Messages

1176

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 15, 2012, 12:14:16 pm »

The updated 3DNA v2.1 (2012dec15) should do the trick for you. In particular, I added the -ring option to 'analyze' and the corresponding 'x3dna_ensemble analyze' script. By default, the -ring option is not set for back compatibility. So other users won't notice any differences.

For example, in directory $X3DNA/examples/ensemble/md, you can run the following command:

x3dna_ensemble analyze -b bpfile.dat -e sample_md0.pdb -r

Note the -r option (short-hand form for -ring).

When you run x3dna_ensemble extract -l, you will find the new parameters are named as below:

Code: [Select]

[ 5] b1c_xyz             [ 6] b1n_vec             [ 7] b2c_xyz             [ 8] b2n_vec
Please verify and report back how it goes.

Xiang-Jun

1177

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 13, 2012, 11:03:57 am »

What do you mean exactly by "position of center of bases"? Geometric center of all base atoms, or just ring atoms? I may consider to add such info that can meet your needs and fit into other parts of 3DNA.

Xiang-Jun

1178

General discussions (Q&As) / Re: Large value of ZpH and inclination for Z-DNA

« on: December 10, 2012, 08:55:22 pm »

Hi Paul,

Thanks for posting back. In addtion to what I said in my previous reply, please note that 3DNA caluclates local step and helical parameters which could thus be strongly influenced by regional distortions. The same argument is true for the derived Xp/Yp/Zp parameters and their corresponding helical variants. The situation is especially noticeable for Z-DNA, as in your case, where the repeating unit is two base pairs but the structure is analyzed the 'normal' way (as in A- and B-DNA).

Other users of the forum may want to comment on your questions. Did you also post your questions on the CCP4 mailing list?

Xiang-Jun

1179

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 10, 2012, 01:17:01 pm »

Hi Ali,

Thanks for providing a sample AMBER MD trajectory file (actg3_md.pdb) which allow me to trace where the problem is.

A portion of PDB ATOM record from the file is as below:

Code: [Select]

         1         2         3         4         5         6         7         8
12345678901234567890123456789012345678901234567890123456789012345678901234567890
ATOM     11  N9  DA5     1       0.781  -4.705  -0.553  0.00  0.00             
ATOM     12  C8  DA5     1      -0.573  -5.051  -0.401  0.00  0.00             
ATOM     13  H8  DA5     1      -0.879  -6.079  -0.526  0.00  0.00             
ATOM     14  N7  DA5     1      -1.349  -3.982  -0.328  0.00  0.00             
ATOM     15  C5  DA5     1      -0.421  -2.977  -0.164  0.00  0.00             
ATOM     16  C6  DA5     1      -0.512  -1.582   0.086  0.00  0.00       
According to the PDB format specification, columns 55 to 60 [Real(6.2)] is for atom occupancy. Normally, it is 1.00 -- check one of PDB entries (e.g., 355d) for an example. Now, as you can see, AMBER puts 0.00 there. It happens that in the 2012dec09 release of 3DNA, I added checking for atom occupancy (see "What's new?" for details). It is in that release that atoms with zero occupancy are ignored -- so find_pair would judge that your MD trajectories contains no base pairs. That explains your observation:

By the previous release (2012Nov26) there is not any problem!

You raised an interesting question:

Is it not possible that 3DNA analyze directly the output trajectory file of famous MD codes such as Amber?

Surely I'll like to make 3DNA directly applicable to the output trajectory file of the famous AMBER package. So I have now made checking for -occupancy optional that can be turned on by user explicitly. Download the updated 3DNA v2.1 2012dec10 release, and it should have solved the problem.

Xiang-Jun

1180

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 10, 2012, 08:02:48 am »

Thanks for letting me know the new issues you experienced -- I will look not them and get back to you shortly.

Xiang-Jun

1181

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 09, 2012, 08:48:26 pm »

Hi Ali,

Thanks for clarifying your question and providing further details on what you aim to achieve.

Following my previous reply, I've revised the Ruby script associated with x3dna_ensemble to parse the section "Origin (Ox, Oy, Oz) and mean normal vector (Nx, Ny, Nz) of each base-pair in the coordinate system of the given structure". The two newly-added parameters to extract are: bpo_xyz and bpn_vec. Please download 3DNA v2.1(2012dec09) and let me know if that fits the bill.

