Re: [Jmol-users] Failure to recognize RNA?

[Eric Martz]

Dear Bob,

I understand the present behavior in Jmol, but I don't think it is optimal.

I think this behavior in Jmol is left over from years ago when the PDB format did not distinguish DNA from RNA. That was a problem in RasMol and Chime. But since 2007, the PDB format has distinguished DNA from RNA. So "A" is RNA and "DA" is DNA. The list of standard nucleotides in the PDB format is here:
http://www.wwpdb.org/documentation/format33/sect4.html#HET

Currently in Jmol, "select A" selects both DNA and RNA, while "select DA" selects only DNA. RNA can be selected with "select A and RNA".

In the case cited (4olb) there are two partial RNA residues (atoms missing), part of a chain, that are not deemed RNA by Jmol, A8 and A9 in chain B. I do not see any way to select these residues as "RNA". Clearly the usual "select A and RNA" fails. They can be selected with "select nucleic and not (RNA, DNA)" but isn't that bizarre? And who would think to check that?

If an atom belongs to a residue with a standard RNA or DNA residue name, I think Jmol should accept and use that identification, even when key identifier atoms are missing, such as in the case cited. In such cases, the present behavior is confusing at best or misleading at worst. It causes FirstGlance to give incorrect or discrepant reports.

If a change is worrisome for backwards compatibility, then I'd like to request a new settable parameter that makes Jmol obey the current PDB format in this regard for standard nucleotide residues. Thus, selecting "RNA" in 4olb would select A8 and A9 in chain B as well as the other RNA residues that have no missing atoms. That seems to me the simple solution. (For non-standard nucleotides, the present behavior can stand, I suppose, e.g. 1evv.)

When the proposed new parameter is set, both "select A" and "select A and RNA" would work, and the problematic A's would be also be selected with "select RNA". However, "select A" would no longer select "DA" (DNA).

Thanks, Eric

At 6/2/14, Robert Hanson wrote:

"nucleic" is based on the residue name being an identified nucleic residue when there is not other evidence that it is RNA or DNA.

"dna" and "rna" depend upon specific sets of atoms.

In  this case, the label "A" identifies these as nucleic, but nothing identifies them as RNA or DNA.



On Sun, Jun 1, 2014 at 11:16 PM, Eric Martz < ema...@microbio.umass.edu> wrote:

4olb contains a 9-residue poly-A RNA chain "B". The last 2 residues  have missing atoms:

REMARK 470                                                                     

REMARK 470 MISSING ATOM                                                        

REMARK 470 THE FOLLOWING RESIDUES HAVE MISSING ATOMS (M=MODEL NUMBER;          

REMARK 470 RES=RESIDUE NAME; C=CHAIN IDENTIFIER; SSEQ=SEQUENCE NUMBER;         

REMARK 470 I=INSERTION CODE):                                                  

REMARK 470   M RES CSSEQI  ATOMS                                               

REMARK 470       A B   8    N9   C8   N7   C5   C6   N6   N1                   

REMARK 470       A B   8    C2   N3   C4                                       

REMARK 470       A B   9    C5'  C4'  O4'  C3'  O3'  C2'  O2'                  

REMARK 470       A B   9    C1'  N9   C8   N7   C5   C6   N6                   

REMARK 470       A B   9    N1   C2   N3   C4                                  

ATOM   6537  P     A B   8     -34.932  11.724  41.002  1.00 77.78           P 

ATOM   6538  OP1   A B   8     -35.604  10.974  42.094  1.00 77.35           O 

ATOM   6539  OP2   A B   8     -33.719  12.526  41.308  1.00 77.37           O 

ATOM   6540  O5'   A B   8     -34.568  10.702  39.831  1.00 76.10           O 

ATOM   6541  C5'   A B   8     -33.247  10.198  39.673  1.00 75.99           C 

ATOM   6542  C4'   A B   8     -32.692  10.535  38.309  1.00 77.26           C 

ATOM   6543  O4'   A B   8     -33.130  11.875  37.967  1.00 78.06           O 

ATOM   6544  C3'   A B   8     -31.168  10.587  38.180  0.25 78.25           C 

ATOM   6545  O3'   A B   8     -30.558   9.342  37.861  1.00 79.23           O 

ATOM   6546  C2'   A B   8     -30.955  11.611  37.072  1.00 78.19           C 

ATOM   6547  O2'   A B   8     -31.153  11.016  35.797  1.00 77.91           O 

ATOM   6548  C1'   A B   8     -32.097  12.584  37.326  1.00 78.36           C 

ATOM   6549  P     A B   9     -30.915   7.974  38.628  1.00121.99           P 

ATOM   6550  OP1   A B   9     -32.069   7.374  37.921  1.00121.50           O 

ATOM   6551  OP2   A B   9     -30.963   8.222  40.084  1.00121.86           O 

ATOM   6552  O5'   A B   9     -29.637   7.051  38.359  1.00 86.84           O 

These residues are deemed NOT to be RNA by Jmol (14.0.13, 14.1.16)

load =4olb

restrict not protein

selectionhalos on

select rna

but they are deemed to be "nucleic"

select nucleic

Thus they are nucleic but neither RNA, DNA, nor ligand.

Is this a bug, e.g. undesirable behavior?

Compare with 1bdx which contains 64 DNA phosphorus atoms with no other atoms in these residues.

ATOM      1  P    DG J   2      37.837  31.943 -10.756  1.00 30.00           P 

ATOM      2  P    DC J   3      43.246  28.658  -8.793  1.00 30.00           P 

ATOM      3  P    DA J   4      46.879  26.096  -3.879  1.00 30.00           P 

ATOM      4  P    DT J   5      47.286  23.980   2.386  1.00 30.00           P 

ATOM      5  P    DG J   6      44.265  21.858   7.884  1.00 30.00           P 

Jmol deems these sidechain-less, pentose-less, oxygen-less P atoms to be both DNA and nucleic.

load =1bdx

selectionhalos on

select dna

select nucleic

Eric

------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
Jmol-users mailing list
Jmol-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jmol-users