you still find mislabeled O1P/O2P atoms in the currently "remediated" PDB files. I have quickly checked [mono:3dv13ht8]10MH[/mono:3dv13ht8] --  the two terminal phosphate groups (DC-402  on chain B and DG-422  on chain C) apparently still have O1P/O2P mislabeled.

On my side, I will continue to check the files, it might well be that some modified residues are also mislabeled in original files.

I think it is certainly a good idea to continue to check the files you care about. The command-line driven [mono:3dv13ht8]o1p_o2p[/mono:3dv13ht8] utility can be applied to any compliant PDB file, not just the one from PDB/NDB. Please report back if meet further issues, or have a request for added functionality :wink:.  Hopefully, I'd be able to "spare" more time to address 3DNA-related questions in the future   .

Xiang-Jun

1432

General discussions (Q&As) / Re: O1P_O2P program

« on: November 08, 2010, 10:24:15 pm »

Hi Pascal,

First, some background information: the utility program [mono:2uhtgbv8]o1p_o2p[/mono:2uhtgbv8] was written for a simple, specific purpose: long time ago, while calculating the RMSD value between the A-DNA NDB entry [mono:2uhtgbv8]adh026[/mono:2uhtgbv8] and the corresponding 3DNA rebuilt structure (with sugar-phosphate backbone), I noticed this RMSD was much larger than expected. Further inspection revealed that the issue was due to a mislabeling of O1P and O2P atoms for this specific NDB entry at that time (still available in directory [mono:2uhtgbv8]X3DNA/examples/analyze_rebuild[/mono:2uhtgbv8]). So [mono:2uhtgbv8]o1p_o2p[/mono:2uhtgbv8] was designed to check for proper O1P/O2P labeling, and to swap them if necessary, given a PDB file. Overall, the utility program works for its purpose, and has been released as part of 3DNA from the very beginning.

As a side note, current PDB/NDB entries have changed O1P/O2P to OP1/OP2. The mislabeling of O1P/O2P for [mono:2uhtgbv8]adh026[/mono:2uhtgbv8] has been corrected. Also, 3DNA v2.0 identifies OP1/OP2 labeling internally, but I still prefer to use O1P/O2P in 3DNA output.

[hr:2uhtgbv8][/hr:2uhtgbv8]
Now back to your question: I am glad that you've found [mono:2uhtgbv8]o1p_o2p[/mono:2uhtgbv8] useful. I know the problem you refer to, regarding header removal from [mono:2uhtgbv8]o1p_o2p[/mono:2uhtgbv8] output PDB file. However, current version of [mono:2uhtgbv8]o1p_o2p[/mono:2uhtgbv8] does not have an option to keep header as is. Conceivably, it should be feasible to add such functionality. I'll consider to put this point in my to-do list for future release of 3DNA (I am busy for my job until the following couple of weeks   ). In the meantime, you may prefer to write a script to extract the header from original PDB file, combined with "corrected" O1P/O2P coordinates.

HTH,

Xiang-Jun

1433

General discussions (Q&As) / Re: Re:stacking diagrm

« on: November 06, 2010, 08:03:17 pm »

Hi Raman,

Thanks for posting questions related to stacking diagrams. You are actually the first to ask such details, and the documentation is sketchy, thus no wonder you "couldn't locate it".

How we can generate different color diagram other than the default

Assuming you are using 3DNA v2.0, the config files are stored in the [mono:24t3qvd3]$X3DNA/config[/mono:24t3qvd3] directory, named [mono:24t3qvd3]ps_image.par[/mono:24t3qvd3], [mono:24t3qvd3]fig_image.par[/mono:24t3qvd3] and [mono:24t3qvd3]raster3d.par[/mono:24t3qvd3]  for PostScript, Xfig, and Raster3D formats, respectively. They are pure ASCII text files and should not be that difficult to follow. You may  copy a chosen-format-config file to a local directory for the stacking diagrams and tweak it as you see fit. Note the format is rigid -- make your change by directly overwriting the default one. [See FAQ http://3dna.rutgers.edu/x3dna/faqs#find_pair on local vs global effects for a config file.]

how one can hide hydrogen atoms in the generated stacking diagram.

You can run "[mono:24t3qvd3]get_part -d[/mono:24t3qvd3]" to remove H atoms first, and then feed the resulting PDB file to [mono:24t3qvd3]stack2img[/mono:24t3qvd3].