Since you are interested in base stacking interactions, your may also want to check the section "Overlap area in Angstrom^2 between polygons defined by atoms on successive bases. Polygons projected in the mean plane of the designed base-pair step."

HTH,

Xiang-Jun

1182

MD simulations / Re: Base stacking from x3dna_ensemble

« on: December 08, 2012, 08:48:13 pm »

Hi Ali,

Thanks for using 3DNA and posting your questions on the forum. With feedbacks from users like you, 3DNA can only become more useful to the community.

I think I understand your question: the 3DNA analyze program includes a section as below:

****************************************************************************
Origin (Ox, Oy, Oz) and mean normal vector (Nx, Ny, Nz) of each base-pair in
   the coordinate system of the given structure

      bp        Ox        Oy        Oz        Nx        Ny        Nz
    1 C-G      17.13     25.96     25.88     -0.10     -0.44     -0.89
    2 G-C      16.56     24.81     22.95     -0.21     -0.33     -0.92
    3 C-G      16.16     23.97     19.28     -0.18     -0.48     -0.86
...........................................................................

However, the parameters herein are not in x3dna_ensemble output. Am I right?

That section is currently not parsed by the x3dna_ensemble script -- you are the first to make this request. I will consider to add the portion in x3dna_ensemble output in the next release of 3DNA v2.1(beta) in the near future. What format do you prefer?

Xiang-Jun

1183

General discussions (Q&As) / Re: Large value of ZpH and inclination for Z-DNA

« on: December 03, 2012, 12:10:56 am »

Hi paul,

Thanks for using 3DNA and posting your questions on the forum.

As detailed in the 2003 3DNA NAR paper, the set of helical parameters (X-disp, Y-disp, h-Rise, Incl., Tip, h-Twist) parallels that of the stacking parameters (Shift, Slide, Rise, Tilt, Roll, Twist); each represents a rigorous mathematical description of the relative geometry (position and orientation) of the two base pairs involved. You can verify the parameters by rebuilding an atomic model using the output file "bp_helical.par" -- the RMSD of base atoms between 3DNA rebuilt structure and the original one should be virtually zero.

From your list of the parameters, it's clear that two CG/CG steps differ from each other: what is the RMSD between them? Do you see "something specific" there by using a visualization tool (PyMOL/Jmol)? As always, 3DNA outputted parameters should be checked along with other means and your understanding of your structure to make "sense".

As elaborated in the thread "A-DNA definition" on the forum, the Zp vs ZpH parameters are mainly used to distinguish right-handed A-, B- and TA-DNA forms. I have not checked how the ZpH values vary in left-handed Z-DNA.

Z-DNA has dinucleotide as a repeating unit, so you need to add (-6 + -54) or analyze alternating base pairs to get the ~-60° twist angle.

To get started with DNA structures, I highly recommend the book "Understanding DNA: The Molecule and How it Works" by Calladine. You may check publications by Dr. Remo Rohs on DNA shape (minor groove width).

HTH,

Xiang-Jun

1184

RNA structures (DSSR) / Re: Can 3DNA be used to build a miRNA-target structure?

« on: November 21, 2012, 12:25:59 pm »

Hi Guido Leoni,

Thanks for providing such a detailed explanation of your miRNA-mRNA pairment question. I can now clearly see what you want to achieve, and I understand the importance of your research area.

Am I right to assume that the RNA structure you want to build is (with code tag for monospace)?

Code: [Select]

U A A A G U G C U U A U
| | | | | | | |     | |
A U U U C A C G C C U A

There are many tools dedicated to building three-dimensional RNA structures, see for example "RNA-Puzzles: A CASP-like evaluation of RNA three-dimensional structure prediction". You may want to check some of the tools included in the evaluation paper, or try other (newer) ones.

