Methods, Acknowledgements and References
Atomic Coordinate Files
- The 19 base-pair DNA segment in 1D66.PDB was chosen because the helix is nearly straight.
All hydrogens are missing since the atomic coordinates in 1D66.PDB were determined by X-ray diffraction (at 2.7 Å resolution).
- The protein and water molecules were deleted from the PDB file, speeding up the rotation frames/second (by about 6-fold in MDL Chime).
- The end base pairs were remarked-out so as to produce 3' hydroxyl's and 5' phosphates on the ends, leaving 17 base pairs in 1D66-PWZ.PDB.
- As an alternative model for the Jmol version of this tutorial, the DNA.PDB file was prepared by editing 1d66-pwz.pdb:
- C28 base was flipped around the glycosididc bond into correct Watson-Crick pairing with G11, using Accelrys DS Visualizer. This fixes the
erroneous orientation contained in the 1d66.pdb file (which is the basis for question #17 in the question set).
- The hydrogen on terminal 3'OH and the third oxygen on terminal 5'P were relocated using Accelrys DS Visualizer for a proper distance and orientation, and renamed to belong to the proper residue.
- The AT and GC base pairs are residues A27 T12 and C33 G6 taken from 1D66.PDB.
- The free program WebLab (which like Chime is also derived in part from RasMol's source code) from Molecular Simulations Inc. was used to extract each base pair into a separate atomic coordinate file. WebLab allows atom-by-atom mouse picking as well as selecting by named residue.
- For the MDL Chime version (version numbers up to 3.0):
- Each base pair was saved in a MOL format file because this, unlike WebLab's PDB-saving, preserved double bond information.
- Some manual correction of bond orders was required.
- WebLab was then used to add hydrogen atoms (a few inappropriately added were removed).
- In the case of the GC, the hydrogens involved in hydrogen bonding were not well oriented towards their bonded oxygens/nitrogens, so WebLab's bond torsion capability was used to position these hydrogens in a more favorable orientation.
- Both real (from H to N/O) and simplified hydrogen bonds (from N to N/O) were added manually to the bond list.
- For the Jmol version (version numbers from 3.0.1 upwards), AT and GC pairs were re-edited:
- Hydrogens were added using Accelrys DS Viewer (the successor to WebLab).
- Hydrogens involved in base pairing were moved manually in Accelrys DS Visualizer, so that they were aligned to N and O.
- Files were saved in PDB format, rather than MOL as before.
- Double bonds are generated via script (dna_at-2.spt) using Jmol "connect" commands; positions of double bonds are those automatically determined by Accelrys DS Visualizer. Then, double bond rendering is activated with "set bonds on"
(Another approach would be to use theoretically-generated atomic coordinate files. These can be generated for any specified sequence, in A- or B-form DNA or A-form RNA, at Pittsburgh Supercomputer Center Web Tools. Although Eric Martz did not know about this resource in 1996 when the nonlinear tutorial on DNA was implemented, he likes the "real world" irregularities in the empirical 1D66 structure.)
Thanks to MDLI (now Elsevier MDL) for donating Chime to the world.
Thanks to Tim Maffett at MDLI for making script-enhanced Chime available during its development.
Thanks to Roger Sayle who pioneered the molecular structure rendering methods in RasMol, and whose source code was used in the development of Chime.
Thanks to the Jmol Development Team for creating Jmol, for offering it under open source and for their altruistic work and continuous improvements.
Contents and Design
Prof. Anne Simon suggested the page on ends and antiparallelism.
Dr. Frieda Reichsman (moleculesinmotion.com) made several good suggestions adopted in version 2.8, and caught several bugs.
The unique "MoveTo" utility server for Chime by Todd A. Carlson and Clark Wells was invaluable in scripting the zooms from the double helix to single base pairs, and back.
Jmol version was built upon the scripts and internal documentation authored by Prof. Martz in the Chime version.
Both Dr. Frieda Reichsman, Prof. Eric Martz and Dr. Gonzalo Claros (Univ. of Málaga in Spain) tested the Jmol versions, caught bugs and suggested improvements.
Dr. Anne B. Hodgson (University of York, UK) suggested improvements in the strands and helical backbone page, adopted in version 4.2.
- Original English version by Prof. Eric Martz (University of Massachusetts, Amherst MA, U.S.A.).
- Updates in English for Jmol version by Dr. Angel Herráez (Universidad de Alcalá, Spain).
- Spanish by Dr. Angel Herráez.
- Brazilian Portuguese by Prof. Sérgio Ceroni da Silva (Universidade Federal do Rio Grande do Sul, Brazil).
- German by Hendrik Buddensiek (student at the Westphalian-Wilhelms University in
Muenster, Germany) and Dr. Peter Wehling (Julius Kuehn Institute, Quedlinburg, Germany).
- French by Dr. Marta Herráez (Universidad de Valladolid, Spain), Martin Moebs, Elisa Alonso Martín and Jean Pierre Chouzenoux.
The structure 1D66.PDB [PDB entry], from which the DNA coordinates were excerpted, was determined by:
R. Marmorstein, M. Carey, M. Ptashne, and S.C. Harrison; DNA recognition by Gal4: Structure of a protein/DNA complex. Nature 356:408, 1992 [Medline entry].
Original methods page for the MDL Chime version of this tutorial, maintained by Eric Martz.
OpenRasMol (freeware, GNU GPL)
MDL Chime (freeware)
Jmol (freeware, open source, GNU LGPL)
WebLab Viewer Lite, by Molecular Simulations Inc. -- discontinued; substituted by DS Visualizer
Accelrys Discovery Studio Visualizer (freeware)