[hr:24t3qvd3][/hr:24t3qvd3]
Needless to say, the above-outlined procedures are not that user-friendly. However, they should work as expected, and allow for reproducible results with a script. Given an opportunity, I'd make the config file more flexible in future versions of 3DNA.

Have a try and report back if you have any problem.

Xiang-Jun

1434

General discussions (Q&As) / Re: Rebuild from sugar-phosphate backbone

« on: October 30, 2010, 08:56:41 pm »

As far as I understood, "rebuild" builds structure from base parameters. So you wrote in your blog "The structural parameters can be used to rebuild the structure, which is virtually identical in base geometry (i.e., without taking consideration of the sugar-phosphate backbone) to the original structure"

Yes, as shown in Table 3 of the 2003 3DNA NAR papers, and the corresponding examples in the directory [mono:3i9kwmym]$X3DNA/examples/analyze_rebuild/[/mono:3i9kwmym].

Tell me please, is it possible somehow to rebuild structure using only sequence and sugar-phosphate backbone parameters? I.e. without time-consuming minimization to define approximately Calladine parameters for each nucleotide pair of a given sequence, having only backbone?

No, as far as 3DNA is concerned. Also, what do you mean exactly by "sugar-phosphate backbone parameters"?
[hr:3i9kwmym][/hr:3i9kwmym]
Please note that the 3DNA [mono:3i9kwmym]analyze/rebuild[/mono:3i9kwmym] pair is purely geometry-based, without any "time-consuming minimization" involved. Building a DNA structure with or without the sugar-phosphate backbone should be (virtually) equally efficient.

HTH,

Xiang-Jun

1435

General discussions (Q&As) / Re: How to make a nice visualization for angle distribution?

« on: October 12, 2010, 11:30:56 pm »

Thanks for your interest in my GpU paper. I am glad that your find Fig 1C "a very nice representation for plotting the histogram of angle".

Per your request, here is the detailed step-by-step procedure I originally used to generate Fig 1C, up to the "raw" histogram of the phase angle of sugar ring pseudorotation.

• The data file named "key_pars_gu193.txt" is attached. It contains the phase angle of pseudorotation of the U and G sugar rings.
  
• The following R script is used to generate the "raw" figure (attached).
  Code: [Select]
  library("circular")

pdat <- read.delim("key_pars_gu193.txt")
attach(pdat)

xfig(file="sugar_raw.fig", onefile=TRUE)
par(mfrow = c(1,2))
plot(circular(p1, units="degrees", rotation="clock"), zero=pi/2,
     stack=TRUE, bins=150, shrink=1.5)
plot(circular(p2, units="degrees", rotation="clock"), zero=pi/2,
     stack=TRUE, bins=150, shrink=1.5)
par(mfrow = c(1,1))
dev.off()
Please note the followings:
  • The R 'circular' package is used (http://cran.r-project.org/web/packages/ ... index.html)
  • The xfig (http://www.xfig.org/) output device is used for later manually editing (not data points, but only cosmetic styles, of course) of the generated figure.
   

[attachment=1:1vq0n9no]key_pars_gu193.txt[/attachment:1vq0n9no]
[attachment=0:1vq0n9no]sugar_raw.png[/attachment:1vq0n9no]
HTH,

Xiang-Jun

1436

General discussions (Q&As) / Re: DNA Structure prediction with a heteroatom !

« on: October 10, 2010, 10:34:26 pm »

Thanks for using 3DNA and your kind words about it.

But, I'm interested to build the 3D structure of the DNA with a heteroatom which is a dye in my case. Can 3DNA be used to accomplish the task ? We have a NMR structure of the DNA with the dye but thats 20 mer. Is there a way to cut n paste the dye to the predicted 30 mer from the NMR using any tool ?

If I understand it correctly/clearly, I do not think 3DNA can be applied directly for your purpose. However, if you are very specific as to what you have tried, what problem you have experienced and what you'd like to achieve, others and I may help you more concretely. For one thing, I am interested in expanding 3DNA's capability for model building.

Other tools you may try include coot and NAB -- I have not used either of them, but I have heard of them mentioned quite a bit.