If you want to try out 3DNA, here are the procedures to build an approximate structure with the above sequence and secondary structure:

# build an RNA duplex with Watson-Crick base pairs
fiber -se=UAAAGUGCUUAU -r RNA-WC-duplex.pdb
find_pair RNA-WC-duplex.pdb stdout
# as shown below:
NA-WC-duplex.pdb
RNA-WC-duplex.out
    2         # duplex
   12         # number of base-pairs
    1    1    # explicit bp numbering/hetero atoms
    1   24  0 #    1 | ....>A:...1_:[..U]U-----A[..A]:..24_:B<....  0.11  0.06 10.52  8.94 -4.24
    2   23  0 #    2 | ....>A:...2_:[..A]A-----U[..U]:..23_:B<....  0.11  0.06 10.52  8.94 -4.24
    3   22  0 #    3 | ....>A:...3_:[..A]A-----U[..U]:..22_:B<....  0.11  0.06 10.52  8.94 -4.24
    4   21  0 #    4 | ....>A:...4_:[..A]A-----U[..U]:..21_:B<....  0.11  0.06 10.51  8.94 -4.24
    5   20  0 #    5 | ....>A:...5_:[..G]G-----C[..C]:..20_:B<....  0.24  0.07 10.52  8.94 -4.10
    6   19  0 #    6 | ....>A:...6_:[..U]U-----A[..A]:..19_:B<....  0.11  0.06 10.52  8.95 -4.24
    7   18  0 #    7 | ....>A:...7_:[..G]G-----C[..C]:..18_:B<....  0.24  0.07 10.52  8.94 -4.09
    8   17  0 #    8 | ....>A:...8_:[..C]C-----G[..G]:..17_:B<....  0.24  0.07 10.52  8.95 -4.10
    9   16  0 #    9 | ....>A:...9_:[..U]U-----A[..A]:..16_:B<....  0.11  0.06 10.52  8.94 -4.24
   10   15  0 #   10 | ....>A:..10_:[..U]U-----A[..A]:..15_:B<....  0.11  0.06 10.52  8.94 -4.24
   11   14  0 #   11 | ....>A:..11_:[..A]A-----U[..U]:..14_:B<....  0.11  0.06 10.52  8.94 -4.24
   12   13  0 #   12 | ....>A:..12_:[..U]U-----A[..A]:..13_:B<....  0.11  0.06 10.52  8.94 -4.24
##### Base-pair criteria used:     4.00     0.00    15.00     2.50    65.00     4.50     7.50 [ O N]
##### 0 non-Watson-Crick base-pairs, and 1 helix (0 isolated bps)
##### Helix #1 (12): 1 - 12

# now mutate A15 and A16 on chain B to C
mutate_bases 'c=B s=16 m=C; c=B s=15 m=C' RNA-WC-duplex.pdb RNA-ok-duplex.pdb
find_pair RNA-ok-duplex.pdb stdout
# now you get the desired miRNA-mRNA duplex with two U--C mis-matches: 
RNA-ok-duplex.pdb
RNA-ok-duplex.out
    2         # duplex
   12         # number of base-pairs
    1    1    # explicit bp numbering/hetero atoms
    1   24  0 #    1 | ....>A:...1_:[..U]U-----A[..A]:..24_:B<....  0.11  0.06 10.52  8.94 -4.24
    2   23  0 #    2 | ....>A:...2_:[..A]A-----U[..U]:..23_:B<....  0.11  0.06 10.52  8.94 -4.24
    3   22  0 #    3 | ....>A:...3_:[..A]A-----U[..U]:..22_:B<....  0.11  0.06 10.52  8.94 -4.24
    4   21  0 #    4 | ....>A:...4_:[..A]A-----U[..U]:..21_:B<....  0.11  0.06 10.51  8.94 -4.24
    5   20  0 #    5 | ....>A:...5_:[..G]G-----C[..C]:..20_:B<....  0.24  0.07 10.52  8.94 -4.10
    6   19  0 #    6 | ....>A:...6_:[..U]U-----A[..A]:..19_:B<....  0.11  0.06 10.52  8.95 -4.24
    7   18  0 #    7 | ....>A:...7_:[..G]G-----C[..C]:..18_:B<....  0.24  0.07 10.52  8.94 -4.09
    8   17  0 #    8 | ....>A:...8_:[..C]C-----G[..G]:..17_:B<....  0.24  0.07 10.52  8.95 -4.10
    9   16  0 #    9 | ....>A:...9_:[..U]U-**--C[..C]:..16_:B<....  0.11  0.06 10.52  8.99  0.76
   10   15  0 #   10 | ....>A:..10_:[..U]U-**--C[..C]:..15_:B<....  0.11  0.06 10.51  8.99  0.76
   11   14  0 #   11 | ....>A:..11_:[..A]A-----U[..U]:..14_:B<....  0.11  0.06 10.52  8.94 -4.24
   12   13  0 #   12 | ....>A:..12_:[..U]U-----A[..A]:..13_:B<....  0.11  0.06 10.52  8.94 -4.24
##### Base-pair criteria used:     4.00     0.00    15.00     2.50    65.00     4.50     7.50 [ O N]
##### 2 non-Watson-Crick base-pairs, and 1 helix (0 isolated bps)
##### Helix #1 (12): 1 - 12

Note that you need to use the current version of 3DNA v2.1 from the download page. For your reference, attached below are the two PDB files RNA-WC-duplex.pdb and RNA-ok-duplex.pdb mentioned above.

