The Rainey Lab

Documentation Contents
A summary of script options Some notes on STR files Some notes on PDB files Some notes on SDB files Contacts References Return to script interface.

Summary of script options
Output bargraph of secondary shifts
Outputs a customizable bargraph of secondary chemical shifts, as calculated by CS Chemeleon. Note that, for the sake of clarity, individual atom types and individual random coil databases are given separate graphs. Bargraphs are produced in GIF format.
Output plaintext secondary shifts
Outputs a text file of secondary shifts calculated from selected random coil databases, and the supplied STR file. Each random coil database is output to separate sections of the same file. Columns are delimited by tab space.
Output comparison of secondary structure from shifts with that from dihedral angles
Calculates the apparent average secondary structure from both comparison of secondary shifts with supplied threshold values, and comparison of phi and psi angles to expected structure in a Ramachandran plot-style analysis. The output text file is compatable as input for a script supplying overall percent agreement, and percent agreement on a per-amino acid basis. This script may be found here.
Output plaintext torsion angles
Outputs a text file of phi and psi torsion angles calculated from PDB cartesian coordinates. Each model in the PDB ensemble is treated separately. Columns are delimited by tab space.
Upload own random coil chemical shift file
Allows the user to supply his or her own table of random coil chemical shifts, in the form of an SDB file, for calculation of secondary chemical shifts. See notes on SDB files for format information.
Correct random coil shifts from this data set for primary sequence
f selected, secondary shifts produced with the relevant random coil database will be corrected for deviations due to primary sequence based on work by Schwarzinger et al.
Consider only solvent-exposed or transmembrane segments
Allows the user to limit comparison between the structure determined by secondary shifts and that from torsion angles to transmembrane or solvent-exposed residues. This analysis requires a non-standard PDB file, modified to contain an indicator off residue location ("AQ" or "TM") in columns 95 through 98 of the PDB. A script has been produced to create these modified PDB files from data found in the PDB TM in the form of PDBTM XML files. This script may be found here.
Average PDB structure according to local consensus
If selected, the secondary structure produced by the PDB file will be modified based on the average structure of residues nearby in primary sequence. The sequence will be scanned iteratively, considering floor(n/2) residues on either side (where n is defined by the Length of frame (residues) option). Should the fraction of residues within this frame be greater-than or equal-to the supplied minimum fraction (Minimum frame agreement for structural averaging), the residue will be considered to feature this type of secondary structure for all analyses. Note that this averaging is NOT recursive. That is, only the original sequence will be considered in determining average local structure, and all changes to sequence are made simultaneously after this calculation has completed.
Minimum agreement amongst ensemble members for defined secondary structure
This is the minimum fraction of models which must demonstrate a specific secondary structure at a given residue for that residue to be considered as having said structure.
Shifts to consider in analysis
The atom types to consider in structure calculations. Note that a given PDB and STR file must contain data for a minimum of n - 1 atom types for any comparison to be made.
Shift significance thresholds
The threshold secondary shift values for a residue to be deemed significant for alpha helix or beta sheet structure. alpha helix thresholds are given first, followed by beta sheet thresholds in the form of x, y. Note that the traits of different atom types are native to the script: HA/CB shifts are significant for alpha helix structure when less than the helix threshold, and significant for beta sheet when greater than the sheet threshold. HB/CA demonstrate the opposite trend.
Use own legends for bargraph
Allows the user to supply his or her own legend for each random coil database, for use during bargraph production.
Scale of graph
Allows the user to determine where grid lines are placed on the bargraph (1/10 will give 1 line per 10 ppm).
Height of graphical output
Allows the user to set the height of the bargraph image, in pixels. Width is determined by the number of secondary shifts calculated
Bar colour
The colour of the bars on the bargraph, for the first random coil database. In RGB format: (R, G, B) where R, G, and B are integer values between 0 and 256.
Iteration for subsequent SDB
Each successive random coil database may have some value added to the original RGB value of the bar colours supplied for the first random coil database.
Background colour
The colour of the bargraph background. Accepts standard colour names, or RGB format.
Text colour
The colour of the bargraph text. Accepts standard colour names, or RGB format.
Line colour
The colour of the scale lines. Accepts standard colour names, or RGB format.
Threshold colour
The colour of the lines at significance threshold positions. Accepts standard colour names, or RGB format.