Xiang-Jun

1437

General discussions (Q&As) / Re: Global Helical Axis Information Missing

« on: September 24, 2010, 08:12:47 pm »

...so I tried changing the parameter from 0.6 to 2.0 but I am still unable to obtain the values for the helical axis. Does this mean that I should also change other parameters?

You've identified a bug in 3DNA v2.0  , so a big thank you! No other parameters need to be changed.

Technically, the bug is related to the tag [mono:hqpg41e9]"std_curved"[/mono:hqpg41e9] match; in the source code, it was mistaken as [mono:hqpg41e9]"std_curved "[/mono:hqpg41e9], i.e., with an extra space in the end. I've fixed this bug in the source; so the next release of 3DNA won't have this issue. As a temporary work around, simply add a suffix space in the tag, as below and it will work as expected:

Code: [Select]

<std_curved >0.6</std_curved >   # add a space in the end!As a side note, v1.5 does not have this bug.

HTH,

Xiang-Jun
[hr:hqpg41e9][/hr:hqpg41e9]
PS. As always, the more bugs reported, the merrier ! In future release of 3DNA, I will also add an acknowledgement list.

1438

General discussions (Q&As) / Re: Global Helical Axis Information Missing

« on: September 23, 2010, 09:57:20 pm »

That's because 1E3M is curved, thus fitting a global linear helical axis would be meaningless. More specifically, the section:

Code: [Select]

Global linear helical axis defined by equivalent C1' and RN9/YN1 atom pairs                    
Deviation from regular linear helix: 3.04(1.34)
shows a "deviation" of 1.34, which is larger than the default cutoff of 0.6, as defined in file 'misc_3dna.par'.

Code: [Select]

# Section 4: is this double helix curved?
#   criterion to decide if a helix is strongly curved.
#   if relatively straight, 'analyze' will output a set of
#   global parameters, the normalized axis vector, and the
#   two end points the helical axis passes through
<std_curved>0.6</std_curved>
HTH,

Xiang-Jun

1439

General discussions (Q&As) / Re: triplex building

« on: September 17, 2010, 10:34:42 pm »

Hi Alpay,

You may find it helpful to have a look of the recently published article "Designing Triple Helical Fragments: The Crystal Structure of the Undecamer d(TGGCCTTAAGG) Mimicking T.AT Base Triplets" by Van Hecke et al in Cryst. Growth Des. (DOI: 10.1021/cg1009048).

Xiang-Jun

1440

General discussions (Q&As) / Re: triplex building

« on: September 11, 2010, 12:13:12 am »

135D and 136D pdb structures do have the GGC triplex with GG base pair step similar to my repeat. May be that is a good place to start.

I had a quick look of 135D/136D and senses that the GGC triplet could serve a good starting point.

A simple way I can think of is: build B-DNA helix of 5 GAG repeats. cp the file, separate the the strands in one file and than dock the single strand on the double helix. I am going to be working this angle as well. ( as soon as I get my hands on a decent docking software  )

That appears to be a good/practical idea. Remember that [mono:3b4grcve]find_pair[/mono:3b4grcve] and [mono:3b4grcve]rebuild[/mono:3b4grcve] works for single strand as well. As long as you have a duplex, a single strand, and a triplet building block, the utility program [mono:3b4grcve]rotate_mol[/mono:3b4grcve] should be of help in assembling the two parts together, even though it may not qualify as "a decent docking software". Have a try, and see how far you can go. Once you get a prototype for GGC, it should be straightforward to apply it to AAT.

Xiang-Jun

[hr:3b4grcve][/hr:3b4grcve]
PS. Please be a bit patient -- I am quite occupied recently, and could be slow in getting back to you, especially on issues that require more explorations. Hopefully, I will able to devote more time/efforts on 3DNA-related stuffs to respond to the user community more promptly and to the depth/extent I'd like to.

1441

General discussions (Q&As) / Re: single-stranded base stacking

« on: September 10, 2010, 11:28:18 pm »

I am wondering if this software provides the information of possible single-stranded base stacking?

In addition to what Alpay has suggested, you can also run 3DNA as follows:

Code: [Select]

find_pair -s your.pdb stdout | analyzewhich will generates a file named [mono:3vpvq2pu]your.outs[/mono:3vpvq2pu], containing overlap areas between neighboring base rings. As an example, see "RNA Tetraloop Folding Reveals Tension between Backbone Restraints and Molecular Interactions" by Mohan et al. (J Am Chem Soc. 2010 Sep 15;132(36):12679-89).