1185

w3DNA -- web interface to 3DNA / Re: O3'-P distance too long after reconstruction

« on: November 21, 2012, 11:20:27 am »

- The default O3'-P distance is 4.5A. From the ADIT server of the PDB, the expected value is 1.6A. What is in fact the range of the possible values?

The O3'-P covalent bond distance is ~1.6 A, as you noticed on the ADIT server of the PDB. The 3DNA default of 4.5 A is purely an empirical value used to decide if to include the bond in a corresponding CONECT record in the rebuilt PDB file; it has no 'chemical' meaning. Have a look of build.pdb file using a text editor, you will see my point.

- Could you indicate some tools alllowing to do energy minimization?

AMBER should help. Some other MD packages or Phenix should also do the trick. Yet, I still failed to find a command-line tool that can 'regularize' the backbone to a reasonable geometry while keeping the base atoms fixed.

HTH,

Xiang-Jun

1186

MD simulations / Re: analyzing longer DNA sequences

« on: November 21, 2012, 11:05:56 am »

I am not sure if this is too much to ask you.

No, it's not; I always welcome user questions such as this one, and I strive to be as helpful as I could.

Now I see the problem you are experiencing. Strictly speaking, and as I mentioned in my previous reply, it's not a 3DNA problem but at the interface between 3DNA and Curves+. Since the purpose of providing the find_pair c+ option is to build a bridge between the two commonly used software programs for analyzing nucleic acid structures, I'd like to dig the issue further to see if anything can be done from 3DNA's perspective.

Your Curves+ input file curves.inp, as generated with find_pair -c+, has the following content:

&inp file=sel.pdb,
     lis=sel,
     fit=.t.,
     lib=./standard,
     isym=1,
&end
    2    1   -1    0    0
    2    3    4    5    6    7    8    9   10   11   12   13   14   15   16   17   18   19   20   21   22   23   24
   47   46   45   44   43   42   41   40   39   38   37   36   35   34   33   32   31   30   29   28   27   26   25

which has 23 base pairs (note that bases 1 and 48 are not paired). Yet when the file is fed to Curves+, only the first 15 bps are recognized.

Combined strands have   15 levels ...

  Strand  1 has  15 bases (5'-3'): GTGTGAGCGTGGGCG
  Strand  2 has  15 bases (3'-5'): CACACTCGCACCCGC

To help solve the problem, could you try the following and report back (in detail) what you get?

• Instead of 23 bps, shorten the list to < 15, say 10, as below:
  

      2    3    4    5    6    7    8    9   10   11
     47   46   45   44   43   42   41   40   39   38

  Run Curves+ on it again, do you get what you expect?
  
• Since the nucleotide numbers are continuous, you can use the short-hand form to specify paired bases:
  

      2:24
     47:25

  Run Curves+, what do you get?
  

Xiang-Jun

1187

RNA structures (DSSR) / Re: Can 3DNA be used to build a miRNA-target structure?

« on: November 21, 2012, 10:33:48 am »

Could you be more specific, better with an example (or reference), to illustrate exactly what want to achieve?

Xiang-Jun

1188

w3DNA -- web interface to 3DNA / Re: O3'-P distance too long after reconstruction

« on: November 21, 2012, 10:31:54 am »

Hi Damien,

The fact that "O3'-P distance too long after reconstruction" is well expected; 3DNA built structures are accurate for the bases, with only an approximate sugar-phosphate backbone. See FAQ "How do I build nucleic acid structures with sugar-phosphate backbone?"

For cases with longer than the default 4.5 A O3'(i)--P(i+1) distance, the 3DNA rebuild program outputs info message as below (using your example, and with a B-DNA backbone conformation):

Code: [Select]

O3' (#317) and P (#331) on chain A have distance 5.3 over 4.5: no linkage assigned
O3' (#768) and P (#782) on chain B have distance 5.2 over 4.5: no linkage assigned
This approximate structure may serve as a starting point for energy minimization.

