Tool: Add Hydrogens

Add Hydrogens adds hydrogen atoms to atomic models, as well as OXT atoms where missing from peptide C-termini. The corresponding command is addh. See also: H-Bonds, Build Structure, Add Charges, Check Waters, Dock Prep

Add Hydrogens can be opened from the Structure Editing section of the Tools menu and manipulated like other panels (more...).

• Consider each model in isolation from all others – whether hydrogen placement should be affected by atoms within the same model only. Otherwise, other models in the vicinity (except submodels of the same model) may affect hydrogen placement, regardless of whether they were chosen for hydrogen addition.

The Method for adding hydrogens can be:

• steric only – based on atom types and clash avoidance
• also consider H-bonds (slower) (default) – based on atom types, clash avoidance, and hydrogen bond formation. Considering H-bonds increases the calculation time. Although hydrogens are placed to avoid clashes and form hydrogen bonds where possible, they are not energy-minimized, and a globally optimal network in terms of the number of H-bonds or total H-bonding energy is not necessarily found.

Clicking the Options button shows/hides additional options:

Protonation states: By default, if amino acids have standard residue names, each histidine sidechain will be protonated based on its local environment, whereas the sidechains of other residue types will be assigned protonation states reasonable at physiological pH, regardless of the local environment: negative aspartic acid and glutamic acid, positive arginine and lysine, and neutral cysteine and tyrosine. However, alternative protonation states of certain amino acids can be specified by changing their residue names beforehand (with setattr or by text-editing the input structure file):

• Choices for histidine:
  • Residue-name-based (default) – residue names will be used to determine which histidine sidechain nitrogens should be protonated: the δ-nitrogen in residues named HID, the ε-nitrogen in HIE, and both nitrogens in HIP. Residues named HIS will be treated as unspecified, and may end up with either or both sidechain nitrogens protonated, depending on the method and the local environment.
  • Unspecified (determined by method) – regardless of which of the names above are used for histidine residues, all will be treated as unspecified, and may end up with either or both sidechain nitrogens protonated, depending on the method and the local environment.
• Choices for aspartic acid:
  • Residue-name-based (default) – residue names will be used to determine sidechain charge state: ASP negatively charged and ASH neutral, OD2-protonated
  • Charged – negatively charged
• Choices for cysteine:
  • Residue-name-based (default) – residue names will be used to determine sidechain charge state: CYS unspecified and CYM negatively charged
  • Unspecified (determined by method) – regardless of which of the names above are used for cysteine residues, all will be treated as unspecified and the sidechain sulfur may be protonated or not depending on the method and the local environment.
• Choices for glutamic acid:
  • Residue-name-based (default) – residue names will be used to determine sidechain charge state: GLU negatively charged and GLH neutral, OE2-protonated
  • Charged – negatively charged
• Choices for lysine:
  • Residue-name-based (default) – residue names will be used to determine sidechain charge state: LYS positively charged and LYN neutral
  • Charged – positively charged

• Do not protonate electronegative atom X within [distance] Å of metal M if X-H-M angle would be >120° (default 3.95 Å)
• Use idealized template to guess atom types in nonstandard residues (default off) – whether to use idealized coordinates to discern atom types in nonstandard residues, instead of their actual coordinates in the structure. The idealized coordinates are looked up by residue name in the PDB Chemical Component Dictionary. This option is useful for structures with poorly (structurally) determined ligands that could be mistyped, potentially causing addition of the wrong number of hydrogens or hydrogens in incorrect positions.

Clicking OK adds hydrogens and dismisses the dialog, while Cancel merely dismisses the dialog. Clicking Help shows this page in the Help Viewer.

The positions of pre-existing atoms are not changed, but any lone pairs and atoms of unidentifiable element are deleted. Hydrogens are named in accordance with the PDB standard, if possible, and positioned to produce the expected bond lengths. An attempt is made to preserve coloring schemes, coloring the added hydrogens by element (white) and/or to match their bond partners.

The default VDW radii of carbon, nitrogen, oxygen, and sulfur atoms depend on whether hydrogen atoms are present; thus, adding hydrogens will change the VDW radii of some atoms.

[back to top: Add Hydrogens]

Protonation States

Add Hydrogens aims to generate protonation states reasonable for pH ~7 (such as in blood). For example, hydrogens are not added to the phosphodiester backbone of DNA and RNA, and in proteins, the default is to make aspartic acid and glutamic acid sidechains negatively charged and arginine and lysine sidechains positively charged. The following groups typically have pKa values near 7, however, so how they get protonated depends on the details of the structure:

• imidazoles such as histidine sidechains; histidine protonation states can be guessed by the method or specified by the user (by changing residue names beforehand with setattr or by text-editing the input structure file)
• terminal phosphates (the third ionization); if one P–O bond is at least 0.05 Å longer than the others around that same phosphorus atom, that oxygen will be protonated

Potentially ambiguous or rare (shifted-pKa) protonation states, especially in binding sites and nonstandard residues, should be verified and corrected as needed:

• hydrogens can be deleted
• unusual protonation states of certain amino acids can be specified by changing their residue names beforehand (see options)
• atom types in nonstandard residues can be edited (also before hydrogen addition) with setattr or Build Structure

Residues at the ends of connected peptide chains are inspected to determine whether they are real termini, based on any SEQRES information in the input PDB file (or the mmCIF equivalent) and the presence or absence of additional chains with the same IDs. Real N-termini are assumed to be positively charged (⁺H₃N–) and real C-termini are assumed to be negatively charged (–CO₂^–). If a C-terminal carboxylate is missing an oxygen (OXT), it will be added. End residues that are not real termini are terminated like other chain-internal residues, with N(H)– and –C(=O). The position of the N-end “amide” hydrogen in such cases is not fully determined by the positions of the existing atoms; Add Hydrogens places this hydrogen to produce a φ angle equal to that of the subsequent residue.

[back to top: Add Hydrogens]

Bond Lengths

Bond lengths for X-H (X = C/N/O/S) are taken from the Amber parm99 parameters:

X	atom types	X-H bond length (Å)
sp3 carbon	C3	1.0900
sp2 carbon	C2,Car	1.0800
sp carbon	C1	1.0560
nitrogen	N3+,N3,Npl,Ng+	1.0100
sp3 oxygen	O3 (except water)	0.9600
sp3 oxygen	O3 (water)	0.9572
sulfur	S3	1.3360

Bond lengths to other X are approximate, obtained by adding the covalent bond radii of elements X and H.

[back to top: Add Hydrogens]

Recommended Alternative: Reduce

When a more intensive approach is desired, the program Reduce provided as part of MolProbity is a good alternative. It places hydrogens to optimize local H-bonding networks and avoid steric overlaps, while flipping certain sidechains 180° as deemed appropriate to fulfill these criteria. Asparagine and glutamine sidechains may be flipped to switch their terminal N and O atoms, and the imidazole ring of histidine may be flipped to switch N and C identities. The protonation state of histidine is adjusted based on the local environment. The method is described in:

Asparagine and glutamine: using hydrogen atom contacts in the choice of side-chain amide orientation. Word JM, Lovell SC, Richardson JS, Richardson DC. J Mol Biol. 1999 Jan 29;285(4):1735-47.