Cheminformatics Introduction
Medicinal chemistry is the strategic design, synthesis, and refinement of biological active
molecules with the aim of discovering novel drug candidates. 1
Cheminformatics is defined as a technique that enables throughout examination specific and
characteristic properties of a natural product and provides a comparison with an analogue
feature of other compounds or drugs. 2 And the person who is considered one of the early
pioneers and founder of cheminformatics is Frank Brown, who did in, fact, publish his first
paper in this field. 3
Bioinformatics is a tool used to comprehend information form a particular biological system
e.g. differentiation, cell growth etc. 4
Paulien Hogeweg, who is credited for coining the term “bioinformatics”, emphasizes the
field's origins as a response to the evolving epistemic demands within the life sciences,
spurred by key discoveries in molecular biology during the 20th century. 5,6
It’s important to investigate the characteristics of the data distribution and the relationship
between the Journal Impact Factor (JIF) and other bibliometric measures to gauge its
potential a new bibliometric indicator.7
Examples of journals with their JIFs in the three disciplines are BMC bioinformatics (JIF
3.0), Journal of Chemical Information and Modeling (JIF 5.9) and Journal of Medicinal
Chemistry (JIF 7.3) 8,9
,Here are a few cheminformatics briefly described.
ChEMBL, which stands for Chemical Biological Database, is a large, open scale database of
bio-active molecules with drug-like properties. ChEML, originated from European Molecular
Biology Laboratory European Bioinformatics Institute (EMBL-EBI), provides access to
chemical, physicochemical and bioactivity data of small molecules and their target. This data
source is a reliable resource for drug discovery and development, validation and application
of artificial intelligence and other silico methods 10
PubChem is a public data search engine for chemical molecules and biological activities,
maintained by the National Center of Biotechnology Information of the National Libraries of
Medicine, an institute within the National institute of health (NIH), U.S. Pubchem collects
data from numerous data sources and this data is mostly used in many machine learning
studies. 11
Furthermore, there are other publicly accessible cheminformatics worth mentioning;
DrugBank and NIH Chemical Genomics Center (NCGC) Pharmaceutical Collection (NPC).12
Both are widely used by scientist in drug discovery or pharmacology and chemical biology
research.
While both, ChEMBL and PubChem, are valuable resources they differ in their curation
regarding data quality. Whilst ChEMBL undergoes several manual curations of the data and
on the opposite end PubChem curation is automated, resulting in a mix of curated and
uncurated data.
, Explanations
LogP
The logP value describes the lipophilicity of a compound and is the partition-coefficient of an
un-ionized solute in two immiscible phases (water and n-octanol) in equilibrium.13
However, the distribution of a highly lipophilic or hydrophilic compound is limited to detect
in the two immiscible phases. Lipophilicity has been an important physicochemical property
that describes a load of information about the intermolecular forces and molecular structure
of a compound. For a compound/drug to be orally administrated and have desirable features it
should consist of a logP smaller than 5, exactly like Lipinski’s rule suggested.14
SMILES notation
The simplified molecular input system (SMILES) is widely used in cheminformatics because
of its simplicity and compactness in representing molecular structures. Whilst the smiles
method encodes and represent molecular structure, the LINGO method is used to calculate
similarities and predict the structure-related properties. 15
InChi
InChi, which stands for International Chemical Identifier, is a standardized machine-reading
string of symbols that provides a unique representation of a chemical compound. It allows
computers to represent chemical structures in an unambiguous manner and InChis are like a
barcode.16
Furthermore, InChiKey is a chemical identifier derived from InChi, is 27 characters long and
is a better identifier for exploring the internet and indexing database. 17
Rule-of-5
The rule-of-5 (Ro5) or the influential analysis of physicochemical property space of Lipinski.
