STING
Features
STING allows one to load a PDB file
(molecular structure) and to
receive information about underlying molecular sequences. This is a KEY feature
of the STING PDB Viewer: relationship between linear
sequence and 3D position for the residues/nucleotides.
|
PC ONLY features are written in this way: within yellow box!
|
Unique STING Features
Instant Display of Residue/Nucleotide Number Within Sequence
Linear Sequence to 3D Fold STING Link
Database Linking
Gaps in Sequence Clearly Indicated
Chains Separated and Displayed
Secondary structure elements identified
Color Coded Residues:
Hydrophobicity/Charge
STINGpaint: WWW tool for sequence
and MSA coloring
|
STING's Control Panel Commands
Clear All
Wireframe
Ribbon
Color CPK
Color by Chain
Color by Structure
HOH on
HOH off
Ligand on
Ligand off
Ligand Pocket
HOH + Ligand
Charged Residues
Interface on
HOH + Interface
Interface: 1st half
Interface: 2nd half
HOH+ 1st half
HOH+ 2nd half
Sequence color coding
Sequence Frame brings linear protein sequence color coded
with respect to Hydrophobicity and charge groups!
- Residues: AVLIMFP are colored gray! [small & hydrophobic]
- Residues: STYNQWG are colored green [polar]
- Residues: D E are colored red! [negatively charged]
- Residues: R K are colored blue! [positively charged]
- Residues: C is colored yellow! [disulfide bridge forming]
Nucleotide sequence is also color coded:
- Nucleotides: T G are colored green [purins]
- Nucleotide: C are colored yellow [pyrimidine]
- Nucleotide: A are colored gray [pyrimidine]
Sequence Frame also shows the numbering of residues in the
sequence, gaps in the PDB sequence, chain identifier and secondary structure elements
identifier. Each residue in the Sequence Frame
is "clickable", resulting in a CPK presentation of its position in 3D in the
Graphics Frame. Blue and Red lines below the sequence
are also "clickable" resulting in a graphical
RIBBON presentation of the specified sequence region!
STINGpaint
STINGpaint was developed to allow the presentation of residue characteristics in the Sequence Frame.
As a consequence, during development of STING project,
we have slightly expanded on STINGpaint idea and adopted it for use
with Multiple Sequence Alignment (MSA) coloring. It turns out that this tool
was very interesting for people wanting to easily grasp specifically colored regions
along the MSA. In addition, our STINGpaint is also a part of our ongoing
work for STING-2, a package that will be able to show both sequnce alignmnts
(in the Sequence Frame) and structures (in the Graphics Frame)
for respective sequences!
STINGpaint now supports following sequence and MSA formats:
- Coloring sequence of any PDB entry
- Coloring any sequence presented to STINGpaint in FASTA format
- Coloring MSA in PRISM (An-Suei Yang) output format
PRISM is a sequence/structure /threading/homology modeling program
developed byAn-Suei Yang in the Honig laboratory
- Coloring MSA in PSI-BLAST output format
- Coloring MSA in GCG output format
For test purposes, a user can access sample formats
(given for each option at STINGpaint)
and just copy_and_paste them
into the working area of STINGpaint. It is our experience that
analysis of MSA can be greatly enhanced using STINGpaint!
Note: We strongly suggest that user tries:
Using STINGpaint with different background color
and experiment with the size of the MSA to be STINGpainted!
It is our experience that only CPU with 400MHz have
acceptable speed for large MSA files, intended for STINGpaint with
background color option! However, results might be very informative
for the MSA analysis! Related to this issue,
see work by
W.R. Taylor (Protein Eng 1997 Jul;10(7):743-746;
"Residual colours: a proposal for aminochromography").
