Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Jan 17;27(2):570.
doi: 10.3390/molecules27020570.

Approach for the Design of Covalent Protein Kinase Inhibitors via Focused Deep Generative Modeling

Atsushi Yoshimori  1 , Filip Miljković  2 , Jürgen Bajorath  2
Affiliations

Approach for the Design of Covalent Protein Kinase Inhibitors via Focused Deep Generative Modeling

Atsushi Yoshimori et al. Molecules. .

Abstract

Deep machine learning is expanding the conceptual framework and capacity of computational compound design, enabling new applications through generative modeling. We have explored the systematic design of covalent protein kinase inhibitors by learning from kinome-relevant chemical space, followed by focusing on an exemplary kinase of interest. Covalent inhibitors experience a renaissance in drug discovery, especially for targeting protein kinases. However, computational design of this class of inhibitors has thus far only been little investigated. To this end, we have devised a computational approach combining fragment-based design and deep generative modeling augmented by three-dimensional pharmacophore screening. This approach is thought to be particularly relevant for medicinal chemistry applications because it combines knowledge-based elements with deep learning and is chemically intuitive. As an exemplary application, we report for Bruton's tyrosine kinase (BTK), a major drug target for the treatment of inflammatory diseases and leukemia, the generation of novel candidate inhibitors with a specific chemically reactive group for covalent modification, requiring only little target-specific compound information to guide the design efforts. Newly generated compounds include known inhibitors and characteristic substructures and many novel candidates, thus lending credence to the computational approach, which is readily applicable to other targets.

Keywords: Bruton’s tyrosine kinase; covalent inhibitors; deep machine learning; generative modeling; kinase inhibitor design.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Structure of ibrutinib. The covalent drug ibrutinib contains an acrylamide warhead, colored magenta in (a), which forms a covalent bond to the thiol group of a Cys481 in BTK, shown in (b).
Figure 2
Figure 2
Key 2, Value 2, and Value 1 fragment assembly. Shown is the DeepSARM fragment composition of (a) ibrutinib and (b) a candidate compound. Key 2, Value 2 (R2), and Value 1 (R1) are displayed in green, magenta, and blue, respectively.
Figure 3
Figure 3
Generation of new Key 2 fragments. (a) The computational workflow for generating Key 2 fragments using the Seq2Seq model (Key 2) and pharmacophore filtering. In the lower left image, pharmacophore features including hydrogen bond donor, hydrogen bond acceptor, and aromatic groups are shown as a green arrow, red arrow, and blue circle, respectively. (b) The structure of 18 newly generated Key 2 fragments. Below each structure, the identification number and log-likelihood score (in parentheses) are provided.
Figure 3
Figure 3
Generation of new Key 2 fragments. (a) The computational workflow for generating Key 2 fragments using the Seq2Seq model (Key 2) and pharmacophore filtering. In the lower left image, pharmacophore features including hydrogen bond donor, hydrogen bond acceptor, and aromatic groups are shown as a green arrow, red arrow, and blue circle, respectively. (b) The structure of 18 newly generated Key 2 fragments. Below each structure, the identification number and log-likelihood score (in parentheses) are provided.
Figure 4
Figure 4
Generation of Value 2 fragments. (a) Workflow for generating Value 2 fragments using the Seq2Seq model (Value 2) and warhead filtering, and (b) new Value 2 fragments. Below each structure, the identification number and log-likelihood score (in parentheses) are provided. * indicates the fragment attachment point.
Figure 5
Figure 5
[Value 2 × Value 1] matrix for each of the 18 prioritized Key 2 fragments. The matrix cells represented unique ([Key 2 − Value 2] − Value 1) combinations (candidate compounds) color-coded by cumulative log-likelihood scores.
Figure 6
Figure 6
Hypothetical complexes of candidate inhibitors from DeepSARM and BTK. (a) Superposition of a candidate inhibitor containing Key 2-21 onto the crystallographic binding mode of ibrutinib. Pharmacophore features including two hydrogen bond acceptors, one hydrogen bond donor, one residue bonding point, and two optional hydrophobic features are represented as red arrows, green arrow, and orange/yellow sphere, respectively. (b) A corresponding diagram of candidate inhibitor–BTK interactions is shown.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Bajorath J., Kearnes S., Walters W.P., Meanwell N.A., Georg G.I., Wang S. Artificial Intelligence in Drug Discovery: Into the Great Wide Open. J. Med. Chem. 2020;63:8651–8652. doi: 10.1021/acs.jmedchem.0c01077. - DOI - PubMed
    1. Segler M.H.S., Kogej T., Tyrchan C., Waller M.P. Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks. ACS Cent. Sci. 2018;4:120–131. doi: 10.1021/acscentsci.7b00512. - DOI - PMC - PubMed
    1. Skalic M., Jiménez J., Sabbadin D., De Fabritis G. Shape-based Generative Modeling for De Novo Drug Design. J. Chem. Inf. Model. 2019;59:1205–1214. doi: 10.1021/acs.jcim.8b00706. - DOI - PubMed
    1. Blaschke T., Arús-Pous J., Chen H., Margreitter C., Tyrchan C., Engkvist O., Papadopoulos K., Patronov A. REINVENT 2.0: An AI Tool for De Novo Drug Design. J. Chem. Inf. Model. 2020;60:5918–5922. doi: 10.1021/acs.jcim.0c00915. - DOI - PubMed
    1. Singh J., Petter R.C., Baillie T.A., Whitty A. The Resurgence of Covalent Drugs. Nat. Rev. Drug Discov. 2011;10:307–317. doi: 10.1038/nrd3410. - DOI - PubMed

MeSH terms

Substances