void Run(Vergil* vergil) {
std::string input_directory = "/Users/Violet/data/peptide_design/2D_lattice/C4459/P422/structure";
std::string output_directory = "/Users/Violet/data/peptide_design/2D_lattice/C4459/P422/structure/P422_a31.2_Cu/";
std::string pdb_filename = input_directory + "/C4459_P422_a31.2_mP_type_wCap_W_H_segX0.pdb";
//std::string pdb_filename = input_directory + "/C4459_P422_a31.7fix_19W_27H.pdb";
//Geometry variables
double chi_incr = 1.0;
std::vector<double> chi_stdev;
chi_stdev.push_back(180.0);
chi_stdev.push_back(180.0);
std::string lp_atom = "NE2";
double unit_cell_length = 31.2;
double angle_tolerance = 5.0;
//Create a local Timer instance
Timer timer;
//Load force field and design topology
vergil->LoadAMBERUnitedAtom();
vergil->LoadSavenDesignStandards();
unsigned int max_dunbrack_rots = 9;
vergil->LoadDunbrackRotamerLibrary(2002, max_dunbrack_rots);
std::string scaffold = "protein and backbone or not site 23";
InputPDBScaffold input(vergil->domain(), scaffold);
input.KeepSitesWithMissingAtoms();
input.Read(pdb_filename);
Log->print_tag("INPUT", "PDB scaffold -- " + pdb_filename);
std::string typelist[] = { "ASP", "GLN", "GLU", "ILE", "ARG", "GLN", "MET", "ALA", "GLU", "TRP", "ILE", "LYS", "LYS",
"MET", "ALA", "GLN", "MET", "ILE", "ASP", "LYS", "MET", "ALA", "HIS", "ARG", "ILE", "ASP", "ARG", "GLU", "ALA",
"NH2" };
// std::string typelist[] = { "ASP", "GLN", "GLU", "ILE", "ARG", "GLN", "MET", "ALA", "GLU", "GLU", "ILE", "ARG", "LYS",
// "MET", "ALA", "GLN", "GLN", "ILE", "TRP", "GLU", "MET", "ALA", "LYS", "ARG", "ILE", "ASP", "HIS", "GLU", "ALA", "NH2"};
size_t i = 0;
for (Domain::SiteIterator it = vergil->domain()->SiteIterator_Begin(); it != vergil->domain()->SiteIterator_End();
++it) {
it->Add(typelist[i]);
++i;
}
vergil->BuildDomain();
timer.Stamp();
Conformer* ligand = &vergil->domain()[0]["X0"]["A"][0][23]["HIS"][0];
OptimizeLigandGeomety(*ligand, chi_incr, chi_stdev, lp_atom, unit_cell_length, angle_tolerance, output_directory);
Log->print_tag("RUNTIME", "Optimize ligand geometry time: " + timer.ElapsedToString());
vergil->domain()->Clear();
}
void Design(Vergil* vergil) {
//set design parameters
double beta = 0.5;
double ref_beta = ROOM_TEMPERATURE_BETA;
double twobody_energy_cap = 30;
double a_list[] = { 31.7 };
std::string input_directory = "/Users/Violet/data/peptide_design/tetramer_29/fullseq_design_approved_by_Jeff";
//std::string input_directory = "/Users/Violet/data/peptide_design/2D_lattice/C4459/P422/structure";
std::string output_directory = "/Users/Violet/data/peptide_design/2D_lattice/C4459/P422/vergil_design";
std::string pdb_filename = input_directory + "/C4459a_chainA_fixed_adg_NH2cap_for_vergil.pdb";
//std::string pdb_filename = input_directory + "/C4459_P422_a31.2_mP_type_wCap_W_H_segX0.pdb";
//Create a local Timer instance
Timer timer;
//Load force field and design topology
vergil->LoadAMBERUnitedAtom();
vergil->LoadSavenDesignStandards();
unsigned int max_dunbrack_rots = 81;
vergil->LoadDunbrackRotamerLibrary(2002, max_dunbrack_rots);
//Compute and time the reference energy
timer.Stamp();
Log->print("Computing reference energies....");
PotentialUnfoldedFreeEnergy unfolded_compute(PotentialUnfoldedFreeEnergy::AMBER_KONO,
vergil->forcefield_parameter_library(), vergil->topology_library(),
vergil->conformer_library());
unfolded_compute.set_beta(ref_beta);
unfolded_compute.AddPotential(vergil->potential("dihedral"));
unfolded_compute.AddPotential(vergil->potential("vanderwaals"));
unfolded_compute.AddPotential(vergil->potential("electrostatic"));
unfolded_compute.GenerateUnfoldedFreeEnergies();
Log->print_tag("RUNTIME", "Reference energy calculation: " + timer.ElapsedToString());
for (int idx = 0; idx < 1; ++idx) {
std::string output_fileprefix = output_directory + "/C4459_P422_a" + Log->to_str(a_list[idx]) + "fix_19W_27H";
// Load scaffold
std::string scaffold = "protein and backbone or site 1 4 7 8 11 14 15 18 21 22 25 28 29 30";
// std::string scaffold = "protein and backbone or site 1 to 15 or site 17 to 30";
InputPDBScaffold input(vergil->domain(), scaffold);
input.KeepSitesWithMissingAtoms();
