chainsaw xyzin foo.pdb alignin foo_ali.ali xyzout foo_out.pdb
Chainsaw is a Molecular Replacement utility which takes an alignment between target and model sequences and modifies the model pdb file by pruning non-conserved residues. The pruning can be done in one of three ways: back to the gamma atom, back to the beta atom, or in such a way as to retain all atoms common to the target and model residues. The names of the retained atoms will be changed if necessary to match the target. Conserved residues are left unchanged.
The residues in the output pdb file will be numbered in a way consistent with the target sequence i.e. if a residue in the target corresponds to a gap in the model, the residue numbers in the output pdb file will contain a gap, but if a residue in the model corresponds to a gap in the target, the residue numbers in the output pdb file will be consecutive.
If there are alternate conformations in the input pdb file, chainsaw will choose the most probable conformation, and assign it an occupancy of 1 in the output pdb file.
Chainsaw accepts several alignment file formats, which it identifies by the file extension: PIR (.pir), Fasta (.fas), Clustal (.aln), MSF/GCG (.msf), interleaved Phylip (.phy), BLAST (.bla) and OCA (.oca). Suggestions for additional formats are welcome.
A .bla file consists of output from the BLAST server:
Score = 164 bits (416), Expect = 6e-42, Method: Composition-based stats. Identities = 85/120 (70%), Positives = 101/120 (84%), Gaps = 0/120 (0%) Query 4 LVGVIMGSTSDWETMKYACDILDELNIPYEKKVVSAHRTPDYMFEYAETARERGLKVIIA 63 +VG+IMGS SDWETM++A +L EL IP+E +VSAHRTPD + +YA TA ERGL VIIA Sbjct 23 VVGIIMGSQSDWETMRHADALLTELEIPHETLIVSAHRTPDRLADYARTAAERGLNVIIA 82 Query 64 GAGGAAHLPGMVAAKTNLPVIGVPVQSKALNGLDSLLSIVQMPGGVPVATVAIGKAGSTN 123 GAGGAAHLPGM AA T LPV+GVPV+S+AL G+DSLLSIVQMPGGVPV T+AIG +G+ N Sbjct 83 GAGGAAHLPGMCAAWTRLPVLGVPVESRALKGMDSLLSIVQMPGGVPVGTLAIGASGAKN 142
while a .oca file contains output from the OCA database:
>>PDB:1U11 _A mol:protein length:182 Pure (N5-Carboxy (182 aa) initn: 605 init1: 605 opt: 606 Z-score: 729.8 bits: 141.5 E(): 1.3e-33 Smith-Waterman score: 606; 65.972% identity (65.972% ungapped) in 144 aa overla p (4-147:23-166) 10 20 30 40 SEARCH MKSLVGVIMGSTSDWETMKYACDILDELNIPYEKKVVSAHR .::.:::: ::::::..: .: ::.::.: .::::: PDB:1U MSETAPLPSASSALEDKAASAPVVGIIMGSQSDWETMRHADALLTELEIPHETLIVSAHR 10 20 30 40 50 60 50 60 70 80 90 100 SEARCH TPDYMFEYAETARERGLKVIIAGAGGAAHLPGMVAAKTNLPVIGVPVQSKALNGLDSLLS ::: . .::.:: ::::.::::::::::::::: :: : :::.::::.:.::.:.::::: PDB:1U TPDRLADYARTAAERGLNVIIAGAGGAAHLPGMCAAWTRLPVLGVPVESRALKGMDSLLS 70 80 90 100 110 120 110 120 130 140 150 160 SEARCH IVQMPGGVPVATVAIGKAGSTNAGLLAAQILGSFHDDIHDALELRREAIEKDVREGSELV ::::::::::.:.::: .:. ::.::::.::. .. . :: : PDB:1U IVQMPGGVPVGTLAIGASGAKNAALLAASILALYNPALAARLETWRALQTASVPNSPITE 130 140 150 160 170 180 PDB:1U DK
The alignment file should contain the target sequence first and the model sequence second.
In practice, the model sequence in the alignment file will often be different from the model sequence in the pdb file. This may be because the alignment only uses part of the model sequence, or because the structure determination has not resolved all residues. Chainsaw is capable of handling such differences automatically.
Chainsaw will work with both monomer and multimer search models. If you wish to use a monomer model and the model pdb file contains more than one chain, you will have to delete the surplus chains manually. If you use a multimer model, chainsaw will apply the same alignment to each successive chain in the input pdb file.
Chainsaw will output a list of conserved/mutated/deleted residues and an estimate of the sequence identity. If this estimate is less than you expect, it is a sign that something has gone wrong, usually a problem with missing residues.
Possible keywords are:
chainsaw xyzin 1a80_1.pdb alignin 1mzr_1a80.pir xyzout 1a80_1_chainsaw.pdb <<eof END eof
Randy Read, Eleanor Dodson, Martyn Winn.