Constants

HHBLITS Mapping to Amino Acid Type

Name

Type

Description

Value

HHBLITS_AA_TO_ID

dict

Use the hhblits convention to map amino acids to corresponding numbers. Map B to D. Map J and O to X. Map U to C. Map Z to E. other letters are numbered in sequence.

HHBLITS_AA_TO_ID

ID_TO_HHBLITS_AA

dict

Inverse mapping of HHBLITS_AA_TO_ID, mapping the number to the amino acid type.

ID_TO_HHBLITS_AA

MAP_HHBLITS_AATYPE_TO_OUR_AATYPE

tuple

Mapping table from hhblits convention number to standard amino acid number, where the standard number is the alphabetical order of amino acids with three characters.

(0, 4, 3, 6, 13, 7, 8, 9, 11, 10, 12, 2, 14, 5, 1, 15, 16, 19, 17, 18, 20, 21)

The obtained MSA is converted from a letter to a number through HHBLITS_AA_TO_ID , and then converted to a restype id through MAP_HHBLITS_AATYPE_TO_OUR_AATYPE when it is sent to the network, consistent with other amino acid numbers in the network.

Amino Acid Correlation Constants

Name

Type

Description

Value

residue_atom_renaming_swaps

dict

Four of the 20 amino acids have partial atoms that are symmetrical. This dict records the mapping information of atomic pairs with symmetric invariance in each amino acid.

residue_atom_renaming_swaps

residue_atoms

dict

20 types of amino acids correspond to their atom types.

residue_atoms

restype_name_to_atom14_names

dict

Atom types each amino acid contains. If there are fewer than 14, use empty type "''" to make up for it. The actual atoms are of the same type as residue_atoms .

restype_name_to_atom14_names

restype_1to3

dict

The mapping information of amino acids represented by one characters with amino acids represented by three characters. Key is amino acids represented by one characters and value is amino acids represented by three characters.

restype_1to3

restype_3to1

dict

The mapping information of amino acids represented by three characters with amino acids represented by one characters. Key is amino acids represented by three characters and value is amino acids represented by one characters.

restype_3to1

restype_atom14_mask

array

Mask of densely encoding of 21 types of amino acid (20 amino acid types and 1 unknown amino acid UNK ).

restype_atom14_mask

restype_atom37_mask

array

Mask of sparse encoding of 21 types of amino acid (20 amino acid types and 1 unknown amino acid UNK ).

restype_atom37_mask

restype_num

scalar

Total number of amino acid type, excluding unknown amino acid UNK .

20

resnames

list

List of amino acid types represented by three characters, including 20 types of amino acid and one type of unknown amino acid represented by UNK .

resnames

resname_to_idx

dict

Index of the amino acid types represented by three characters. Order is the same as the order of amino acids in the list of resnames.

resname_to_idx

restypes

list

List of amino acids, including a total of 20 amino acids, excluding UNK.

restypes

restype_order

dict

The index of amino acid. Order is consistent with the order of amino acid in the list of restypes .

restype_order

restypes_with_x

list

List of amino acids including 'UNK'. 'UNK' means unknown amino acid and is represented by 'X'. Totally (20 + 1) types of amino acid are in the list.

restypes_with_x

restype_order_with_x

dict

Index of the amino acid types represented by one characters. Order is the same as the order of amino acids in the list of restypes_with_x.

restype_order_with_x

order_restype_with_x

dict

Index corresponds to the amino acid. Order is the same as the order of amino acids in restypes , where '20' represents the unknown amino acid number.

order_restype_with_x

restypes_with_x_and_gap

list

List of amino acid including GAP and UNK . GAP is represented by - . UNK means unknown amino acid and is represented by X. Totally (20 + 1 + 1) types of amino acid are in the list.

restypes_with_x_and_gap

Constants Associated with Atoms in Amino Acids

Name

Type

Description

Value

atom_order

dict

The name of all atoms of the amino acid and their index. Key is the atom name and value is its index.

atom_order

atom_type_num

scalar

The total number of atoms in an amino acid.

37

atom_types

list

List of names of all atoms of amino acid, including 37 types of atom.

atom_types

Van der Waals Atomic Radius

Name

Type

Description

Value

van_der_waals_radius

dict

Van der Waals radius of atom C, atom N , atom O and atom S .

van_der_waals_radius

Constants Associated with torsion angle

Name

Type

Description

Value

chi_angles_atoms

dict

The name of the atomic group that determines the side chain torsion angle in amino acids. Each amino acid has a maximum of four side chain torsion angles, among which ALA and GLY have no side chain torsion angle. Each torsion angle is determined by the position of 4 atomic groups (Dihedral Angle).

chi_angles_atoms

chi_angles_mask

list

If there is a mask of corresponding torsion angle in the amino acid. A two-dimensional array, first dimension is 20, consistent with the order of 'restype_order'. The second dimension has a length of 4, with a value of 1.0 indicating the existence of the torsion angle and 0.0 indicating the absence of the torsion angle.

chi_angles_mask

chi_groups_for_atom

dict

Indexes of all atoms in amino acids in the corresponding side chain torsion angle encoding, where the side chain torsion angle encodings are shown in 'chi_angles_atoms'.

chi_groups_for_atom

chi_pi_periodic

list

Whether torsion angles in amino acids is symmetric invariant, that is, rotating the corresponding torsion angle 180 degrees does not affect the overall three-dimensional structure.

chi_pi_periodic

Constants Associated with Local Coordinate System in Amino Acids

Name

Type

Description

Value

restype_atom14_rigid_group_positions

array

Densely encoding 21 amino acid types. Each amino acid contains the coordinates of the atoms in the local coordinate system of the rigid group to which it belongs. Shape is \((21, 14, 3)\) .

restype_atom14_rigid_group_positions

restype_atom14_to_rigid_group

array

Densely encoding 21 types of amino acid. Each amino acid contains the rigid group the atoms belong to. Shape is \((21, 14)\) .

restype_atom14_to_rigid_group

restype_atom37_rigid_group_positions

array

Sparse encoding 21 amino acid types. Each amino acid contains the coordinates of the atoms in the local coordinate system of the rigid group to which it belongs. Shape is \((21, 37, 3)\) .

restype_atom37_rigid_group_positions

restype_atom37_to_rigid_group

array

Sparse encoding 21 types of amino acid. Each amino acid contains the rigid group the atoms belong to. Shape is \((21, 37)\) .

restype_atom37_to_rigid_group

restype_rigid_group_default_frame

array

Affine transformation matrix of each rigid group of 21 amino acids, that is, the coordinate transformation matrix from the local coordinate system of each rigid group to the local coordinate system of the previous adjacent rigid group. Shape is \((21, 8, 4, 4)\) .

restype_rigid_group_default_frame

rigid_group_atom_positions

dict

The coordinates of the atoms in each amino acid in the local coordinate system of rigid group to which the atom belongs.

rigid_group_atom_positions