Glossary

candidate morphs

Morphs generated from the leaves of an exploration tree, but not yet attached to it.

chemical operators

Also called morphing operators, are a set of predefined structural modifications that a molecule can undergo during molecular morphing.

chemical space

Chemical space has many definitions, but in this documentation we define it as space spanned by all possible (i.e. energetically stable) molecules and chemical compounds [1].

chemical space path

A consecutive sequence of morphs created by iteratively applying a chemical operator on leaves in the exploration tree.

exploration parameters

Those define how the exploration tree behaves (i.e. what similarity measure to use and what fingerprint, how the candidate morphs are filtered or how often the tree is pruned). See the description of the ExplorationData class for details.

exploration tree

This is the data structure Molpher generates as it progresses through chemical space. It is represented in the API as an instance of the ExplorationTree class.

high-throughput screening

High-throughput screening (HTS) is a specialized method of chemical biology which relies on robotics to conduct many experiments in a very short time (usually in the order of tens of thousands a day) with the goal to find compounds that have the potential to affect a biological target and, thus, could become leads in a drug discovery project.

molecular fingerprint

Molecular fingerprints are binary strings that encode certain structural features of a molecule.

molecular morphing

Molecular morphing is an atom-based de novo method of computer-aided drug design and it was first implemented in the Molpher program [2]. It uses a set of chemical operators that modify structures of compounds to ‘travel’ through chemical space and sample certain biologically interesting areas.

Molpher

Software developed as a collaboration between the SIRET Research Group (Faculty of Mathematics and Physics, Charles University in Prague,), Laboratory of Informatics and Chemistry (UCT Prague) and CZ-OPENSCREEN [2]. The main goal of the project is to implement an effective tool for chemical space exploration with molecular morphing. Molpher itself is a standalone program and is available from GitHub.

morph

Any molecule generated by Molpher or Molpher-lib.

morph generation

The morphs attached to the tree upon committing a morphing iteration.

morphing iteration

Any sequence of operations that finishes with attaching new generation of morphs to a tree. For example, an iteration is committed when ExplorationTree.extend() is called.

morphing parameters

A set of restrictions and rules the exploration tree follows when certain operations are performed.

non-producing molecule

A molecule that has not produced any morphs that would have a value of the objective function lower than itself.

SAScore

Available from molpher.core.MolpherMol instances via the sascore parameter. It is the synthetic availability score according to Ertl et al. [3]. The higher the score, the more difficult the compound should be to synthesize. Compounds with SAScore of more than 6 are considered synthetically inaccessible and are discarded by the FilterMorphsOper operation and filterMorphs() method by default.

SAScore.dat

The data file used for computation of the synthetic feasibility scores.

selectors

Located in the selectors module, these serve the purpose of selecting various options for the morphing algorithm. See ExplorationData for more information on how to affect the way morphs are generated.

similarity measure

A similarity measure measures similarity between two molecules using a molecular fingerprint.

source molecule

The root of the exploration tree. All morphs in the tree are descendants of this molecule.

target molecule

Target molecule of the whole exploration process. The algorithm attempts to generate a path in chemical space leading from the source molecule to this one.

target molecule

A molecule the chemical space exploration is driven towards. Each morph is evaluated according to its distance (using a molecular fingerprint and a similarity measure) from this molecule.

tree operation

Any operation that manipulates the exploration tree. Can be defined by implementing the molpher.core.operations.TreeOperation interface.

tree snapshot

A tree snapshot is a file on disk that contains all the information needed to create an exploration tree instance. It can be saved and loaded using the save() and create() methods.

virtual screening

Virtual screening (VS) can be regarded as a computational alternative to HTS. The goal of virtual screening is the same: to find bioactive compounds, but in this case the compounds are represented in silico and saved in a database. Various computational techniques can then be used to automatically probe the database for structures that could have biological activity in the real world. Therefore, the point of VS is to create plausible hypotheses about the usefulness of the compounds in the database and discard all that very likely do not satisfy the requirements for a bioactive compounds in a given project.

XML template

A file in XML format that can be used as a configuration file and is loaded the same way as a tree snapshot (using the create() method).

[1]	https://en.wikipedia.org/wiki/Chemical_space

[2]	(1, 2) Hoksza, P. Škoda, M. Voršilák, and D. Svozil, “Molpher: a software framework for systematic chemical space exploration,” Journal of Cheminformatics, vol. 6, no. 1, p. 7, Mar. 2014. DOI: 10.1186/1758-2946-6-7

[3]	Peter Ertl, P & Schuffenhauer, Ansgar. (2009). Estimation of Synthetic Accessibility Score of Drug-Like Molecules Based on Molecular Complexity and Fragment Contributions. Journal of cheminformatics. 1. 8. DOI:10.1186/1758-2946-1-8.