/// \brief TODOCUMENT
size_t geometric_mean_length_getter::do_get_length(const protein &arg_protein_a, ///< TODOCUMENT
const protein &arg_protein_b ///< TODOCUMENT
) const {
const size_t length_a = arg_protein_a.get_length();
const size_t length_b = arg_protein_b.get_length();
const double geom_mean_doub = sqrt( numeric_cast<double>( length_a * length_b ) );
return numeric_cast<size_t>( round( geom_mean_doub ) );
}
/// \brief TODOCUMENT
void cath::test::dssp_wolf_file_test_suite_fixture::check_pdb_and_dssp_built_protein(const string &arg_example_id ///< TODOCUMENT
) {
ostringstream build_prot_warn_stream;
const auto the_pdb_file = read_pdb_file ( pdb_file_of_example_id( arg_example_id ) );
const auto the_dssp_file = read_dssp_file( dssp_file_of_example_id( arg_example_id ) );
const auto pdb_prot = build_protein_of_pdb( the_pdb_file, reference_wrapper<ostream>{ build_prot_warn_stream } );
const auto num_non_null_dssp_residues = get_num_non_null_residues( the_dssp_file );
const auto num_pdb_residues = pdb_prot.get_length();
BOOST_CHECK_EQUAL( build_prot_warn_stream.str(), "" );
ostringstream test_ss;
const log_to_ostream_guard the_guard{ test_ss };
const protein combi_prot_with_all_pdb_residues = protein_from_dssp_and_pdb(the_dssp_file, the_pdb_file, dssp_skip_policy::DONT_SKIP__BREAK_ANGLES );
const protein combi_prot_with_dssp_only_residues = protein_from_dssp_and_pdb(the_dssp_file, the_pdb_file, dssp_skip_policy::SKIP__BREAK_ANGLES );
BOOST_REQUIRE_EQUAL( num_non_null_dssp_residues, combi_prot_with_dssp_only_residues.get_length() );
BOOST_REQUIRE_EQUAL( num_pdb_residues, combi_prot_with_all_pdb_residues.get_length() );
// Compare the combi's residues with the DSSP residues
//( this requires two counters to handle skipping null DSSP residues)
size_t combi_residue_ctr = 0;
for (const residue dssp_residue : the_dssp_file) {
// If this is a null DSSP residue then just move onto the next one
if ( is_null_residue(dssp_residue) ) {
continue;
}
const residue built_residue = combi_prot_with_dssp_only_residues.get_residue_ref_of_index( combi_residue_ctr );
BOOST_CHECK_EQUAL( dssp_residue.get_pdb_residue_id(), built_residue.get_pdb_residue_id() );
BOOST_CHECK_EQUAL( dssp_residue.get_sec_struc_number(), built_residue.get_sec_struc_number() );
BOOST_CHECK_EQUAL( dssp_residue.get_sec_struc_type(), built_residue.get_sec_struc_type() );
// cerr << dssp_residue << endl;
// cerr << built_residue << endl;
// cerr << endl;
++combi_residue_ctr;
}
for (size_t pdb_residue_ctr = 0; pdb_residue_ctr < num_pdb_residues; ++pdb_residue_ctr) {
const pdb_residue the_pdb_residue = the_pdb_file.get_residue_cref_of_index__backbone_unchecked(pdb_residue_ctr);
const residue built_residue = combi_prot_with_all_pdb_residues.get_residue_ref_of_index( pdb_residue_ctr );
BOOST_CHECK_EQUAL( the_pdb_residue.get_residue_id(), built_residue.get_pdb_residue_id() );
BOOST_CHECK_EQUAL( get_carbon_alpha_coord( the_pdb_residue ), built_residue.get_carbon_alpha_coord() );
BOOST_CHECK_EQUAL( get_or_predict_carbon_beta_coord_of_residue( the_pdb_residue ), built_residue.get_carbon_beta_coord() );
// cerr << pdb_residue << endl;
// cerr << built_residue << endl;
// cerr << endl;
}
}
/// \brief TODOCUMENT
inline single_struc_res_pair_vec build_single_rep_pairs(const protein &arg_protein ///< TODOCUMENT
) {
const auto num_residues = debug_unwarned_numeric_cast<index_type>( arg_protein.get_length() );
const auto scan_phi_angles = make_scan_phi_angles ( arg_protein );
const auto scan_psi_angles = make_scan_psi_angles ( arg_protein );
const auto scan_view_coords = make_scan_view_coords ( arg_protein );
