Homology groups in Markush structures

Currently JChem enables the searching of Markush structures containing homology groups only with specific molecule queries (with no query features). Homology groups are supported only on the target side; query side support will be implemented in a future version, as well as properties of the homology groups constraining the possible structures.

Definition of the groups

The homology groups in Markush structures are represented by pseudo atoms labelled with the common chemical annotation of the groups. Most groups have alias names which allow shorter names. The names are case insensitive, spaces might be inserted. The pseudo atoms can be most easily drawn in Marvin Sketch using the Homology Groups template group.

There are two major types of homology groups regarding their way of definition:
  1. Built-in groups are defined by specific structural properties of the group. These groups are not enumerated during searching, but the query structure is recognized as fulfilling the requirements for such a structure. The possible number of covered structures is usually infinite, unless the number of atoms is limited. Examples of built-in groups are alkyl, aryl, heterocycle, etc.
  2. User-defined groups are explicitly defined and only the listed structures can match these homology groups. The definition is given in the form of an R-group definition, and any of the generic features discussed in the Markush chapter can be used in the definition. These definitions can be customized by the user, and may be context-specific. (E.g. protecting group definition depends on which functional group it is protecting.)

Built-in groups

Table 1. shows the properties of built-in groups. Built in groups are recognized some specific features, these are shown in the table as "compulsory" parts. Optional parts are structures that are not necessary can be part of the structure that matches on the homology group. "Incomplete case" means structures that are substructures of structures that would represent a homology group. It is required that even such incomplete structures can be extended to a complete homology group. Search options regulate if incomplete structures can match on a homology group or complete structures are required.

The "Example" column shows complete structures representing the homology groups.

Table 1. Built-in groups

Group name (alias names) Compulsory Optional Incomplete case Example
Alkyl
(chk)		 - minimum of one carbon atom
- only carbon and hydrogen atoms
- single bonds
- no ring bonds		 substitution at arbitrary position(s) same requirements
Alkenyl
(che)		 - at least one double bond
- minimum of 2 carbon atoms
- otherwise same as for Alkyl		 same as above same as at compulsory, but the matching structure does not need to have any double bond
Alkynyl
(chy)		 - at least one triple bond
- minimum of 2 carbon atoms
- otherwise same as for Alkyl		 same as above, double bond same as at compulsory, but the matching structure does not need to have any triple bond
Cycloalkyl
(cyc)		 - monocyclic or fused aliphatic rings
- only carbon and hydrogen atoms		 - substitution by (saturated) alkyl chains
- double or triple bonds in the ring but not aromatic
- several connection points but all must be on a ring (can't have external connection on an alkyl chain)		 - any carbon structure without aromatic bonds
- the substituting alkyl chain can be unsaturated
Aryl
(ary)		 - monocyclic or fused rings
- among this rings at least one should be aromatic
- only carbon and hydrogen atoms		 - substitution by (saturated) alkyl chains
- double bonds/triple bonds in the aliphatic rings
- several connection points but all must be on an aromatic ring (can't have external connection on an alkyl chain or on an aliphatic ring)		 - similar to cycloalkyl but the atoms can have aromatic bonds:
- any carbon structure where the external connection is on an atom that has aromatic bond or has only one bond.
- the matching structure doesn't need to have a ring.
AliphaticHeterocyclyl
(het, heterocycle, heterocyclyl)		 - similar to a cycloalkyl but the ring(system) should contain at least one hetero atom
- no hetero atoms outside the rings		 same as cycloalkyl similar to cycloalkyl but here hetero atoms are accepted as well, which means any structure without aromatic bonds
Heteroaryl
(hea)		 - similar to aryl but the aromatic ring(system) should contain at least one hetero atom.
- no hetero atoms outside the rings		 same as aryl Similar to aryl but here hetero atoms are accepted as well. Condition for the externally connecting atom holds as in case of aryl.
FusedHetero
(hef)		 - Fused ringsystem having at least one hetero atom. same as aryl Any structure having hetero, carbon and hydrogen atoms, with any bonds.
Unknown group
(unk)		 - Any structure. Unknown structures are enumerated as the union of all other homology groups. - -
Metal
(mx)		 Any metal - - U, K, Fe, Na, Ni, Al
AlkaliMetal
(amx)		 Alkali and alkaline earth metals - - Na, K, Ca, Mg
OtherMetal
(a35)		 Group IIIa-Va metals - - Al, Ga
TransitionMetal
(trm)		 transition metals excluding Lanthanum - - Fe, Ni, Zn, Co, Hg, W
Lanthanide
(lan)		 lanthanides (including Lanthanum) - - Nd, Ce, Pr
Actinide
(act)		 actinides (including Actinium) - - U, Th, Pa

User-defined homology groups

These homology groups are predefined and are represented by R-group definitions. During search the pseudo atoms are translated to the corresponding R-group definitions.

The group definitions are customizable, the user can override them or can make new definitions as well. Group names are treated case insensitive, but in case sensitive file systems the definition files should be lowercase.

User defined groups readily available in the system are the following:

Halogen

Halogen elements: F, Cl, I and Br.
JChem's group name: halogen
alias names: hal, x

Protecting

Protecting groups' definition file contains several definitions, each for protecting different functional groups. The protected functional group is defined by the neighbourhood of the R-atom. When the R-atom has the same neighbourhood as the "protecting" pseudo atom, then the group is replaced by the R-atom.