input.Read(pdb_filename);
Log->print_tag("INPUT", "PDB scaffold -- " + pdb_filename);
// Type each site in the domain
int num_amino_acids = 18;
std::string amino_acid_set[] = {"ALA", "ARG", "ASN", "ASP", "GLN", "GLU", "GLY", "HIS", "ILE", "LEU", "LYS","MET",
"PHE", "SER", "THR", "TRP", "TYR", "VAL"};
std::string typelist[] = {"ASP", "ILE", "MET", "ALA", "ILE", "MET", "ALA", "ILE", "MET", "ALA", "ILE", "GLU", "ALA"};
//std::string typelist_ex[] = { "GLN", "GLU", "ARG", "GLN", "GLU", "TRP", "LYS", "LYS", "LEU", "MET", "ASP", "LYS", "HIS", "ARG", "ASP", "ARG", "NH2" };
size_t i = 0;
// size_t j = 0;
for (Domain::SiteIterator it = vergil->domain()->SiteIterator_Begin(); it != vergil->domain()->SiteIterator_End();
++it) {
size_t heptad_id = it->name() % 7;
if (heptad_id == 1 || heptad_id == 4 || heptad_id == 0) {
it->Add(typelist[i]);
++i;
// } else {
// //must only add the reference type to the fixed sites
// it->Add(typelist_ex[j]);
// ++j;
// }
} else if (it->name() == 27) {
it->Add("HIS");
} else if (it->name() == 19) {
it->Add("TRP");
} else if (it->name() == 30) {
it->Add("NH2");
} else {
for (int j = 0; j < num_amino_acids; ++j)
it->Add(amino_acid_set[j]);
}
}
// Build all conformers
vergil->BuildDomain();
vergil->TrimDomain(30.0);
//Offset the domain(ASU) to the correct position in the unit cell
const Vector3 offset(a_list[idx] / 2, 0, 0);
vergil->domain()->TransformBy(Geometry::Transformation(Vector3(0, 0, 0), 0.0, offset));
//Set up symmetry and build lattice
vergil->domain()->set_unit_cell_parameters("P422", a_list[idx], a_list[idx], 1000, 90.0, 90.0, 90.0);
SymmetryGenerator<Domain> symexp("P422", a_list[idx], a_list[idx], 1000, 90.0, 90.0, 90.0);
symexp.BuildLattice(*vergil->domain(), -1, -1, 0, 1, 1, 0);
std::vector<Domain>* symmetry_related_elements = symexp.lattice();
//Trim rotamers that clash with the scaffold
vergil->TrimDomain(30.0, symmetry_related_elements);
// Set up energy function
timer.Stamp();
Log->print("Setting up energy function...");
FunctionMeanFieldEnergyLattice energy(vergil->domain(), symmetry_related_elements);
energy.set_pairwise_energy_cap_values(twobody_energy_cap, twobody_energy_cap);
energy.AddPotential(vergil->potential("dihedral"));
energy.AddPotential(vergil->potential("vanderwaals"));
energy.AddPotential(vergil->potential("electrostatic"));
energy.ComputeEnergies();
Log->print_tag("RUNTIME", "Energy Matrix Fill Time: " + timer.ElapsedToString());
// Initialize probabilities
vergil->domain()->SetUniformConformerProbabilities();
// Set up the problem and solve the unconstrained problem
FunctionFreeEnergyWithReference free_energy(&energy, beta, &unfolded_compute);
vergil->InitProbabilityProblem(free_energy);
vergil->SolveProbabilityProblem();
//Output standard files (pdb, psf, seq, csv)
vergil->StandardOutput(output_fileprefix);
for (Domain::SiteIterator it = vergil->domain()->SiteIterator_Begin(); it != vergil->domain()->SiteIterator_End(); ++it) {
it->SetAggregatedTypeProbability();
}
vergil->RenameSymmetrySegname(symmetry_related_elements);
OutputPDB output1(output_fileprefix + "_lattice.pdb", vergil->domain());
output1.WriteLattice(symmetry_related_elements);
Log->print("Most Probable PDB -- " + output_fileprefix + "_lattice.pdb");
vergil->domain()->Clear();
}
}
void Design(Vergil* vergil) {
//Variables for the design
double design_beta = 0.5;
unsigned int max_dunbrack_rots = 10;
std::string pdb_filename = VERGIL_PATH "/working/input/4JKV.pdb";
std::string output_fileprefix = VERGIL_PATH "/working/output/example_output";
//Designable sites
int num_designable_sites = 22;
int designable_sites[] = { 243, 246, 253, 316, 362, 363, 369, 370, 373, 410, 413, 420, 424, 454, 457, 458, 460, 464,
471, 537, 542, 543 };
//Construct scaffold selection text from the array of designable sites
std::string scaffold_seltext = "chain A and resid 190 to 550"; //Only consider residues 190-550 on chain A
scaffold_seltext += " and (backbone or not resid "; //Fix the backbone atoms, or all the atoms in sites that are NOT designable
for (int i = 0; i < num_designable_sites; ++i) {