STINGpaint output exemple:
1 50
WRP25 SGPWSWCDPA TGYQVSALTG CRAMVKLQCV KSQVPEAVLR DCCQQLADIN
WRP26 SGPWMWCDPA TGYQVSALTG CRAMVKLQCV GSQVPEAVLR DCCQQLADIN
WRP24 SGPWMWCYPG QAFQVPALPA CRPLLRLQCN GCQVPEAVLR DCCQQLAHIS
WRP27 SGPWMWCDPA MGHRVRPLMG CRAMVKLQCV GNQVPEAIQR DCCQELANIT
AI1FAT ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~
AI2FAT ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~
51 100
WRP25 NEWCRCGDLS SMLRSVYQEL GVREGKVLPG CRKEVMKLTA ASVPEVCKVP
WRP26 NEWCRCGDLS SSLRSVYQEL GVREGKVLPG CRKEVMKLTA ASVPEVCKVP
WRP24 NEWCRCG~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~
WRP27 NNWCRCHDLG SMLNSVYQEL GAREGTVFPG CRKEVMKLTV ASVPAVCKVP
AI1FAT ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~
AI2FAT ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~ ~~~~~~~~~~
Residue/nucleotide
sequence number
One of the KEY features of STING
is the facility of deciphering the residue sequence number: It is meant to
aid users in sequence analysis with respect to 3D positioning/fold! User can
slide mouse over the sequence and get on SINTG's Status frame instant
identification of the sequence number and single letter code.
Warning: residue number is separated by ":" from the chain identifier. Chain
identitier can be either letter or number or first letter of any 3-letter code used in
PDB fiel as chain identifier! For more details, user should consult PDB file formats!
Linear to 3D Link
As mentioned earlier, the 3D Graphics Frame and Sequence Frame
are interconnected so that user can have control of
both types of information (linear and 3D);
|
User can search the Graphics Frame and ask
questions such as: which residue
is at this particular spot pointed to by the mouse within the 3D frame?
|
The user can place the mouse
over the single letter code in the sequence
and ask question such as: where in the 3D fold is
this residue placed?
Secondary structure regions
Similarly, the user can slide the mouse over the secondary structure colored bars
Helices (red lines below the
sequence) and Extended Sheets
(blue lines below sequence) and see on STING's Status Frame
sequence region covered by this element of secondary structure.
Sequence Gaps in PDB file
Gaps are clearly indicated in the Sequence Frame by use of "-----"
in place of residue/nucleotide single letter code,
for indication of residue missing at that position. This is also one of the
key features of
STING presentation.
Sequence Chains in PDB file
Chains are also clearly indicated in the Sequence Frame so that
the user can have access to
residues separated by distinct chain identifier! This key feature is very
handy, once a user would like to examine an interface between two protein chains; digital access
to residues (belonging to different chains) can then produce graphical CPK positions of critical
residues in the protein fold.
Warning: residue number is separated by ":" from the chaine identifier. Chain
identifier can be either letter or number, or first letter of wany 3-letter code used in
PDB file as chain identifier! For more details, user should consult PDB file formats!
Chain identification in STING's PDB Info Frame follows alphanumerical order;
However, appearence of the chains in the PDB file is NOT always following such order! This is
very important to recognize, specially in case of "Interface ON" STING script, where
first two chains encountered in PDB file are taken for interface building {next version of
STING will have option for chain choice}!
Mouse
marked contiguous sequence regions: CPK displayed in Graphics Frame
One of the key features of the STING
presentation is one stroke (mouse slide over contiguous sequence region)
action to get a series of residues (within contiguous stretch of the sequence)
presented in CPK within the Graphics Frame. This key feature
makes it possible to see 3D fold positioning of the sequence profiles
such as those indicated in the BLOCKS
database (see also the tutorial on this subject:
STING: Enzyme/Inhibitor Interface and Structural Waters ). |
STING's Control Frame/Panel Commands
Note: Some of the commands presented here are simple copy of the standard CHIME commands.
We have chosen to put them in the Control Frame because of easier access.
In STING _SGI version, commands as "Ligand on"
are only possible by using CHIME scripting.
The same command in the PC CHIME version 2.0beta2 32bit, however, is actual
command (available from the pool-down menu) and not a script.
Most usable STING commands are actual scripts, made to
facilitate molecular structure and macromolecular interface analysis. We found
"Interface on", "Ligand Pocket", "HOH+Interface",
"HOH+ligand", Interface: 1st/2nd half and "Charged Residues" very conveniently,
one stroke apart from viewing on the Graphics Frame.
Clear All
This is the only way to refresh the graphics frame (if starting
from scratch is desired).
Wireframe
If your graphics screen becomes too cluttered, use Clear All and then restart by using
the Wireframe command. You will end up with (you guessed right) wireframe representation
of the molecule.
Note: if previously you have used "Color by chain", these
color codes are retained, which we found useful once you enter into chain of analysis
procedures. If you really desire to start with original CPK colors, use option "Color CPK"
in addition to "Clear All" + "Wireframe"!