Note that under the hook, w3DNA calls exactly the above commands.

HTH,

Xiang-Jun

1442

General discussions (Q&As) / Re: triplex building

« on: September 08, 2010, 08:05:31 pm »

I only have sequence info that is predicted to form a triplex, no structure information at all!. I am assuming that the triplex will form regular W-C bps on the B-DNA double strand and regular Hoogsten bps with the single strand.

It is helpful to know this background information. Then the question becomes what the "regular Hoogsten bps with the single strand" look like for the G.G-C and A.A-T triplet building blocks? While the A.U Hoogsteen pair is well-characterized, how about the G.G and A.A Hoogsteen pairs? They are prerequisite to build the triplexes you want. Wherever possible, I'd like to help you get started.

Xiang-Jun

PS. Thanks for your kind words about 3DNA, and for sharing your Python script for the analysis along the md trajectories.

1443

General discussions (Q&As) / Re: triplex building

« on: September 02, 2010, 09:45:50 pm »

Hi Alpay,

Thanks for posing an interesting question which is, however, beyond 3DNA's current capability. The triplex fiber models with 3DNA are based on experimental data collected from literature, and they require fixed sequences. Are your purine-purine-pyrimidine (RRY) triplex models (AAT and GGC) regular?  If so, do you have the structure data of the corresponding triplet repeats, A.A.T and G.G.C, and how such triplets stack over one another (twist, rise)? I am always interested in extending 3DNA's capabilities to meet users' needs. You may also find NAB (Nucleic Acid Builder) from Dr. David Case's group useful.

Xiang-Jun

1444

General discussions (Q&As) / Re: What do numbers in *.inp files mean?

« on: August 26, 2010, 12:06:32 am »

Well, this is a clearly defined question with a concrete example  and I am in a unique position to answer it   .

Could you please tell me what all five numbers in "0.32 0.17 17.11 8.91 -0.83" exactly mean?

The content after the # sign is not documented because it was initially used for debugging purpose only. Since you asked, here is a brief account of what the five numbers mean, using your specified example with unit in () at the end:

• the distance between origins of the two base reference frames (0.32 Å).
• the vertical distance between the base planes, i.e., |stagger| (0.17 Å).
• the angle between the two base normals (17.11°).
• distance between RN9 and YN1 of the pair (8.91 Å).
• an empirical measure combining a few other quantities; for deciding if a pair should be included in a double helix (-0.83).

Overall, the first four numbers are for identifying a base pair, while the fifth number is for checking its inclusion/exclusion in a duplex.

Over the years, [mono:3bemt574]find_pair[/mono:3bemt574] has turned out to be an essential 3DNA application that makes analyzing nucleic acid structure straightforward.  See my blog post "What find_pair in 3DNA can do". As I've mentioned in several occasions, a detailed account of the underlying algorithm of  [mono:3bemt574]find_pair[/mono:3bemt574] has still to be written. I've been continuously refining [mono:3bemt574]find_pair[/mono:3bemt574] internally, and hopefully I will be able to write a paper on it in the not too distant future.

HTH,

Xiang-Jun

1445

General discussions (Q&As) / Re: Transformation of DNA keeping it's curvature constant

« on: August 24, 2010, 08:03:51 pm »

Thanks for your clarification. It is certainly clearer. However, I am still confused how one could make base-pair movement while "keeping absolutely constant the backbone shape". Are you trying to make rigid body transformation; i.e., moving bp1 to bp2 while keeping the internal structure (base + backbone) unchanged? As suggested in my previous post, a simple/concrete/reproducible example would make your point unambiguous.

At this point, I do not think 3DNA has direct support to what you presumably want to do. However, if I could have a clear understanding of your question, I may be able to help in a more concrete way. Of course, always keep your mind open and try to explore other available tools.

Xiang-Jun

1446

General discussions (Q&As) / Re: Transformation of DNA keeping it's curvature constant

« on: August 23, 2010, 10:35:56 pm »

Hi Mikhail,

Thanks for posting such an interesting question. As you guessed, the operation you'd like to make "may seem strange to some people". Indeed, I do not fully understand what you are trying to achieve, exactly. Could you use a simple, concrete example so others can see your point clearly? I am always interested in adapting 3DNA in ways that make sense to me and helpful to users like you.

