PFRMAT TS TARGET T0413 AUTHOR 4008-1775-0004 METHOD The SAM-T06 hand predictions use methods similar to SAM_T04 in CASP6 and METHOD the SAM-T02 method in CASP5. METHOD METHOD We start with a fully automated method (implemented as the SAM_T06 server): METHOD METHOD Use the SAM-T2K and SAM-T04 methods for finding homologs of the METHOD target and aligning them. The hand method also uses the METHOD experimental new SAM-T06 alignment method, which we hope is both METHOD more sensitive and lass prone to contamination by unrelated sequences. METHOD METHOD Make local structure predictions using neural nets and the METHOD multiple alignments. METHOD METHOD We currently use 10 local-structure alphabets: METHOD DSSP METHOD STRIDE METHOD STR2 an extended version of DSSP that splits the beta strands METHOD into multiple classes (parallel/antiparallel/mixed, METHOD edge/center) METHOD ALPHA an discretization of the alpha torsion angle: METHOD CA(i-i), CA(i), CA(i+1), CA(i+2) METHOD BYS a discretization of Ramachandran plots, due to Bystroff METHOD CB_burial_14_7 a 7-state discretization of the number of C_beta METHOD atoms in a 14 Angstrom radius sphere around the C_beta. METHOD near-backbone-11 an 11-state discretization of the number of METHOD residues (represented by near-backbone points) in a METHOD 9.65 Angstrom radius sphere around the sidechain proxy METHOD spot for the residue. METHOD DSSP_EHL2 CASP's collapse of the DSSP alphabet METHOD DSSP_EHL2 is not predicted directly by a METHOD neural net, but is computed as a weighted METHOD average of the other backbone alphabet predictions. METHOD O_NOTOR2 an alphabet for predicting characteristics of hydrogen METHOD bonds from the carbonyl oxygen METHOD N_NOTOR2 an alphabet for predicting characteristics of hydrogen METHOD bonds from the amide nitrogen METHOD We hope to add more networks for other alphabets over the summer. METHOD METHOD We make 2-track HMMs with each alphabet (1.0 amino acid + 0.3 METHOD local structure) and use them to score a template library of about METHOD 8000 (t06), 10000 (t04), or 15000 (t2k) templates. METHOD The template libraries are expanded weekly, but old template HMMs METHOD are not rebuilt. METHOD METHOD We also used a single-track HMM to score not just the template METHOD library, but a non-redundant copy of the entire PDB. METHOD METHOD One-track HMMs built from the template library multiple alignments METHOD were used to score the target sequence. METHOD METHOD All the logs of e-values were combined in a weighted average (with METHOD rather arbitrary weights, since we still have not taken the time METHOD to optimize them), and the best templates ranked. METHOD METHOD Alignments of the target to the top templates were made using METHOD several different alignment methods (mainly using the SAM hmmscore METHOD program, but a few alignments were made with Bob Edgar's MUSCLE METHOD profile-profile aligner). METHOD METHOD Generate fragments (short 9-residue alignments for each position) METHOD using SAM's "fragfinder" program and the 3-track HMM which tested METHOD best for alignment. METHOD METHOD Residue-residue contact predictions are made using mutual METHOD information, pairwise contact potentials, joint entropy, and other METHOD signals combined by a neural net. The contact prediction method METHOD is expected to evolve over the summer, as new features are METHOD selected and new networks trained. METHOD METHOD Then the "undertaker" program (named because it optimizes burial) METHOD is used to try to combine the alignments and the fragments into a METHOD consistent 3D model. No single alignment or parent template was METHOD used as a frozen core, though in many cases one had much more METHOD influence than the others. The alignment scores were not passed METHOD to undertaker, but were used only to pick the set of alignments METHOD and fragments that undertaker would see. Helix and strand METHOD constraints generated from the secondary-structure predictions are METHOD passed to undertaker to use in the cost function, as are the METHOD residue-residue contact prediction. METHOD METHOD One important change in this server over previous methods is that METHOD sheet constraints are extracted from the top few alignments and METHOD passed to undertaker. METHOD METHOD After the automatic prediction is done, we examine it by hand and try METHOD to fix any flaws that we see. This generally involves rerunning METHOD undertaker with new cost functions, increasing the weights for METHOD features we want to see and decreasing the weights where we think the METHOD optimization has gone overboard. Sometimes we will add new templates METHOD or remove ones that we think are misleading the optimization process. METHOD METHOD New this year, we are also occasionally using ProteinShop to METHOD manipulate proteins by hand, to produce starting points for undertaker METHOD optimization. We expect this to be most useful in new-fold all-alpha METHOD proteins, where undertaker often gets trapped in poor local minima by METHOD extending helices too far. METHOD METHOD Another new trick is to optimize models with gromacs to knock them out METHOD of a local minimum. The gromacs optimization does terrible things to METHOD the model (messing up sidechains and peptide planes), but is good at METHOD removing clashes. The resulting models are only a small distance from METHOD the pre-optimization models, but score much worse with the undertaker METHOD cost functions, so undertaker can move them more freely than models it METHOD has optimized itself. METHOD METHOD Although the superfamily for T0413 was fairly obvious (c.69.1.* = METHOD alpha/beta-Hydrolases), there are at least 35 families in that METHOD superfamily, and choosing the right family is not obvious. The METHOD HMMs built from different multiple alignments favored different METHOD families. METHOD METHOD Closing gaps turned out to be a bit of a problem with most of the METHOD models. I ended up doing some cut-and-paste work to combine pieces of METHOD different models, then optimizing in a dimeric context to clean up METHOD clashes between the monomers. The dimers were initially created by METHOD superimposing the monomer models on the 2hu5 pdb file. METHOD METHOD I did further cut-and-paste work to clean up the dimer interface, and METHOD optimized with undertaker to try to reduce breaks. METHOD METHOD METHOD Model METHOD METHOD 1 dimer/try20-opt3 < dimer-chimera2-try12 METHOD METHOD Optimized from a dimer that superimposed chimera2 monomers on the METHOD dimer/try12 model METHOD chimera2 is mainly try9-opt3, but R232-L263 from MQAU1 METHOD METHOD 2 dimer/try12-opt3.unpack.gromacs0.repack-nonPC