Notes on STR Files
Assigned chemical shift lists: To be functional, provided STR files must contain an "Assigned chemical shift lists" section, with appropriate header. Further, STR files must contain column identification as outlined in the STR file documentation. CS Chemeleon is tolerant of unassigned chemical shifts. However, provided information must contain: - Author-assigned sequence codes - Residue labels following 3-letter amino acid code conventions - Atom name - Atom type - Chemical shift Chemical shift referencing: Random coil shift databases provided are referenced to DSS, and this will be assumed of any user-supplied tables. To this end, some attempt has been made to allow re-refrencing to TMS from several common reference compounds. CS Chemeleon is tolerant of the following reference compounds, provided a "Chemical shift referencing" section is supplied in the STR file: - Hydrogen shifts may be referenced to: TSP, TMS, or water. - Carbon shifts may be referenced to: TSP, TMS, or Dioxane. If no Chemical shift referencing section is provided, the shifts are assumed to be referenced to DSS. Other issues: - Note that non-standard amino acids are not supported at this time, save hydroxyproline. - Cysteine residues are assumed to be reduced, and proline residues are assumed to be trans. - At this time, to ensure proper 1:1 mapping of PDB and STR files, the supplied STR file must contain only residues present in the supplied PDB file.

Notes on STR Files

Assigned chemical shift lists:
     To be functional, provided STR files must contain an "Assigned chemical shift lists" section, with appropriate header. Further, STR files must contain column identification as outlined in the STR file documentation. CS Chemeleon is tolerant of unassigned chemical shifts. However, provided information must contain:
          - Author-assigned sequence codes
          - Residue labels following 3-letter amino acid code conventions
          - Atom name
          - Atom type
          - Chemical shift

Chemical shift referencing:
     Random coil shift databases provided are referenced to DSS, and this will be assumed of any user-supplied tables. To this end, some attempt has been made to allow re-refrencing to TMS from several common reference compounds. CS Chemeleon is tolerant of the following reference compounds, provided a "Chemical shift referencing" section is supplied in the STR file:
          - Hydrogen shifts may be referenced to: TSP, TMS, or water.
          - Carbon shifts may be referenced to: TSP, TMS, or Dioxane.
     If no Chemical shift referencing section is provided, the shifts are assumed to be referenced to DSS.

Other issues:
     - Note that non-standard amino acids are not supported at this time, save hydroxyproline.
     - Cysteine residues are assumed to be reduced, and proline residues are assumed to be trans.
     - At this time, to ensure proper 1:1 mapping of PDB and STR files, the supplied STR file must contain only residues present in the supplied PDB file.

Notes on PDB Files
PDB files should be formatted according to conventions for PDB files, and must contain: - Standard 3 letter amino acid code - Atom type - Residue ID. This does not necessairly need to match the residue numbering scheme found in the supplied STR file. - x, y, and z Cartesian coordinates. Note that the supplied PDB file MUST contain all residues found in the supplied STR file.

Notes on SDB Files
SDB files should contain random coil shifts for all atoms in all 20 standard amino acids. Each atom type must be denoted as follows: - On a new line, type #, followed by atom type. Atom types must be named as follows: HA for alpha hydrogen, HB for beta hydrogen, CA for alpha carbon, CB for beta carbon. - Shifts for each amino acid should follow, with each being given its own line prefixed by the standard one letter code. Shifts should follow, enclosed in square brackets ("[ ]"), with diastereotopic shifts listed from most deshielded to most shielded, and delimited by commas. - Lines prefixed with ! will be ignored as comments. - Order of atom types, or amino acids within atom type headings, is irrelevant. - If no shift is available, or the particular amino acid lacks that atom type, leave the square brackets empty: e.g.: # HB G [] A [1.43] L [1.67, 1.67] ! A comment ...etc. - Note that trans proline should be written as Ptrans, cis proline as Pcis. - Reduced cysteine should be written as Cred, oxidized as Cox.

Notes on SDB Files

SDB files should contain random coil shifts for all atoms in all 20 standard amino acids. Each atom type must be denoted as follows:
- On a new line, type #, followed by atom type. Atom types must be named as follows: HA for alpha hydrogen, HB for beta hydrogen, CA for alpha carbon, CB for beta carbon.
- Shifts for each amino acid should follow, with each being given its own line prefixed by the standard one letter code. Shifts should follow, enclosed in square brackets ("[ ]"), with diastereotopic shifts listed from most deshielded to most shielded, and delimited by commas.
- Lines prefixed with ! will be ignored as comments.
- Order of atom types, or amino acids within atom type headings, is irrelevant.
- If no shift is available, or the particular amino acid lacks that atom type, leave the square brackets empty:
e.g.:

# HB
	G	[]
	A	[1.43]
	L	[1.67, 1.67]
! A comment

...etc.
- Note that trans proline should be written as Ptrans, cis proline as Pcis.
- Reduced cysteine should be written as Cred, oxidized as Cox.

Contact information
Questions or concerns regarding CS Chemeleon may be directed to the following individuals, via e-mail: Aaron Banks Marie Tremblay Dr. Jan K. Rainey

References
[[[ References go here ]]]