CoChat
Sign Up
Cited by 118Computational Drug Discovery MethodsMachine Learning in Materials ScienceSARS-CoV-2 and COVID-19 Research

The DOI matches its Crossref record; 57 of the 58 references it cites were found in a registry or an index.

De Novo Design of New Chemical Entities for SARS-CoV-2 Using Artificial Intelligence

Open access
Bung, Navneet, Krishnan, Sowmya R, Bulusu, Gopalakrishnan, Roy, Arijit·2021·Future Medicinal Chemistry
DOI 10.4155/fmc-2020-0262PMC PMC7888348

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Background: The novel coronavirus SARS-CoV-2 has severely affected the health and economy of several countries. Multiple studies are in progress to design novel therapeutics against the potential target proteins in SARS-CoV-2, including 3CL protease, an essential protein for virus replication. Materials & methods: In this study we employed deep neural network-based generative and predictive models for de novo design of small molecules capable of inhibiting the 3CL protease. The generative model was optimized using transfer learning and reinforcement learning to focus around the chemical space corresponding to the protease inhibitors. Multiple physicochemical property filters and virtual screening score were used for the final screening. Conclusion: We have identified 33 potential compounds as ideal candidates for further synthesis and testing against SARS-CoV-2.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.4155/fmc-2020-0262

Cites

57 of 57 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1. 3D structure of 3CL protease from SARS-CoV-2. The active site residues His41 and Cys145 which are crucial for the catalytic process of 3CL protease are shown in magenta sticks.
Figure 1. 3D structure of 3CL protease from SARS-CoV-2. The active site residues His41 and Cys145 which are crucial for the catalytic process of 3CL protease are shown in magenta sticks.CC BY · PubMed Central
Figure 2. De novo drug design pipeline for generating small molecules against a target of interest. (A) Pretrained generative model. (B) Transfer learning (TL) to learn the features of small molecules specific to the target protein and reinforcement learning (RL) to optimize the property of interest. (C) Different physico-chemical property filters, structural alerts and virtual screening score were used for the final screening.
Figure 2. De novo drug design pipeline for generating small molecules against a target of interest. (A) Pretrained generative model. (B) Transfer learning (TL) to learn the features of small molecules specific to the target protein and reinforcement learning (RL) to optimize the property of interest. (C) Different physico-chemical property filters, structural alerts and virtual screening score were used for the final screening.CC BY · PubMed Central
Figure 3. New chemical entities similar to the HIV protease inhibitors. Structurally similar new chemical entities (NCEs) with better virtual screening score than existing HIV protease inhibitors (ASC09, darunavir, indinavir and saquinavir) that are in clinical trials against SARS-CoV-2. The virtual screening score and the Tanimoto similarity of the NCEs are shown in blue and black within the bracket. The virtual screening scores of existing HIV protease inhibitors in clinical trials are also shown in blue.
Figure 3. New chemical entities similar to the HIV protease inhibitors. Structurally similar new chemical entities (NCEs) with better virtual screening score than existing HIV protease inhibitors (ASC09, darunavir, indinavir and saquinavir) that are in clinical trials against SARS-CoV-2. The virtual screening score and the Tanimoto similarity of the NCEs are shown in blue and black within the bracket. The virtual screening scores of existing HIV protease inhibitors in clinical trials are also shown in blue.CC BY · PubMed Central
Figure 4. New chemical entities with highest virtual screening score. Generated new chemical entities with highest virtual screening score (shown in blue) against the 3CL protease of SARS-CoV-2. The 3CL protease is shown in surface representation. The active site of the 3CL protease is highlighted using an ellipse.
Figure 4. New chemical entities with highest virtual screening score. Generated new chemical entities with highest virtual screening score (shown in blue) against the 3CL protease of SARS-CoV-2. The 3CL protease is shown in surface representation. The active site of the 3CL protease is highlighted using an ellipse.CC BY · PubMed Central
Figure 5. Comparison of generated new chemical entities and natural products. 2D structures of generated new chemical entities (NCEs, left) and Aurantiamide (right). The similarity between the NCEs and Aurantiamide was quantified using the Tanimoto coefficient.
Figure 5. Comparison of generated new chemical entities and natural products. 2D structures of generated new chemical entities (NCEs, left) and Aurantiamide (right). The similarity between the NCEs and Aurantiamide was quantified using the Tanimoto coefficient.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 58 references

Cited by

Sorted by citations

See all 118

Related

Metadata

Authors
Bung, N., Krishnan, S. R., Bulusu, G., & Roy, A.
Affiliations
Tata Consultancy Services (India) (IN)
Corresponding author
  • Bulusu, Gopalakrishnan
  • Roy, Arijit
ORCID
Venue
Future Medicinal Chemistry
Publisher
Informa UK Limited
Published
2021-02-16
Volume
13
Issue
6
Pages
575-585
ISSN
  • 1756-8919 (ISSN-L)
  • 1756-8927
DOI
10.4155/fmc-2020-0262
PMC
PMC7888348
OpenAlex
W3090171308
Cited by
118
References
58
Citations by year
2026: 9 · 2025: 11 · 2024: 17 · 2023: 13 · 2022: 24 · 2021: 29
FWCI
10.81
Type
journal article
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • Future Medicinal Chemistry (publishedVersion)
  • ChemRxiv (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Royal College of Surgeons in Ireland (submittedVersion, cc-by-nc-nd)
Topics
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
  • Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
  • Health Sciences › Medicine › Infectious Diseases › SARS-CoV-2 and COVID-19 Research
Keywords
SARS-CoV-2, main protease, Mpro inhibitors, de novo drug design, generative models, deep neural network, transfer learning, reinforcement learning, virtual screening, antiviral agents
MeSH terms
COVID-19, SARS-CoV-2, Coronavirus 3C Proteases, COVID-19 Drug Treatment, Antiviral Agents, Artificial Intelligence, Humans, Ligands, Protease Inhibitors, Drug Design, Drug Discovery, Molecular Docking Simulation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review