Xiang-Jun
 

1189

MD simulations / Re: analyzing longer DNA sequences

« on: November 20, 2012, 04:29:58 pm »

Hi Shyno,

Thanks for providing details of the commands you used and attaching three relevant files. However, I fail to see what're wrong here; things are working as expected from my understanding.

With the -c+ option, you get what's desired, as in your attached curves.inp file. Running find_pair in its default settings on your PDB file sel.pdb gives expected results:

find_pair sel.pdb stdout
sel.pdb
sel.out
    2         # duplex
   23         # number of base-pairs
    1    1    # explicit bp numbering/hetero atoms
    2   47  0 #    1 | ....>A:...2_:[..G]G-----C[..C]:..48_:B<....  0.77  0.76 28.01  8.85 -1.31
    3   46  0 #    2 | ....>A:...3_:[..T]T-----A[..A]:..47_:B<....  0.32  0.02 11.50  9.26 -4.07
    4   45  0 #    3 | ....>A:...4_:[..G]G-----C[..C]:..46_:B<....  0.83  0.64  4.42  9.11 -2.68
    5   44  0 #    4 | ....>A:...5_:[..T]T-----A[..A]:..45_:B<....  0.54  0.32 21.12  8.91 -2.76
    6   43  0 #    5 | ....>A:...6_:[..G]G-----C[..C]:..44_:B<....  0.45  0.27 27.61  8.88 -2.64
    7   42  0 #    6 | ....>A:...7_:[..A]A-**+-T[..T]:..43_:B<....  3.71  0.96 28.75  7.04  4.07
    8   41  0 #    7 | ....>A:...8_:[..G]G-----C[..C]:..42_:B<....  0.45  0.32  3.39  8.97 -3.75
    9   40  0 #    8 | ....>A:...9_:[..C]C-----G[..G]:..41_:B<....  1.05  0.47  8.17  8.82 -2.61
   10   39  0 #    9 | ....>A:..10_:[..G]G-----C[..C]:..40_:B<....  0.76  0.50 21.19  8.86 -2.17
   11   38  0 #   10 | ....>A:..11_:[..T]T-----A[..A]:..39_:B<....  0.34  0.27  5.44  9.01 -3.84
   12   37  0 #   11 | ....>A:..12_:[..G]G-----C[..C]:..38_:B<....  0.58  0.48 12.17  9.02 -2.85
   13   36  0 #   12 | ....>A:..13_:[..G]G-----C[..C]:..37_:B<....  0.24  0.03 16.54  9.08 -3.88
   14   35  0 #   13 | ....>A:..14_:[..G]G-----C[..C]:..36_:B<....  0.69  0.69 14.12  9.14 -2.22
   15   34  0 #   14 | ....>A:..15_:[..C]C-----G[..G]:..35_:B<....  0.41  0.09 14.35  9.02 -3.70
   16   33  0 #   15 | ....>A:..16_:[..G]G-----C[..C]:..34_:B<....  0.40  0.39 26.56  9.03 -2.49
   17   32  0 #   16 | ....>A:..17_:[..T]T-----A[..A]:..33_:B<....  0.70  0.62  8.01  9.49 -2.65
   18   31  0 #   17 | ....>A:..18_:[..A]A-----T[..T]:..32_:B<....  0.26  0.21 18.44  8.96 -3.41
   19   30  0 #   18 | ....>A:..19_:[..C]C-----G[..G]:..31_:B<....  0.26  0.09 25.81  9.27 -3.27
   20   29  0 #   19 | ....>A:..20_:[..A]A-----T[..T]:..30_:B<....  0.63  0.21 19.54  8.96 -2.98
   21   28  0 #   20 | ....>A:..21_:[..C]C-----G[..G]:..29_:B<....  0.57  0.33  6.64  9.06 -3.43
   22   27  0 #   21 | ....>A:..22_:[..A]A-----T[..T]:..28_:B<....  1.09  1.01 10.89  8.63 -1.34
   23   26  0 #   22 | ....>A:..23_:[..C]C-----G[..G]:..27_:B<....  0.66  0.37 28.93  9.04 -2.16
   24   25  0 #   23 | ....>A:..24_:[..A]A-----T[..T]:..26_:B<....  2.72  2.29 25.44  7.75  5.57
##### Base-pair criteria used:     4.00     0.00    15.00     2.50    65.00     4.50     7.50 [ O N]
##### 1 non-Watson-Crick base-pair, and 1 helix (0 isolated bps)
##### Helix #1 (23): 1 - 23

Certainly, find_pair is working properly, as designed. Is there anything I am missing here?

