Main-chain/Side-chain Database commands
Use a protein database to model a structure. O can access both a main chain database (Jones et al, 1992; Jones and Thirup,1986) and side-chain rotamer database (Kleywegt & Jones, 1998).The data base files are loaded from the public area when first used in the session. They go into the user's O database but are not written back when the database is saved because they have a status of 'R' (readonly).
The main chain database consists of 32 protein structures that have been refined at high resolution. The side chain rotamers are also based on an analysis made on high resolution structures.
Lego_setup <Ca_trace>
<colour_ramp>
<main_chain_database>
<database_directory>
<rotamer_database>
Define necessary parameters for the menu. This allows the user to select an option to show or not show Ca traces, the colour codes to be used to display them, a file with the definition of the precalculated Ca diagonal distance matrix, the directory where this database is kept and the file with the side chain rotamer database. Most of the values will be correct by default, but it is necessary to define these parameters the first time the Proleg commands are used with a given database file.
Lego_Ca <id two atoms> or <mol res1 res2>
Look through the fragment database to find similar pieces of structure. A dial becomes activated to pass through the top 20 best fits. The rms fit, protein and sequences are shown on the prompt window. The user is also able to specify any particular protein in the data base. Use Yes / No to pick the best matching main chain fit to one's own Ca atoms. Only the main chain atoms are extracted from the database. The coordinates of the first and last residues in the user's zone are not affected
lego_side_chain <id an atom> or <mol res atom>
Pick a side chain rotamer for a specified residue from the library of preferred side chain conformations. A residue is specified by identifying any atom in the residue. A dial becomes activated to change the rotamer. This command requires the coordinates for 'N', 'CA', 'C' atoms for each residue. Use Yes to choose the rotamer you want, No to end without changing the coordinates.
lego_loop <id two atoms> or <mol res1 res2>
Make a Ca data base search for a zone but only use atoms that have been selected with select_on or select_off After accepting coordinates, the selection datablock is not changed so sometime remember to use select_on again. Otherwise, when you eventually write out the coordinates you might not get what you expected.
Example, insert an extra residue into a molecule called a28 after residue 100. Use the databases to construct main and side chain coordinates
mutate_insert a28 100 100a leu ;
sel_on a28 ;
sel_off a28 100 101
leg_loop a28 98 103
yes
sel_on a28 ;
leg_side a28 100
yes
The loop chosen is the best fit to the CA coordinates of residues 98 99 102 and 103. The new leucie residue has the most common rotamer conformation.
lego_auto_mc <id two atoms> or <mol res1 res2>
Automatically build the main chain for a zone of atoms, using the database. This option goes along the chain searching for the best fitting pentapeptide, keeping the main chain atoms of the middle 3 residues and overlapping with the next pentapeptide. This means that the first and last residues are not changed. The option requires coordinates for the Ca atom of each residue.
lego_auto_sc <id two atoms> or <mol res1 res2>
Move along the chain and automatically build side chain by choosing the most frequent rotamer from the database. This requires the coordinates for 'N', 'CA', 'C' atoms for each residue. Useful for putting on starting side chains that may then be checked individually or with RSR commands. Because ALA and GLY residues have no rotamers, they will cause a warning that can be ignored.
lego_bones <id>i
This command is meant for use with skeleton Bones atoms but is not enforced by the program. Identify atoms to be taken as guide Ca's. End by specifying Yes. Use Yes / No to pick best fit to Id'ed (Ca equivalent) atoms. We do not recommend using this option anymore, rather we recommend using the Bones commands to make a Ca trace which can then be built automatically.