scaffold_seltext += Log->to_str(designable_sites[i]) + " ";
}
scaffold_seltext += ")";
//Create a local Timer instance
Timer timer;
//Load force field, dunbrack library, and scaffold from PDB file
vergil->LoadAMBERUnitedAtom();
vergil->LoadSavenDesignStandards();
vergil->LoadDunbrackRotamerLibrary(2002, max_dunbrack_rots);
vergil->LoadScaffoldFromPDB(pdb_filename, scaffold_seltext);
//The three usual computes are automatically loaded into Vergil: "dihedral", "vanderwaals", and "electrostatic"
// for the AMBER84 modified ff, where "vanderwaals" is the KonoModified compute with hbonding
//Compute and time the reference energy
timer.Stamp();
Log->print("Computing reference energies....");
ComputeUnfoldedFreeEnergy unfolded_compute(ComputeUnfoldedFreeEnergy::AMBER_KONO, vergil->forcefield_parameter_library(),
vergil->topology_library(), vergil->conformer_library());
unfolded_compute.set_beta(ROOM_TEMPERATURE_BETA);
unfolded_compute.set_grid_space(90);
unfolded_compute.AddCompute(vergil->compute("dihedral"));
unfolded_compute.AddCompute(vergil->compute("vanderwaals"));
unfolded_compute.AddCompute(vergil->compute("electrostatic"));
unfolded_compute.GenerateUnfoldedFreeEnergies();
Log->print_tag("RUNTIME", "Reference energy calculation: " + timer.ElapsedToString());
//Type each site in the domain
Path* chain = &(*vergil->domain())["4JKV"]["A"][0];
//Disulfide bridges
(*chain)[193].reference_type()->set_name("CYX");
(*chain)[213].reference_type()->set_name("CYX");
(*chain)[193].ConnectNextSiteWithIterface((*chain)[213], "DISU");
(*chain)[217].reference_type()->set_name("CYX");
(*chain)[295].reference_type()->set_name("CYX");
(*chain)[217].ConnectNextSiteWithIterface((*chain)[295], "DISU");
(*chain)[314].reference_type()->set_name("CYX");
(*chain)[390].reference_type()->set_name("CYX");
(*chain)[314].ConnectNextSiteWithIterface((*chain)[390], "DISU");
(*chain)[490].reference_type()->set_name("CYX");
(*chain)[507].reference_type()->set_name("CYX");
(*chain)[490].ConnectNextSiteWithIterface((*chain)[507], "DISU");
int num_amino_acids = 20; //All 20 natural amino acids
std::string amino_acid_set[] = { "ALA", "ARG", "ASN", "ASP", "CYS", "GLN", "GLU", "GLY", "HIS", "ILE", "LEU", "LYS",
"MET", "PHE", "PRO", "SER", "THR", "TRP", "TYR", "VAL" };
for (int i = 0; i < num_designable_sites; ++i) {
for (int j = 0; j < num_amino_acids; ++j) {
(*chain)[designable_sites[i]].Add(amino_acid_set[j]);
}
}
// Build all conformers
vergil->BuildDomain();
//Trim rotamers that clash with the backbone
vergil->TrimDomain(30.0);
// Set up energy function
timer.Stamp();
Log->print("Setting up energy function...");
FunctionMeanFieldEnergy energy(vergil->domain());
energy.set_pairwise_energy_cap_values(30.0, 30.0);
energy.AddCompute(vergil->compute("dihedral"));
energy.AddCompute(vergil->compute("vanderwaals"));
energy.AddCompute(vergil->compute("electrostatic"));
energy.ComputeEnergies();
Log->print_tag("RUNTIME", "Energy Matrix Fill Time: " + timer.ElapsedToString());
// Initialize probabilities
vergil->domain()->SetUniformConformerProbabilities();
// Set up the problem and solve the unconstrained problem
FunctionFreeEnergyWithReference free_energy(&energy, design_beta, &unfolded_compute);
vergil->InitProbabilityProblem(free_energy);
vergil->SolveProbabilityProblem();
//Environmental Energy function
double Eenv_target = 0.0;
FunctionEnvironmentalEnergy environmental_energy(*vergil->domain(), ENVIRONMENTAL_ENERGY_FILE);
//Print out the Environmental Energy value
VectorN w = vergil->ConformerProbabilityVector();
Log->print(environmental_energy.to_string(w));
//Output standard files (pdb, psf, seq, csv)
vergil->StandardOutput(output_fileprefix + "_unconstrained");
//Now add the Environmental Energy constraint and resolve the problem
vergil->problem()->AddConstraint("Eenv", &environmental_energy, Eenv_target);
vergil->SolveProbabilityProblem();
//Print out the Environmental Energy value to check its been constrainted properly
w = vergil->ConformerProbabilityVector();
Log->print(environmental_energy.to_string(w));
//Output standard files (pdb, psf, seq, csv)
vergil->StandardOutput(output_fileprefix);
}