Xiang-Jun

1190

MD simulations / Re: analyzing longer DNA sequences

« on: November 20, 2012, 02:22:59 pm »

Hi Shyno,

DNAs with 62 base pairs should not be a problem for 3DNA to analyze; specifically, find_pair and other 3DNA components have no default bp limit other than your computer's memory.

As always, please be specific by providing a reproducible example; that will help solve your problem.

Xiang-Jun

1191

General discussions (Q&As) / Re: Problems with installing 3DNA on MinGW/MSYS

« on: November 11, 2012, 11:00:52 pm »

Thanks for posting your problem on installing 3DNA. Am I right to assume that your system is Windows with MinGW/MSYS? Could you be more specific about your Windows systems -- XP, Vista or Windows 7? Is it 32 bit or 64 bit?

Xiang-Jun

1192

General discussions (Q&As) / Re: different geometries

« on: November 06, 2012, 10:18:50 am »

Could you be more specific by providing an example to put your question in context?

Xiang-Jun

1193

General discussions (Q&As) / Re: DNA elastic parameters from NDB database

« on: November 05, 2012, 11:32:27 am »

No, these new values are not included in 3DNA. You're advised to contact the authors of the 2009 Balasubramanian paper for further details.

For those who are interested in the original Olson et al. 1998 dataset, please check the post "Some details on PNAS98 DNA sequence-dependent deformability".

Xiang-Jun

1194

w3DNA -- web interface to 3DNA / Re: server down?

« on: November 03, 2012, 12:28:40 pm »

Hi Ludo,

Thanks for reporting the w3DNA server problem. It is indeed down right now, possibly due to Hurricane Sandy. I've informed those in change at Rutgers, and hopefully w3DNA will be back soon.

Xiang-Jun

1195

General discussions (Q&As) / Re: different geometries

« on: October 29, 2012, 04:12:43 pm »

It's closer, but still not a proper PDB format -- see the thread "how the cartesian coordinates transform to PDB format" for details.

Given the fact that this issue has popped up quite a few times over the years, I may consider adding support in 3DNA to make such conversion more straightforward.

Xiang-Jun

1196

General discussions (Q&As) / Re: different geometries

« on: October 29, 2012, 01:06:52 pm »

Thanks for posting back an example molden file which help clarify the issue. No, that's not the format recognized by 3DNA programs. Currently, 3DNA only accepts PDB format file (coordinate record descriptions) as documented in the RCSB website. That said, it's feasible to convert your molden file via a purpose-oriented script or third-party utility -- Google search may help.

Xiang-Jun

1197

Bug reports / Re: open_file <./Atomic.g.pdb> failed: No such file or directory

« on: October 26, 2012, 01:46:07 pm »

As a follow up, as of 3DNA v2.1 2012oct26, your initial problem has been solved. Now you can have for example only Atomic_A.pdb in your current working directory, and 3DNA will use that file; for the Atomic*.pdb files not found in the current directory, 3DNA will use the default in $X3DNA/config. Overall, this revision allows 3DNA users greater flexibility in choosing the standard base-reference-frame files for analysis and rebuilding.

Xiang-Jun

1198

General discussions (Q&As) / Re: different geometries

« on: October 26, 2012, 01:37:31 pm »

Thanks for using 3DNA and posting your question on the forum!

Could you be more specific by providing a concreate example to show us what a molden file look like?

Xiang-Jun

1199

FAQs / How can I mutate cytosine to 5-methylcytosine

« on: October 26, 2012, 01:23:14 pm »

Methylation of cytosines in DNA is a crucial epigenetic modification that regulate expression of many genes. Chemically, it is the addition of a methyl group to the 5 position of cytosine (C).

The mutate_bases program in 3DNA v2.x performs in silico base mutations given a nucleic acid structure in PDB format. It is not a problem to mutate any C to a 5-methylcytosine (5CM) provided that users set a 5CM in its standard base reference frame. Given the importance of 5CM in epigenetics and the increasing simulation studies to understand its effects, I have included Atomic_5CM.pdb in the 3DNA v2.1 distribution as of 2012oct26.