const auto scan_frames = make_scan_frame_quat_rots( arg_protein );
single_struc_res_pair_vec results;
results.reserve( num_residues * num_residues );
const auto all_residues_range = boost::irange<index_type>( 0, num_residues );
for (const auto &x : common::cross( all_residues_range, all_residues_range ) ) {
const auto &from_res_index = std::get<0>( x );
const auto &to_res_index = std::get<1>( x );
results.emplace_back(
rotate_copy(
scan_frames [ from_res_index ],
scan_view_coords[ to_res_index ] - scan_view_coords[ from_res_index ]
),
rotation_between_rotations(
scan_frames[ from_res_index ],
scan_frames[ to_res_index ]
),
scan_phi_angles[ from_res_index ],
scan_psi_angles[ from_res_index ],
scan_phi_angles[ to_res_index ],
scan_psi_angles[ to_res_index ],
from_res_index,
to_res_index
);
}
return results;
}
void add_structure_to_store(T &arg_store, ///< TODOCUMENT
const index_type &arg_structure_index, ///< TODOCUMENT
const protein &arg_protein, ///< TODOCUMENT
const scan_policy<KPs...> &arg_scan_policy, ///< TODOCUMENT
const scan_role &arg_scan_role ///< TODOCUMENT
) {
const auto &the_keyer = arg_scan_policy.get_keyer();
// BOOST_LOG_TRIVIAL( warning ) << "Keyer is : " << the_keyer;
const auto roled_stride = roled_scan_stride{ arg_scan_role, arg_scan_policy.get_scan_stride() };
const auto num_residues = debug_unwarned_numeric_cast<index_type>( arg_protein.get_length() );
const auto from_rep_strider = get_this_from_strider( roled_stride );
const auto to_rep_strider = get_this_to_strider ( roled_stride );
for (const auto &from_rep_index : get_rep_indices_range( from_rep_strider, num_residues ) ) {
// BOOST_LOG_TRIVIAL( warning ) << "\tFrom rep " << from_rep_index;
for (const auto &to_rep_index : get_rep_indices_range( to_rep_strider, num_residues ) ) {
if ( from_rep_index != to_rep_index ) {
const auto the_res_pair = make_multi_struc_res_rep_pair(
arg_protein.get_residue_ref_of_index( get_index_of_rep_index( from_rep_strider, from_rep_index ) ),
arg_protein.get_residue_ref_of_index( get_index_of_rep_index( to_rep_strider, to_rep_index ) ),
arg_structure_index,
from_rep_index,
to_rep_index
);
// BOOST_LOG_TRIVIAL( warning ) << "\t\tTo rep " << to_rep_index << " - the rep : " << the_res_pair;
arg_store.push_back_entry_to_cell( the_keyer.make_key( the_res_pair ), the_res_pair );
}
}
}
}
/// \brief TODOCUMENT
inline single_struc_res_pair_list_vec build_rep_sets(const protein &arg_protein, ///< TODOCUMENT
const roled_scan_stride &arg_roled_scan_stride ///< TODOCUMENT
) {
const auto num_residues = debug_unwarned_numeric_cast<index_type>( arg_protein.get_length() );
const auto &all_single_res_pairs = build_single_rep_pairs( arg_protein );
const auto from_range = get_indices_range( get_this_from_strider( arg_roled_scan_stride ), num_residues );
const auto to_range = get_indices_range( get_this_to_strider ( arg_roled_scan_stride ), num_residues );
single_struc_res_pair_list_vec results;
results.reserve( from_range.size() * to_range.size() );
for (const auto &x : common::cross( from_range, to_range ) ) {
// std::cerr << "Building rep: [" << std::get<0>( x ) << ", " << std::get<1>( x ) << "]\n";
results.emplace_back();
detail::add_stride_neighbours(
results.back(),
num_residues,
all_single_res_pairs,
std::get<0>( x ),
std::get<1>( x ),
from_co_stride( arg_roled_scan_stride ),
to_co_stride ( arg_roled_scan_stride )
);
}
return results;
}
/// \brief TODOCUMENT
inline frame_quat_rot_vec make_scan_frame_quat_rots(const protein &arg_protein ///< TODOCUMENT
) {
frame_quat_rot_vec results;
results.reserve( arg_protein.get_length() );
for (const auto &x : arg_protein) {
