Abbreviations
QSAR
Quantitative structure–activity relationship
iQSAR
Inverse quantitative structure–activity relationship
vHTS
Virtual high-throughput screening
MIC
Minimum inhibitory concentration
Mtb
Mycobacterium tuberculosis
AAE
Acid alkyl ester
NA
Nucleoside analogue
HIV
Human immunodeficiency virus
MPS
Molecular Priority Score
ARL
Active range length
ARW
Active range weight
ARV
Active range value
MAI
Molecular activity index
IRL
Inactive range length
IRW
Inactive range weight
IRV
Inactive range value
MDI
Molecular de-activity index
SMILES Simplified molecular-input line-entry system
MOL file Molecular structural information file
1 Introduction
Exploring chemical space to discover a compound that elicits a desired pharmacologic response without undesired side effect is like searching a needle in a
haystack problem. The problem arises because we seek to screen a limited subset
that exists among many compounds that elicit a desired pharmacologic response.
Different approaches have therefore evolved to make the problem tractable, namely
effective use of macromolecular target information, if available, use synthesis
tractability of the compounds as guidance, and most importantly, the pharmacological relevance of the compounds selected. While modern advances like targeted
library search or chemogenomics have helped in bringing focus to the drug candidate search, the utility of drug candidate search using serendipity-based approaches has not diminished in face of increasing burden of drug resistance and adverse
side effects. These problems may possibly be addressed by discovering novel
compounds using new drug discovery methods. One of such a new line of thinking
has been proposed by Ruddigkeit et al. [1] who have considered all possible
compounds having 17 atoms taken from C, N, O, S and halogens to create a
database of several billions of compounds. It is tempting to believe that such an
effort of discovering novel drug molecules from such a huge collection of compounds can be useful. However, a method that enables searching of potential drug
72
Md.I. H. Rizvi et al.
QSAR
Quantitative structure–activity relationship
iQSAR
Inverse quantitative structure–activity relationship
vHTS
Virtual high-throughput screening
MIC
Minimum inhibitory concentration
Mtb
Mycobacterium tuberculosis
AAE
Acid alkyl ester
NA
Nucleoside analogue
HIV
Human immunodeficiency virus
MPS
Molecular Priority Score
ARL
Active range length
ARW
Active range weight
ARV
Active range value
MAI
Molecular activity index
IRL
Inactive range length
IRW
Inactive range weight
IRV
Inactive range value
MDI
Molecular de-activity index
SMILES Simplified molecular-input line-entry system
MOL file Molecular structural information file
1 Introduction
Exploring chemical space to discover a compound that elicits a desired pharmacologic response without undesired side effect is like searching a needle in a
haystack problem. The problem arises because we seek to screen a limited subset
that exists among many compounds that elicit a desired pharmacologic response.
Different approaches have therefore evolved to make the problem tractable, namely
effective use of macromolecular target information, if available, use synthesis
tractability of the compounds as guidance, and most importantly, the pharmacological relevance of the compounds selected. While modern advances like targeted
library search or chemogenomics have helped in bringing focus to the drug candidate search, the utility of drug candidate search using serendipity-based approaches has not diminished in face of increasing burden of drug resistance and adverse
side effects. These problems may possibly be addressed by discovering novel
compounds using new drug discovery methods. One of such a new line of thinking
has been proposed by Ruddigkeit et al. [1] who have considered all possible
compounds having 17 atoms taken from C, N, O, S and halogens to create a
database of several billions of compounds. It is tempting to believe that such an
effort of discovering novel drug molecules from such a huge collection of compounds can be useful. However, a method that enables searching of potential drug
72
Md.I. H. Rizvi et al.