According to PDB, the three-letter nucleotide name for 5-methylcytosine is 5CM instead of 5MC -- see for example PDB entries 4mht and 2uz4. The methyl carbon is named " C5A" instead of " C5M" or " C7 ". Thus, the content of the Atomic_5CM.pdb file is:

Code: [Select]

REMARK    3DNA by Dr. Xiang-Jun Lu [2012-10-26] (xiangjun@x3dna.org)
ATOM      1  C1' 5CM A   1      -2.477   5.402   0.000  1.00  0.00           C 
ATOM      2  N1  5CM A   1      -1.285   4.542   0.000  1.00  0.00           N 
ATOM      3  C2  5CM A   1      -1.472   3.158   0.000  1.00  0.00           C 
ATOM      4  O2  5CM A   1      -2.628   2.709   0.001  1.00  0.00           O 
ATOM      5  N3  5CM A   1      -0.391   2.344   0.000  1.00  0.00           N 
ATOM      6  C4  5CM A   1       0.837   2.868   0.000  1.00  0.00           C 
ATOM      7  N4  5CM A   1       1.875   2.027   0.001  1.00  0.00           N 
ATOM      8  C5  5CM A   1       1.056   4.275   0.000  1.00  0.00           C 
ATOM      9  C5A 5CM A   1       2.466   4.961   0.001  1.00  0.00           C 
ATOM     10  C6  5CM A   1      -0.023   5.068   0.000  1.00  0.00           C 
END
With this new addition, it is now very straightforward to mutate Cs to 5CMs with mutate_bases, as illustrated by the following two examples:

• Mutate C1 on chain A and C23 on chain B of the Dickerson B-DNA dodecamer (PDB entry 355d) to 5CMs:
  

  mutate_bases 'chain=A snum=1 m=5CM; chain=B snum=23 m=5CM' 355d.pdb 355d_AC1BC23_5CM.pdb

• Mutate C2 on chain A of the yeast phenylalanine tRNA (PDB entry 6tna) to 5CM:
  

  mutate_bases 'chain=A snum=2 name=C m=5CM' 6tna.pdb 6tna_C2_5CM.pdb

The mutated files 355d_AC1BC23_5CM.pdb and 6tna_C2_5CM.pdb are attached for your verification. For comparison, shown below are the original atomic coordinates of the above tRNA 6tna cytosine and coordinates of its 5CM mutant in red. Note that the coordinates of the backbone atoms are the same, and coordinates of common base atoms are very close.