results.emplace_back( geom::make_quat_rot_from_rotation<frame_quat_rot_type>( x.get_frame() ) );
}
return results;
}
/// \brief TODOCUMENT
inline view_type_vec make_scan_view_coords(const protein &arg_protein ///< TODOCUMENT
) {
view_type_vec results;
results.reserve( arg_protein.get_length() );
for (const auto &x : arg_protein) {
results.emplace_back( x.get_carbon_beta_coord() );
}
return results;
}
/// \brief TODOCUMENT
inline angle_type_vec make_scan_psi_angles(const protein &arg_protein ///< TODOCUMENT
) {
angle_type_vec results;
results.reserve( arg_protein.get_length() );
for (const auto &x : arg_protein) {
results.emplace_back( geom::convert_angle_type<angle_base_type>( x.get_psi_angle() ) );
}
return results;
}
/// \brief TODOCUMENT
view_cache_index_entry_vec view_cache_index_entry_test_suite_fixture::build_random_vcies(const size_t &arg_num_entries, ///< TODOCUMENT
const protein &arg_protein, ///< TODOCUMENT
mt19937 &arg_rng ///< TODOCUMENT
) {
if ( arg_protein.get_length() < 2 ) {
BOOST_THROW_EXCEPTION(invalid_argument_exception("Cannot build random vcies if the protein has fewer than two residues"));
}
const auto &length = arg_protein.get_length();
return generate_n_build<view_cache_index_entry_vec>(
arg_num_entries,
[&] () {
const auto index_pair = pick_random_pair( 0_z, length - 1, arg_rng );
return make_view_cache_index_entry(
arg_protein,
index_pair.first,
index_pair.second
);
}
);
}
/// \brief TODOCUMENT
void cath::test::dssp_wolf_file_test_suite_fixture::compare_residue_frames_from_pdb_vs_wolf(const string &arg_example_id ///< TODOCUMENT
) {
ostringstream build_prot_warn_stream;
const wolf_file the_wolf_file = read_wolf( wolf_file_of_example_id( arg_example_id ) );
const pdb the_pdb_file = read_pdb_file( pdb_file_of_example_id( arg_example_id ) );
const protein pdb_prot = build_protein_of_pdb( the_pdb_file, reference_wrapper<ostream>{ build_prot_warn_stream });
const wolf_file::size_type num_wolf_residues = the_wolf_file.get_num_residues();
const size_t num_pdb_residues = pdb_prot.get_length();
BOOST_CHECK_EQUAL( build_prot_warn_stream.str(), "" );
BOOST_REQUIRE_EQUAL(num_pdb_residues, num_wolf_residues);
doub_angle_vec angles_between_frame_pairs;
for (size_t residue_ctr = 0; residue_ctr < num_pdb_residues; ++residue_ctr) {
const residue &wolf_residue = the_wolf_file.get_residue_of_index( residue_ctr );
const residue &pdb_prot_residue = pdb_prot.get_residue_ref_of_index( residue_ctr );
BOOST_REQUIRE_EQUAL( pdb_prot_residue.get_pdb_residue_id(), wolf_residue.get_pdb_residue_id() );
const rotation raw_wolf_frame = wolf_residue.get_frame();
const rotation wolf_frame = tidy_copy(raw_wolf_frame, 0.001);
const rotation pdb_frame = pdb_prot_residue.get_frame();
const doub_angle the_angle = angle_between_rotations( wolf_frame, pdb_frame );
angles_between_frame_pairs.push_back( the_angle );
}
BOOST_REQUIRE_GT( angles_between_frame_pairs.size(), 0_z );
// Sort the list of angles in ascending order
sort( angles_between_frame_pairs );
// Check that around 95% of the way through the list of angles, the values are still
// less than 3 degrees (in radians)
const size_t ninety_fifth_perc_index = numeric_cast<size_t>(
round( 0.95 * numeric_cast<double>( angles_between_frame_pairs.size() ) )
);
BOOST_CHECK_LT(
angles_between_frame_pairs[ ninety_fifth_perc_index ],
make_angle_from_degrees<double>( 3.0 )
);
}
/// \brief Private static method that implements the process of building the views from proteins
coord_vec_vec view_cache::build_views(const protein &arg_protein ///< The protein which the view_cache should be built to represent
) {
// Grab the number of residues and prepare the views accordingly
