Addressing the Metabolic Stability of Antituberculars through Machine Learning

Thomas P. Stratton, Alexander L. Perryman, Catherine Vilchèze, Riccardo Russo, Shao Gang Li, Jimmy S. Patel, Eric Singleton, Sean Ekins, Nancy Connell, William R. Jacobs, Joel S. Freundlich

Research output: Contribution to journalArticle

Abstract

We present the first prospective application of our mouse liver microsomal (MLM) stability Bayesian model. CD117, an antitubercular thienopyrimidine tool compound that suffers from metabolic instability (MLM t1/2 < 1 min), was utilized to assess the predictive power of our new MLM stability model. The S-substituent was removed, a set of commercial reagents was utilized to construct a virtual library of 411 analogues, and our MLM stability model was applied to prioritize 13 analogues for synthesis and biological profiling. In MLM stability assays, all 13 analogues had superior metabolic stability to the parent compound, and six new analogues had acceptable MLM t1/2 values greater than or equal to 60 min. It is noteworthy that whole-cell efficacy and lack of relative mammalian cell cytotoxicity could not be predicted simultaneously. These results support the utility of our new MLM stability model in chemical tool and drug discovery optimization efforts.

Original languageEnglish (US)
Pages (from-to)1099-1104
Number of pages6
JournalACS Medicinal Chemistry Letters
Volume8
Issue number10
DOIs
StatePublished - Oct 12 2017
Externally publishedYes

Fingerprint

Liver
Learning systems
Digital Libraries
Chemical Models
Drug Discovery
Cytotoxicity
Machine Learning
Assays
Cells

Keywords

  • antitubercular
  • Bayesian
  • chemical tool optimization
  • computer-aided analogue design
  • machine learning
  • mouse liver microsomal stability

ASJC Scopus subject areas

  • Biochemistry
  • Drug Discovery
  • Organic Chemistry

Cite this

Stratton, T. P., Perryman, A. L., Vilchèze, C., Russo, R., Li, S. G., Patel, J. S., ... Freundlich, J. S. (2017). Addressing the Metabolic Stability of Antituberculars through Machine Learning. ACS Medicinal Chemistry Letters, 8(10), 1099-1104. https://doi.org/10.1021/acsmedchemlett.7b00299

Addressing the Metabolic Stability of Antituberculars through Machine Learning. / Stratton, Thomas P.; Perryman, Alexander L.; Vilchèze, Catherine; Russo, Riccardo; Li, Shao Gang; Patel, Jimmy S.; Singleton, Eric; Ekins, Sean; Connell, Nancy; Jacobs, William R.; Freundlich, Joel S.

In: ACS Medicinal Chemistry Letters, Vol. 8, No. 10, 12.10.2017, p. 1099-1104.

Research output: Contribution to journalArticle

Stratton, TP, Perryman, AL, Vilchèze, C, Russo, R, Li, SG, Patel, JS, Singleton, E, Ekins, S, Connell, N, Jacobs, WR & Freundlich, JS 2017, 'Addressing the Metabolic Stability of Antituberculars through Machine Learning', ACS Medicinal Chemistry Letters, vol. 8, no. 10, pp. 1099-1104. https://doi.org/10.1021/acsmedchemlett.7b00299
Stratton TP, Perryman AL, Vilchèze C, Russo R, Li SG, Patel JS et al. Addressing the Metabolic Stability of Antituberculars through Machine Learning. ACS Medicinal Chemistry Letters. 2017 Oct 12;8(10):1099-1104. https://doi.org/10.1021/acsmedchemlett.7b00299
Stratton, Thomas P. ; Perryman, Alexander L. ; Vilchèze, Catherine ; Russo, Riccardo ; Li, Shao Gang ; Patel, Jimmy S. ; Singleton, Eric ; Ekins, Sean ; Connell, Nancy ; Jacobs, William R. ; Freundlich, Joel S. / Addressing the Metabolic Stability of Antituberculars through Machine Learning. In: ACS Medicinal Chemistry Letters. 2017 ; Vol. 8, No. 10. pp. 1099-1104.
@article{4e2e2c8f1d29412caa0579278aafbc13,
title = "Addressing the Metabolic Stability of Antituberculars through Machine Learning",
abstract = "We present the first prospective application of our mouse liver microsomal (MLM) stability Bayesian model. CD117, an antitubercular thienopyrimidine tool compound that suffers from metabolic instability (MLM t1/2 < 1 min), was utilized to assess the predictive power of our new MLM stability model. The S-substituent was removed, a set of commercial reagents was utilized to construct a virtual library of 411 analogues, and our MLM stability model was applied to prioritize 13 analogues for synthesis and biological profiling. In MLM stability assays, all 13 analogues had superior metabolic stability to the parent compound, and six new analogues had acceptable MLM t1/2 values greater than or equal to 60 min. It is noteworthy that whole-cell efficacy and lack of relative mammalian cell cytotoxicity could not be predicted simultaneously. These results support the utility of our new MLM stability model in chemical tool and drug discovery optimization efforts.",
keywords = "antitubercular, Bayesian, chemical tool optimization, computer-aided analogue design, machine learning, mouse liver microsomal stability",
author = "Stratton, {Thomas P.} and Perryman, {Alexander L.} and Catherine Vilch{\`e}ze and Riccardo Russo and Li, {Shao Gang} and Patel, {Jimmy S.} and Eric Singleton and Sean Ekins and Nancy Connell and Jacobs, {William R.} and Freundlich, {Joel S.}",
year = "2017",
month = "10",
day = "12",
doi = "10.1021/acsmedchemlett.7b00299",
language = "English (US)",
volume = "8",
pages = "1099--1104",
journal = "ACS Medicinal Chemistry Letters",
issn = "1948-5875",
publisher = "American Chemical Society",
number = "10",