Ribbon
Ribbon simply turns on ribbon around protein main chain backbone.
Color CPK
This option is made available for the single purpose of convenience: easy, one
stroke action to obtain, for example, Interface built in earlier analysis, color coded CPK (instead of
color coded with respect to chain identifier, e.g.)!
Color by Chain
Convenient coloring by Chain also gives you opportunity to see if any chain is actually
broken by introduction of the gap in the sequence (this info, however, could be easily
observed by looking at the sequence frame.)
Color by Structure
This command colours the molecule by protein secondary structure.
Alpha helices are coloured magenta, [240,0,128], beta sheets are coloured
yellow, [255,255,0], turns are coloured pale blue, [96,128,255]
and all other residues are coloured white. The secondary structure is either
read from the PDB file (HELIX and SHEET
records), if available, or determined using Kabsch and Sander's DSSP algorithm.
Note:regions of protein fold colored by this command DO NOT
NECESSARILY coincide with Secondary Structure identifiers (red and blue lines
within the Graphics Frame). Difference originates in calculated versus
indicated nature of two approaches, respectively.
HOH on
STING's default graphics frame ,
comes with crystal waters turned on. If present, these water molecules might make further analysis
difficult, as they might clog a picture. In this case, use command "HOH off".
Later in an analysis, one might wish to turn them on, for examination of their presence within
interface layer, for example. Once you use "HOH on" button, water molecules will change
from default red color to magenta! This is done in order to facilitate visual identification
of HOH molecules, especially in cases when default HOH color, red, is also used by other chain or
ligand!
HOH off
STING's default graphics frame ,
comes with crystal waters turned on. If present in large number, waters should be removed from
the visual using this command!
Ligand on
This command will turn visual presentation of any present ligands on.
Ligand off
Somewhat surprisingly, this command, really turns ligand visual off. :)
Ligand Pocket
"Ligand Pocket" will erase anything else from the graphics frame but ligand and residues side_chain
atoms in contact with it! Contact is defined by distance, set to 4.0 Angstrems, and measured
between ligand atoms and any other atoms belonging to ligand surrounding chains!
This feature is very convenient, once you turn on the wireframe
display of the rest of the molecule, by using "Wireframe" command.
Position of the Ligand and surrounding molecules will be very clearly displayed for further
analysis. Combine this with Ligand on/off, Color by chain and get better insight
into ligand 3D environment!
Also, very much used in combination with HOH + Ligand!
Note: distance of 4.0 Angstroms will pretty much satisfy requirements for both
hydrogen bond formation, as well as for hydrophobic interactions!
HOH + Ligand
This command will have visualized Ligand and water molecules in contact with it. It is often
necessary to have this information on contacting water molecules around the ligand for
proper H-bond counting. Some structural water molecules are identified easily in this way!
Contact between ligand and HOH molecules is defined here by distance, set to 3.3 Angstrems,
and measured between ligand atoms and any other HOH molecule
(actually, in most cases, an Oxygen atom).
Note: distance of 3.3 Angstroms is considered maximum distance between Hydrogen donor
and acceptor, that could still bring about hydrogen bond formation.
Charged Residues
This command will visualize all charged residues using van der Waals dotted surface.
All lysines and arginines will be blue color coded, aspartic acid and glutamic acid will be red and
histidine will be color coded cyan. This is a very useful feature, once you would like to know
charge distribution in vicinity of ligand or maybe at molecular interface!
Note: this command will visualize (turn on) all charged residues, irrespective to the chain
identifier. We found this convenient in most cases, as user can get CPK presentation of
charged residues in the chain of interest, and dot color-code of charge residues in contact
with interface in focus (and not belonging to the same chain) - therefore having glimpse of
complementarity)!
Interface on
This is one of the most used commands in STING .
What we like about it is the simplicity of getting general information on interfaces between two
molecular chains, all in one mouse stroke. This
command will turn on only atoms at an INTERFACE of the first two chains in a PDB file.
As of now, this is a hard-wired command; in other words, you will be able to see most interfaces
for PDB files having chains A and B, H and L, E and I (which is the most frequently observed case).
We are going to implement choice of chains for which user desires to see Interface,
in the next version of STING.
In combination with command "Color by chain" (issued prior to Interface on)
graphical information is even more emphasized.