This is a set of guidelines used to evaluate the drug likeness of a chemical compound with
respect to their potential oral availability.18 The guidelines to predict adequate absorption are
LogP <5, H-bond donors <5, H-bond acceptors <10, the molecular weight (MW) <500.19
The Ro5 provides medicinal chemist with various insight and principle of the prospective
design of new drugs.18
Medicinal chemistry is the strategic design, synthesis, and refinement of biological active
molecules with the aim of discovering novel drug candidates. 1
Cheminformatics is defined as a technique that enables throughout examination specific and
characteristic properties of a natural product and provides a comparison with an analogue
feature of other compounds or drugs. 2 And the person who is considered one of the early
pioneers and founder of cheminformatics is Frank Brown, who did in, fact, publish his first
paper in this field. 3
Bioinformatics is a tool used to comprehend information form a particular biological system
e.g. differentiation, cell growth etc. 4
Paulien Hogeweg, who is credited for coining the term “bioinformatics”, emphasizes the
field's origins as a response to the evolving epistemic demands within the life sciences,
spurred by key discoveries in molecular biology during the 20th century. 5,6
It’s important to investigate the characteristics of the data distribution and the relationship
between the Journal Impact Factor (JIF) and other bibliometric measures to gauge its
potential a new bibliometric indicator.7
Examples of journals with their JIFs in the three disciplines are BMC bioinformatics (JIF
3.0), Journal of Chemical Information and Modeling (JIF 5.9) and Journal of Medicinal
Chemistry (JIF 7.3) 8,9
,Here are a few cheminformatics briefly described.
ChEMBL, which stands for Chemical Biological Database, is a large, open scale database of
bio-active molecules with drug-like properties. ChEML, originated from European Molecular
Biology Laboratory European Bioinformatics Institute (EMBL-EBI), provides access to
chemical, physicochemical and bioactivity data of small molecules and their target. This data
source is a reliable resource for drug discovery and development, validation and application
of artificial intelligence and other silico methods 10
PubChem is a public data search engine for chemical molecules and biological activities,
maintained by the National Center of Biotechnology Information of the National Libraries of
Medicine, an institute within the National institute of health (NIH), U.S. Pubchem collects
data from numerous data sources and this data is mostly used in many machine learning
studies. 11
Furthermore, there are other publicly accessible cheminformatics worth mentioning;
DrugBank and NIH Chemical Genomics Center (NCGC) Pharmaceutical Collection (NPC).12
Both are widely used by scientist in drug discovery or pharmacology and chemical biology
research.
While both, ChEMBL and PubChem, are valuable resources they differ in their curation
regarding data quality. Whilst ChEMBL undergoes several manual curations of the data and
on the opposite end PubChem curation is automated, resulting in a mix of curated and
uncurated data.
, Explanations
LogP
The logP value describes the lipophilicity of a compound and is the partition-coefficient of an
un-ionized solute in two immiscible phases (water and n-octanol) in equilibrium.13
However, the distribution of a highly lipophilic or hydrophilic compound is limited to detect
in the two immiscible phases. Lipophilicity has been an important physicochemical property
that describes a load of information about the intermolecular forces and molecular structure
of a compound. For a compound/drug to be orally administrated and have desirable features it
should consist of a logP smaller than 5, exactly like Lipinski’s rule suggested.14
SMILES notation
The simplified molecular input system (SMILES) is widely used in cheminformatics because
of its simplicity and compactness in representing molecular structures. Whilst the smiles
method encodes and represent molecular structure, the LINGO method is used to calculate
similarities and predict the structure-related properties. 15
InChi
InChi, which stands for International Chemical Identifier, is a standardized machine-reading
string of symbols that provides a unique representation of a chemical compound. It allows
computers to represent chemical structures in an unambiguous manner and InChis are like a
barcode.16
Furthermore, InChiKey is a chemical identifier derived from InChi, is 27 characters long and
is a better identifier for exploring the internet and indexing database. 17
Rule-of-5
The rule-of-5 (Ro5) or the influential analysis of physicochemical property space of Lipinski.
This is a set of guidelines used to evaluate the drug likeness of a chemical compound with
respect to their potential oral availability.18 The guidelines to predict adequate absorption are
LogP <5, H-bond donors <5, H-bond acceptors <10, the molecular weight (MW) <500.19
The Ro5 provides medicinal chemist with various insight and principle of the prospective
design of new drugs.18