The conversion processes the group definitions in their order in the file. This means that more specific environments should be placed earlier. For example, a carboxyl protecting group definition should precede an alcohol definition, otherwise the alcohol definitions will be applied instead. Currently they are located in the following order:

  1. amino
  2. carboxyl
  3. alcohol

Currently the system can't handle protecting groups having more than one attachment point, or groups where the heavy atoms of the functional group should be changed by the substitution. The readily available definitions contain amine, carboxyl and hydroxyl protecting groups.
JChem's group name: protecting
alias names: prt Some examples with different functional groups protected can be found on Table 2.

Table 2. Protecting group examples

Protecting group Represented examples

Acyl

Residue left after removal of one or more OH groups from an acid. Currently it behaves as simple pseudo atoms: can only be matched by itself and is not enumerated. This behaviour complies with the Thomson-Reuters/Questel acyl group handling.
JChem's group name: acyl
alias names: acy

Any group

The union of all other homology groups except acyl, unknown and protecting.
JChem's group name: any
alias names: xx

Search options and properties

Search options regulating the search behaviour are also available:

Search Options

Currently there is one regulating option: 'completeHG', which specifies if the part of the query side structure matching on the given group should represent an entire homology group or if substructures are also accepted. Of course in the incomplete case an entire structure can also match on the given homology group.
For example, if completeHG is set to true (default) an alkyl chain can't match on a cycloalkyl group, only a ring (system). The detailed behaviour is found at the definition of the groups. And example is shown on Table 3.

Table 3. Complete and incomplete structures of homology groups

target query hit
completeHG:y completeHG:n

Enumeration

To enable the enumeration of homology groups, the homology enumeration option of Markush enumeration has to be switched on. Otherwise the homology groups are kept as pseudo atoms. This latter option might be useful for showing that these structures can't be fully enumerated.

Built-in groups

For the built-in types, example R-group definitions specify the enumerable library, with the same technology as user-defined groups. These structures are characteristic to the homology group and encompass simple and large structures as well. The group definitions are customizable.

We have to emphasize, that these definitions are only used for enumeration and do not affect searching. As noted earlier, arbitrary structures fulfilling the requirements for the homology group will match such a target.

The enumeration definitions contain as default two attachment points. After enumeration these are the atoms which connect to the first two neighbours of the group. If the enumerated homology pseudo atom has more than two connections then further attachment points are added. These are put on atoms that have free valence and comply the requirements for externally connecting atoms of the given group. E.g. for aryl only aromatic ring atoms can be connection points of the aryl. The atoms of the definition are investigated in the order of their numbering. If a definition does not have the sufficient number of such atoms, then it is rejected. In case all the definitions of a homology atoms are rejected an exception is thrown showing that the given homology group does not have any suitable enumeration definition.

User-defined groups

The enumeration of user defined homology groups use the same (customizable) R-group definitions as searching. The user-defined homology atom should have the same number of connections as are shown in the definitions.

Customizing rgroup definitions

Location of the definitions:
User defined groups
Directory chemaxon/enumeration/homology/user_def_groups in the jar file. The name implies that these groups are represented by these definitions during search and enumeration as well.
Built-in groups
Directory chemaxon/enumeration/homology/enumeration_only in the jar file. The name implies that these groups are represented by these definitions during enumeration only.
The definition files are in lower case, customized file names have to be in lower case as well. Overriding of these definition occurs by copying the definition from the jar file to the user (.)chemaxon directory into the subdirectory: homology/user_def_groups (or homology/enumeration_only for built-in groups). Overriding these files will affect searching/enumeration in case of user defined groups and the enumeration in case of built-in groups.
Adding a new group can be achieved by creating/copying a new mrv file containing an R atom and the definitions of the group. The name of the new group is specified by the name of the file.
The overridden definition or the newly added group can also be dependent on the neighbourhood (context-sensitive) as in the case of protecting groups.
If the user would like to have different definitions for searching and enumeration of a user-defined group, then a separate file should be specified in the "enumeration_only" dictionary. In this case the content of the "user_def_groups" will be used during searching and the content of the "enumeration_only" for enumeration.
An example for overriding an amino protecting group definition may consist of the following steps:
  1. Copy protecting group definition to the users (.)chemaxon library: e.g. from .../jchem.jar/chemaxon/enumeration/homology/user_def_groups/protecting.mrv to /home/{USER_NAME}/.chemaxon/homology/user_def_groups/
  2. Open the newly copied file in the user's directory with msketch.
  3. A dialog appears asking the index of molecule to open. Enter 1 because this contains the amino protecting group definition. If the proper molecule number is not known, all the definitions can be displayed using mview.
  4. Overwrite the structures, e.g. delete the FMOC group, see Table 4. The new definition will be used in searching and enumeration, see Table 4.
If a definition is overridden it comes into effect immediately, however the addition of a new group requires a restart of the Java Virtual Machine.

Table 4. Overriding amino protecting group definitions.

overwriting the definition sample markush file enumerations

Planned improvements: Properties of homology groups

These will be implemented in a future version. They regulate the properties of the individual homology groups independently. They include the followings (with the group to which it may be applied):
  1. monocyclic/fused, default: both are allowed, for all ring type groups
  2. saturated, default false, for cycloalkyl and heterocycle groups
  3. chain, default false, for alkyl, alkenyl, alkynyl