const size_t num_residues = arg_protein.get_length();
coord_vec_vec new_views( num_residues );
for (coord_vec &view_of : new_views) {
view_of.reserve( num_residues );
}
// Loop over the all from-versus-to residue pairs and add the resulting views
for (const size_t &from_res_ctr : irange( 0_z, num_residues ) ) {
for (const size_t &to_res_ctr : irange( 0_z, num_residues ) ) {
coord_vec &view_of_from = new_views[ from_res_ctr ];
view_of_from.push_back( view_vector_of_residue_pair(
arg_protein.get_residue_ref_of_index( from_res_ctr ),
arg_protein.get_residue_ref_of_index( to_res_ctr )
) );
}
}
return new_views;
}
/// \brief TODOCUMENT
size_t residue_querier::do_get_length(const protein &arg_protein ///< TODOCUMENT
) const {
return arg_protein.get_length();
}
/// \brief TODOCUMENT
size_t length_of_second_getter::do_get_length(const protein &/*arg_protein_a*/, ///< TODOCUMENT
const protein &arg_protein_b ///< TODOCUMENT
) const {
return arg_protein_b.get_length();
}
/// \brief TODOCUMENT
pair<str_vec, str_vec> check_scan_on_final_alignment::get_rep_name_lists(const protein &arg_protein, ///< TODOCUMENT
const roled_scan_stride &arg_roled_scan_stride ///< TODOCUMENT
) const {
// const auto &num_residues = arg_protein.get_length();
const auto rep_list_indices = get_rep_index_lists( arg_roled_scan_stride, numeric_cast<index_type>( arg_protein.get_length() ) );
const auto residue_name_of_index = [&] (const index_type &x) { return get_pdb_residue_name_string( arg_protein.get_residue_ref_of_index( x ) ); };
return make_pair(
transform_build<str_vec>( rep_list_indices.first, residue_name_of_index ),
transform_build<str_vec>( rep_list_indices.second, residue_name_of_index )
);
}
void dense_add_structure_to_store(T &arg_store, ///< TODOCUMENT
const index_type &arg_structure_index, ///< TODOCUMENT
const protein &arg_protein, ///< TODOCUMENT
const scan_policy<KPs...> &arg_scan_policy, ///< TODOCUMENT
const scan_role &arg_scan_role ///< TODOCUMENT
) {
const auto &the_criteria = arg_scan_policy.get_criteria();
const auto &the_keyer = arg_scan_policy.get_keyer();
// BOOST_LOG_TRIVIAL( warning ) << "Keyer is : " << the_keyer;
const auto roled_stride = roled_scan_stride{ arg_scan_role, arg_scan_policy.get_scan_stride() };
const auto num_residues = debug_unwarned_numeric_cast<index_type>( arg_protein.get_length() );
const auto from_rep_strider = get_this_from_strider( roled_stride );
const auto to_rep_strider = get_this_to_strider ( roled_stride );
for (const auto &from_rep_index : get_rep_indices_range( from_rep_strider, num_residues ) ) {
// BOOST_LOG_TRIVIAL( warning ) << "\tFrom rep " << from_rep_index;
for (const auto &to_rep_index : get_rep_indices_range( to_rep_strider, num_residues ) ) {
// BOOST_LOG_TRIVIAL( warning ) << "\t\tTo rep " << to_rep_index;
if ( from_rep_index != to_rep_index ) {
const auto the_res_pair = make_multi_struc_res_rep_pair(
arg_protein.get_residue_ref_of_index( get_index_of_rep_index( from_rep_strider, from_rep_index ) ),
arg_protein.get_residue_ref_of_index( get_index_of_rep_index( to_rep_strider, to_rep_index ) ),
arg_structure_index,
from_rep_index,
to_rep_index
);
// BOOST_LOG_TRIVIAL( warning ) << "\t\tMade multi_struc_res_rep_pair : " << the_res_pair;
// const auto the_key = the_keyer.make_key ( the_res_pair );
// BOOST_LOG_TRIVIAL( warning ) << "\t\tThe actual key itself is : " << output_key(the_key );
const auto close_keys = the_keyer.make_close_keys( the_res_pair, the_criteria );
// BOOST_LOG_TRIVIAL( warning ) << "\t\t\t(from_phi : "
// << the_res_pair.get_res_pair_core().get_from_phi_angle()
// << ") There are "