Xiang-Jun

1447

General discussions (Q&As) / Re: Re:pdb problem in 3DNA

« on: August 12, 2010, 10:02:26 pm »

Dear Raman,

I was trying to use 3DNA web server for analyzing of the RNA double stranded (CCGCUGCGG)2 but the server seems not to recognize the file.

w3DNA works (or should work) for the most common/standard cases, but it is not that sophisticated yet. Technically, I am not involved in supporting w3DNA per se. See the section "w3DNA web server topics".

I am not able to figure out the error source. I think it might be some atom types are not recognized. I will appreciate if you can give me any insight into how I can modify the files for proper format for 3DNA. I have attached the pdb file.

Thanks for your attached file, the issue becomes clear: it is related to the 'non-standard' PDB format you use -- read carefully the "Coordinate Section" of the standard PDB documentation. For illustration purpose, here is an comparison of two ATOM records, one from your file, another in standard PDB format:

Code: [Select]

# 1st from your file; 2nd in standard format
ATOM     40  O3' C A    2      -6.273 -12.136   8.883  1.00  0.00           O  
ATOM     47  O3'  DC A   3      24.468  20.110  19.126  1.00 15.21           O  
What software package did you use to generate the pseudo-'PDB' file? There should be a utility for converting it to standard PDB format. It is easy to write an ad hoc script for converting as well, see the following Ruby script (note while it works in this case, it serves only as a starting point for further refinement):

Code: [Select]

File.open('cug_new_3_3_ok.pdb', 'w') do |aFile|
    File.open('cug_new_3_3.pdb').each_line do |line|
        if line =~ /^(ATOM|HETATM)/
            sb = line[0, 17] # upto atom name plus alt-loc
            resname, chain_id, resnum = line[17, 8].split
            se = line[26..-1]
            aFile.puts "%17s%3s %1s%4s %s" % [sb, resname, chain_id, resnum, se]
        else
            aFile.puts line
        end
    end
end
With the above script, [mono:2x0a5suh]find_pair[/mono:2x0a5suh] will get the following base-pairing info

Code: [Select]

   1   18  0 #    1 | ....>A:...1_:[..C]C-----G[..G]:..24_:B<....  0.58  0.24 21.25  8.84 -2.88
    2   17  0 #    2 | ....>A:...2_:[..C]C-----G[..G]:..23_:B<....  1.20  0.86 31.10  8.66 -0.52
    3   16  0 #    3 | ....>A:...3_:[..G]G-----C[..C]:..22_:B<....  1.54  1.27 25.83  8.63  0.37
    4   15  0 #    4 | ....>A:...4_:[..C]C-----G[..G]:..21_:B<....  0.21  0.06 44.98  8.77 -2.42
    5   14  0 #    5 | ....>A:...5_:[..U]U-*---U[..U]:..20_:B<....  2.89  1.60 22.74  8.20  6.22
    6   13  0 #    6 | ....>A:...6_:[..G]G-----C[..C]:..19_:B<....  0.84  0.67 23.53  8.88 -1.66
    7   12  0 #    7 | ....>A:...7_:[..C]C-----G[..G]:..18_:B<....  0.65  0.46 35.94  8.95 -1.64
    8   11  0 #    8 | ....>A:...8_:[..G]G-----C[..C]:..17_:B<....  0.21  0.14 25.89  8.78 -3.22
    9   10  0 #    9 | ....>A:...9_:[..G]G-----C[..C]:..16_:B<....  0.84  0.69 25.34  8.83 -1.50HTH,

Xiang-Jun

1448

General discussions (Q&As) / Re: bending angle of DNA

« on: August 05, 2010, 12:38:59 am »

Dear Ralf,

Thanks for posting a question which should be of general interest. Overall, a (bending) angle is defined by two vectors. For DNA bending angle, the question then becomes how the two (helical) axes are calculated. In 3DNA, if the input structure is not heavily curved, the following section is produced (using [mono:330yz4f2]355d/bdl084[/mono:330yz4f2] as an example):

[pre:330yz4f2]Global linear helical axis defined by equivalent C1' and RN9/YN1 atom pairs                                        
Deviation from regular linear helix: 3.30(0.52)
Helix:    -0.127  -0.275  -0.953
HETATM 9998  XS    X X 999      17.536  25.713  25.665
HETATM 9999  XE    X X 999      12.911  15.677  -9.080
Average and standard deviation of helix radius:
      P: 9.42(0.82), O4': 6.37(0.85),  C1': 5.85(0.86)[/pre:330yz4f2]

If your structure is curved overall, but with two relatively straight fragments (via a kink), you will need to analyze the two segments separately to get the above information. With the two unit vectors (defined in Helix: line), it is straight forward to calculate the angle between them. Of course, another way to think of the issue is to fit a perfectly straight B-DNA fiber model onto each of the two relatively straight fragments, and calculate the angle between them. The two approaches should give comparable (but usually not identical results).