}

TY - JOUR

T1 - Addressing the Metabolic Stability of Antituberculars through Machine Learning

AU - Stratton, Thomas P.

AU - Perryman, Alexander L.

AU - Vilchèze, Catherine

AU - Russo, Riccardo

AU - Li, Shao Gang

AU - Patel, Jimmy S.

AU - Singleton, Eric

AU - Ekins, Sean

AU - Connell, Nancy

AU - Jacobs, William R.

AU - Freundlich, Joel S.

PY - 2017/10/12

Y1 - 2017/10/12

N2 - We present the first prospective application of our mouse liver microsomal (MLM) stability Bayesian model. CD117, an antitubercular thienopyrimidine tool compound that suffers from metabolic instability (MLM t1/2 < 1 min), was utilized to assess the predictive power of our new MLM stability model. The S-substituent was removed, a set of commercial reagents was utilized to construct a virtual library of 411 analogues, and our MLM stability model was applied to prioritize 13 analogues for synthesis and biological profiling. In MLM stability assays, all 13 analogues had superior metabolic stability to the parent compound, and six new analogues had acceptable MLM t1/2 values greater than or equal to 60 min. It is noteworthy that whole-cell efficacy and lack of relative mammalian cell cytotoxicity could not be predicted simultaneously. These results support the utility of our new MLM stability model in chemical tool and drug discovery optimization efforts.

AB - We present the first prospective application of our mouse liver microsomal (MLM) stability Bayesian model. CD117, an antitubercular thienopyrimidine tool compound that suffers from metabolic instability (MLM t1/2 < 1 min), was utilized to assess the predictive power of our new MLM stability model. The S-substituent was removed, a set of commercial reagents was utilized to construct a virtual library of 411 analogues, and our MLM stability model was applied to prioritize 13 analogues for synthesis and biological profiling. In MLM stability assays, all 13 analogues had superior metabolic stability to the parent compound, and six new analogues had acceptable MLM t1/2 values greater than or equal to 60 min. It is noteworthy that whole-cell efficacy and lack of relative mammalian cell cytotoxicity could not be predicted simultaneously. These results support the utility of our new MLM stability model in chemical tool and drug discovery optimization efforts.

KW - antitubercular

KW - Bayesian

KW - chemical tool optimization

KW - computer-aided analogue design

KW - machine learning

KW - mouse liver microsomal stability

UR - http://www.scopus.com/inward/record.url?scp=85031311169&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031311169&partnerID=8YFLogxK

U2 - 10.1021/acsmedchemlett.7b00299

DO - 10.1021/acsmedchemlett.7b00299

M3 - Article

AN - SCOPUS:85031311169

VL - 8

SP - 1099

EP - 1104

JO - ACS Medicinal Chemistry Letters

JF - ACS Medicinal Chemistry Letters

SN - 1948-5875

IS - 10

ER -