// << boost::distance( common::cross( close_keys ) )
// << " keys close to "
// << output_key( the_keyer.make_key( the_res_pair ) );
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 0 : \"" << boost::algorithm::join ( std::get<0>( close_keys ) | boost::adaptors::transformed( [](const uint8_t &x) { return std::to_string( static_cast<size_t>( x) ); } ) , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 1 : \"" << boost::algorithm::join ( std::get<1>( close_keys ) | boost::adaptors::transformed( [](const uint8_t &x) { return std::to_string( static_cast<size_t>( x) ); } ) , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 2 : \"" << boost::algorithm::join ( std::get<2>( close_keys ) | boost::adaptors::transformed( [](const uint8_t &x) { return std::to_string( static_cast<size_t>( x) ); } ) , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 3 : \"" << boost::algorithm::join ( std::get<3>( close_keys ) | boost::adaptors::transformed( [](const uint8_t &x) { return std::to_string( static_cast<size_t>( x) ); } ) , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 4 : \"" << boost::algorithm::join ( std::get<4>( close_keys ) | common::lexical_casted<std::string>() , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 5 : \"" << boost::algorithm::join ( std::get<5>( close_keys ) | common::lexical_casted<std::string>() , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 6 : \"" << boost::algorithm::join ( std::get<6>( close_keys ) | common::lexical_casted<std::string>() , "\", \"") << "\"";
// BOOST_LOG_TRIVIAL( warning ) << "\t\tRange 7 : \"" << boost::algorithm::join ( std::get<7>( close_keys ) | common::lexical_casted<std::string>() , "\", \"") << "\"";
// KEYER_PARTS: tuple<
// res_pair_phi_psi_angle_keyer_part<res_pair_from_phi_keyer_part_spec>,
// res_pair_phi_psi_angle_keyer_part<res_pair_from_psi_keyer_part_spec>,
// res_pair_phi_psi_angle_keyer_part<res_pair_to_phi_keyer_part_spec>,
// res_pair_phi_psi_angle_keyer_part<res_pair_to_psi_keyer_part_spec>,
// res_pair_index_dirn_keyer_part,
// res_pair_view_axis_keyer_part<res_pair_view_x_keyer_part_spec>,
// res_pair_view_axis_keyer_part<res_pair_view_y_keyer_part_spec>,
// res_pair_view_axis_keyer_part<res_pair_view_z_keyer_part_spec>
// >
// KEY: tuple<unsigned char, unsigned char, unsigned char, unsigned char, res_pair_dirn, short, short, short>
// CLOSE_KEYS: tuple<
// joined_range<const integer_range<unsigned char>, const integer_range<unsigned char> >,
// joined_range<const integer_range<unsigned char>, const integer_range<unsigned char> >,
// joined_range<const integer_range<unsigned char>, const integer_range<unsigned char> >,
// joined_range<const integer_range<unsigned char>, const integer_range<unsigned char> >,
// array<res_pair_dirn, 1u>,
// integer_range<short int>,
// integer_range<short int>,
// integer_range<short int>
// >;
// ELEMENT_OF_CLOSE_KEYS: tuple<unsigned char, unsigned char, unsigned char, unsigned char, const res_pair_dirn &, short, short, short>
// const bool contains_key = common::contains( common::cross( close_keys ), the_keyer.make_key( the_res_pair ) );
// if ( contains_key ) {
// BOOST_LOG_TRIVIAL( warning ) << "\t\t\tDoes contain the central key";
// }
// else {
// BOOST_LOG_TRIVIAL( warning ) << "\t\t\t***** Doesn't contain the central key";
// }
// size_t counter = 0;
for (const auto &x : common::cross( close_keys ) ) {
// if ( ! contains_key ) {
// BOOST_LOG_TRIVIAL( warning ) << "\t\t\t\t[" << counter << "] : Adding entry for close key " << output_key( x );
// }
arg_store.push_back_entry_to_cell( x, the_res_pair );
// ++counter;
}
}
}
}
arg_store.summarize();
}