• The original atomic coordinates of a cytosine from PDB entry 6tna:
  --------------------------------------------------------------------------------
  ATOM     25  P     C A   2      31.659  20.469  70.978  1.00 10.00           P  
  ATOM     26  OP1   C A   2      32.973  21.044  71.364  1.00 10.00           O  
  ATOM     27  OP2   C A   2      30.973  21.143  69.849  1.00 10.00           O  
  ATOM     28  O5'   C A   2      31.815  18.912  70.652  1.00 10.00           O  
  ATOM     29  C5'   C A   2      30.629  18.184  70.293  1.00 10.00           C  
  ATOM     30  C4'   C A   2      30.507  16.914  71.139  1.00 10.00           C  
  ATOM     31  O4'   C A   2      29.293  17.051  71.947  1.00 10.00           O  
  ATOM     32  C3'   C A   2      30.455  15.607  70.367  1.00 10.00           C  
  ATOM     33  O3'   C A   2      31.724  14.971  70.316  1.00 10.00           O  
  ATOM     34  C2'   C A   2      29.411  14.815  71.146  1.00 10.00           C  
  ATOM     35  O2'   C A   2      29.987  14.227  72.301  1.00 10.00           O  
  ATOM     36  C1'   C A   2      28.473  15.927  71.630  1.00 10.00           C  
  ATOM     37  N1    C A   2      27.474  16.346  70.621  1.00 10.00           N  
  ATOM     38  C2    C A   2      26.658  15.368  70.068  1.00 10.00           C  
  ATOM     39  O2    C A   2      26.802  14.198  70.441  1.00 10.00           O  
  ATOM     40  N3    C A   2      25.726  15.730  69.143  1.00 10.00           N  
  ATOM     41  C4    C A   2      25.601  17.008  68.767  1.00 10.00           C  
  ATOM     42  N4    C A   2      24.682  17.314  67.872  1.00 10.00           N  
  ATOM     43  C5    C A   2      26.436  18.041  69.324  1.00 10.00           C  
  ATOM     44  C6    C A   2      27.351  17.658  70.243  1.00 10.00           C  
  --------------------------------------------------------------------------------
  The coordinates of the mutant 5-methylcytosine generated by 'mutate_bases'
  REMARK    Mutation#1 A:...2@:[..C] to [5CM]
  ATOM     25  P   5CM A   2      31.659  20.469  70.978  1.00 10.00           P  
  ATOM     26  OP1 5CM A   2      32.973  21.044  71.364  1.00 10.00           O  
  ATOM     27  OP2 5CM A   2      30.973  21.143  69.849  1.00 10.00           O  
  ATOM     28  O5' 5CM A   2      31.815  18.912  70.652  1.00 10.00           O  
  ATOM     29  C5' 5CM A   2      30.629  18.184  70.293  1.00 10.00           C  
  ATOM     30  C4' 5CM A   2      30.507  16.914  71.139  1.00 10.00           C  
  ATOM     31  O4' 5CM A   2      29.293  17.051  71.947  1.00 10.00           O  
  ATOM     32  C3' 5CM A   2      30.455  15.607  70.367  1.00 10.00           C  
  ATOM     33  O3' 5CM A   2      31.724  14.971  70.316  1.00 10.00           O  
  ATOM     34  C2' 5CM A   2      29.411  14.815  71.146  1.00 10.00           C  
  ATOM     35  O2' 5CM A   2      29.987  14.227  72.301  1.00 10.00           O  
  ATOM     36  C1' 5CM A   2      28.473  15.927  71.630  1.00 10.00           C  
  ATOM     37  N1  5CM A   2      27.475  16.353  70.620  1.00  1.00           N  
  ATOM     38  C2  5CM A   2      26.651  15.372  70.062  1.00  1.00           C  
  ATOM     39  O2  5CM A   2      26.789  14.195  70.427  1.00  1.00           O  
  ATOM     40  N3  5CM A   2      25.726  15.730  69.141  1.00  1.00           N  
  ATOM     41  C4  5CM A   2      25.610  17.009  68.776  1.00  1.00           C  
  ATOM     42  N4  5CM A   2      24.685  17.316  67.863  1.00  1.00           N  
  ATOM     43  C5  5CM A   2      26.437  18.028  69.328  1.00  1.00           C  
  ATOM     44  C5A 5CM A   2      26.359  19.547  68.947  1.00  1.00           C  
  ATOM     45  C6  5CM A   2      27.347  17.660  70.238  1.00  1.00           C  

1200

Bug reports / Re: open_file <./Atomic.g.pdb> failed: No such file or directory

« on: October 24, 2012, 11:11:33 am »

Hi Mauricio,

Thanks for your report. I can reproduce the 'problem'; it has nothing to do with including $X3DNA/config in the environment settings.

The error message:

Code: [Select]

open_file <./Atomic.g.pdb> failed: No such file or directorymeans 3DNA (find_pair) is trying to locate the file in your current working directory (CWD), instead of the default system setting $X3DNA/config where the file Atomic.g.pdb is available, among other Atomic*.pdb files.

If file Atomic_A.pdb exists in your CWD, 3DNA assumes all other Atomic*.pdb files there as well. That's normally true if you run, e.g.:

Code: [Select]

x3dna_utils cp_std bdna
For your case, you must have Atomic_A.pdb and possibly several other canonical base Atomic_[CGTU].pdb files in your CWD, but not the modified version (in lower case, prefixed with a dot instead of underscore). So simply run the following command in you CWD will do the trick:

Code: [Select]

cp -f $X3DNA/config/Atomic.*.pdb .
Alternatively, you can delete all the Atomic*pdb files from your CWD, and find_pair will work as expected:

Code: [Select]

rm -f Atomic*.pdb
find_pair 1ehz.pdb stdout
Please have a try and report back how it goes.

I may consider to refine 3DNA to check for each Atomic*.pdb file separately, but that would complicate the code. You are actually the first to notice this 'limitation'. Practically, knowing what's happening behind the scene, you can easily work around it, as suggested above.

HTH,

Xiang-Jun