Note:Interface is defined based on a distance, set to 8.0 Angstroms,
and measured between any two atoms in different chains! A value of 8.0 Angstroms was
chosen empirically; we first tried a distance of two times 3.3 Angstroms (Hydrogen aceptor from
one chain, to water molecule, to hydrogen donor on the other chain). This would be distance of 6.6
Angstroms, but we found that graphical presentation of the interfaces is much more "complete" if
distance of 8.0 Angstroms is used. We judged completeness by how well an interface is populated by atoms,
or how many holes we have on the chaines interface.
Obviously, the user should consider Interafce graphics presentation more as a guide,
than as a exact interface definition.
Warning:
Chain identification in STING's PDB Info Frame follows alphanumerical order;
However, appearence of the chains in the PDB file is NOT always following such order! This is
a very important to recognize, especially in case of "Interface ON" STING script, where
first two chains encountered in a PDB file are taken for interface building {next version of
STING will have option for chain choice}!
|
For the latter, we are already implementing exaact Interface definition,
based on Buried surface area upon complex formation, in our package:
HORNET.
|
HOH + Interface
This command is very useful for analysis of the interfaces and water
molecules captured between Interface Forming Residues (IFR). Availability
of such quick identification of these waters may aid in identification of
indirect H-bond formation between two chains (with involvement of structural
water molecules).
Note:HOH molecules visualized by this command are identified here by
a distance, set to 3.3 Angstrems, and measured between a subset of atoms belonging
to the interface from one chain, to any HOH molecule (actually, in most cases,
an Oxygen atom); This is then done for the other chain (its IFR and HOH molecules
at defined distance of 3.3 Angstroms).
Finaly presented waters are actually intersection of water molecule ensambles
defined above! In other words, the only water molecules presented (color coded
magenta), are those that satisfy the geometric condition of being 3.3 Angstrom
(maximum) distance from both chains! These HOH molecules are likely to make
H-bonds with both chains, contributing effectively to the energy of binding!
Note: distance of 3.3 Angstroms is considered here as the maximum distance
between a Hydrogen donor and acceptor that still defines a hydrogen bond.
These water molecules are then easily observable if you use option: "Interface:
1st half" and "Interface: 2nd half". One can actually analyse only one chain
IFR, with HOH molecules which are likely to make H-bonds with both
chains!
Interface: 1st half
This particular command will allow the user to observe only one of the
two chains forming a macromolecular interface. This option allows the user
to examine in detail only half of a complementary surface (see example in
tutorial section!). Again, this is kind of hard-wired scripting (since we
do not have chain independent command, as we do not have it for "Interface
on" button), and so, this feature will not work for chain names other than
the first two ones encountered in PDB file. But hang on, we are working on
it.
As explained above for "Interface on", using the "Interface: 1st half" button
one can actually analyse only one chain IFR, with HOH molecules which
are likely to make H-bonds with both chains!
Interface: 2nd half
Same as command "Interface: 1st half", but obviously tuned for visualization
of the sexond half of the complementary surfaces. Sorry, for the hard wired
commands! We are working on making this command chain_name_independent! See
tutorial for the real power of these two last commands!
Note:As explained above for "Interface on", with "Interface: 2nd half"
button one can actually analyse only one chain IFR, with HOH molecules
which are likely to make H-bonds with both chains!
HOH+ 1st half
This command will conveniently display only one (first) part of the facing
surfaces at molecular interface, in addition to water molecules which are
3.3 Angstroms distant from any of IFR of that chain. Difference between
this command and "Interface: 1st half" /"Interface: 2nd half" is that the
former one will generally show many more HOH molecules than the latter ones.
This is due to the more restrictive condition imposed for the latter commands,
with respect to which water molecules will be shown. Namely, "HOH+ 1st half"
(and so the "HOH+ 2nd half") will show one chain IFR and all HOH at
3.3 Angstroms from it! On the other hand, "Interface: 1st half" and "Interface:
2nd half" will only show those HOH molecules which are 3.3 Angstroms distant
from BOTH chains, which is a much more restrictive condition!
User can easily grasp the difference between two HOH molecule ansambles and
conceptualise the importance of the difference!
HOH+ 2nd half
Same as "HOH+ 1st half", but obviously tuned for visualization of another
half of the complementary surface.
Special feature is a quality of compiled
information EASILY obtainable by combination of above features!
(see tutorial section for some illuminating examples)
|