Several ago while at Rutgers, I helped Chen et al. (in Dr. Berman's group) analyzed some CAP-DNA complexes to quantify bending angles. I sense your case is similar to his. Please see: http://www.ncbi.nlm.nih.gov/pubmed/11724532 and http://www.ncbi.nlm.nih.gov/pubmed/11724533.

The DNA bending question has appeared quite a few times in the forum; you may find it helpful to check "Calculating the angle of DNA curvature".

Specifically, you may want to repeat 3DNA NP2008 paper, protocol (recipe) no. 4 on "Automatic identification of double-helical regions in a DNA–RNA junction", which provides detailed steps on calculating the angle between helices #1 and #3. Note also:

The chosen view allows for easy visualization and protractor measurement of the overall bending angle between the two relatively straight helices (the red horizontal helix h1 and the green vertical helix h3). This intuitive approach, initially developed to quantify the bending angles of a series of catabolic activator protein–DNA complexes67, yields sensible results when other methods fail.

HTH,

Xiang-Jun

1449

General discussions (Q&As) / Some details on PNAS98 DNA sequence-dependent deformability

« on: July 25, 2010, 11:54:58 am »

Recently I received a request from Mauricio, a graduate student from Dr. Olson's laboratory, for a technical question on generating the potential energy contour plots in the Olson et al. 1998 PNAS paper [PNAS98], "DNA sequence-dependent deformability deduced from protein–DNA crystal complexes." Then I vaguely remembered another PNAS98-related request from Dr. Olson several years ago, regarding details of data processing procedures.

PNAS98 is published a dozen years ago, and over the years it has made a big impact on the area of DNA-protein interactions, as evident by its over 400 citations. In retrospect, I was involved in the project mostly from a "technical" perspective: Andrey Gorin collected and analyzed (using CompDNA) the original DNA-protein complexes, and left the Olson lab before I came. I took over the project and performed further processing to generate all the table and figures used in PNAS98.

After publishing PNAS98, I have never done any follow up work regarding energetics of DNA-protein interactions. So all PNAS98-related files and scripts had been sleeping around, until I was contact by Dr. Olson in early 2006. It took me quite a while to dig them out, cleaned up and put them together. Four years later, now when Mauricio asks me a related question again, I tried to trace the emails I communicated with Dr. Olson on this matter. Unfortunately, my rutchem email account ceased to function from around summer 2006. So I have to check  backups from old machines I used ages ago ... It is from this time-consuming process (again) that I've decided to write this post to put together all PNAS98-related information I have collected, for the benefit of easy future reference.

The tarball PNAS98_Data.tar.gz contains the raw data I received from Gorin, and the scripts I used for further processing. See the included README.pdf and index.html for details. The Matlab script Ro_Tw_WKO.m was used to generate the ellipsoids contour plots of potential energies.  Around 1998, I think the version of Matlab I used was 5.x. Note that I do not have access to Matlab on my desktop nowadays.

For reasons detailed above, if you have further questions regarding PNAS98, please post them here instead of sending me emails, and I will try my best to answer them.

Xiang-Jun

1450

General discussions (Q&As) / Re: Methodology for sugar puckering calcualtion

« on: July 23, 2010, 07:34:52 pm »

The reference is: Altona, C. & Sundaralingam, M. (1972) "Conformational analysis of the sugar ring in nucleosides and nucleotides. A new description using the concept of pseudorotation." J. Am. Chem. Soc. 94, 8205-8212. See the SHCNAaP paper for further details; as mentioned in Section SCHNAaP/SCHNArP of the forum, [mono:2tedyoak]analyze[/mono:2tedyoak] in 3DNA is derived directly from SCHNAaP.

HTH,

Xiang-Jun