Data-driven discovery of cardiolipin-selective small molecules by computational active learning

Publication type:
Journal article
Metadata:
Abstract
<jats:p>We present a data-driven approach combining deep learning-enabled active learning with coarse-grained simulations and alchemical free energy calculations to discover small molecules to selectively permeate cardiolipin membranes.</jats:p>
Autoren
  • Bernadette Mohr
  • Kirill Shmilovich
  • Isabel S Kleinwächter
  • Dirk Schneider
  • Andrew L Ferguson
  • Tristan Bereau
DOI
10.1039/d2sc00116k
eISSN
2041-6539
ISSN
2041-6520
Ausgabe der Veröffentlichung
16
Zeitschrift
Chemical Science
Sprache
en
Online publication date
2022
Paginierung
4498 - 4511
Status
Published online
Herausgeber
Royal Society of Chemistry (RSC)
Herausgeber URL
http://dx.doi.org/10.1039/d2sc00116k
Datum der Datenerfassung
2024
Titel
Data-driven discovery of cardiolipin-selective small molecules by computational active learning
Ausgabe der Zeitschrift
13

Data source: Crossref

Other metadata sources:
Abstract
Subtle variations in the lipid composition of mitochondrial membranes can have a profound impact on mitochondrial function. The inner mitochondrial membrane contains the phospholipid cardiolipin, which has been demonstrated to act as a biomarker for a number of diverse pathologies. Small molecule dyes capable of selectively partitioning into cardiolipin membranes enable visualization and quantification of the cardiolipin content. Here we present a data-driven approach that combines a deep learning-enabled active learning workflow with coarse-grained molecular dynamics simulations and alchemical free energy calculations to discover small organic compounds able to selectively permeate cardiolipin-containing membranes. By employing transferable coarse-grained models we efficiently navigate the all-atom design space corresponding to small organic molecules with molecular weight less than ≈500 Da. After direct simulation of only 0.42% of our coarse-grained search space we identify molecules with considerably increased levels of cardiolipin selectivity compared to a widely used cardiolipin probe 10-<i>N</i>-nonyl acridine orange. Our accumulated simulation data enables us to derive interpretable design rules linking coarse-grained structure to cardiolipin selectivity. The findings are corroborated by fluorescence anisotropy measurements of two compounds conforming to our defined design rules. Our findings highlight the potential of coarse-grained representations and multiscale modelling for materials discovery and design.
Addresses
  • Van't Hoff Institute for Molecular Sciences and Informatics Institute, University of Amsterdam Amsterdam 1098 XH The Netherlands t.bereau@uva.nl.
Autoren
  • Bernadette Mohr
  • Kirill Shmilovich
  • Isabel S Kleinwächter
  • Dirk Schneider
  • Andrew L Ferguson
  • Tristan Bereau
DOI
10.1039/d2sc00116k
eISSN
2041-6539
Externe Identifier
  • PubMed Identifier: 35656132
  • PubMed Central ID: PMC9019913
Funding acknowledgements
  • National Science Foundation: DMR-1828629
  • National Science Foundation: DGE-1746045
  • National Science Foundation: DMS-1440415
Open access
true
ISSN
2041-6520
Ausgabe der Veröffentlichung
16
Zeitschrift
Chemical science
Sprache
eng
Medium
Electronic-eCollection
Online publication date
2022
Open access status
Open Access
Paginierung
4498 - 4511
Datum der Veröffentlichung
2022
Status
Published
Publisher licence
CC BY-NC
Datum der Datenerfassung
2022
Titel
Data-driven discovery of cardiolipin-selective small molecules by computational active learning.
Sub types
  • research-article
  • Journal Article
Ausgabe der Zeitschrift
13

Files

https://pubs.rsc.org/en/content/articlepdf/2022/sc/d2sc00116k https://europepmc.org/articles/PMC9019913?pdf=render

Data source: Europe PubMed Central

Abstract
Subtle variations in the lipid composition of mitochondrial membranes can have a profound impact on mitochondrial function. The inner mitochondrial membrane contains the phospholipid cardiolipin, which has been demonstrated to act as a biomarker for a number of diverse pathologies. Small molecule dyes capable of selectively partitioning into cardiolipin membranes enable visualization and quantification of the cardiolipin content. Here we present a data-driven approach that combines a deep learning-enabled active learning workflow with coarse-grained molecular dynamics simulations and alchemical free energy calculations to discover small organic compounds able to selectively permeate cardiolipin-containing membranes. By employing transferable coarse-grained models we efficiently navigate the all-atom design space corresponding to small organic molecules with molecular weight less than ≈500 Da. After direct simulation of only 0.42% of our coarse-grained search space we identify molecules with considerably increased levels of cardiolipin selectivity compared to a widely used cardiolipin probe 10-N-nonyl acridine orange. Our accumulated simulation data enables us to derive interpretable design rules linking coarse-grained structure to cardiolipin selectivity. The findings are corroborated by fluorescence anisotropy measurements of two compounds conforming to our defined design rules. Our findings highlight the potential of coarse-grained representations and multiscale modelling for materials discovery and design.
Date of acceptance
2022
Autoren
  • Bernadette Mohr
  • Kirill Shmilovich
  • Isabel S Kleinwächter
  • Dirk Schneider
  • Andrew L Ferguson
  • Tristan Bereau
Autoren-URL
https://www.ncbi.nlm.nih.gov/pubmed/35656132
DOI
10.1039/d2sc00116k
Externe Identifier
  • PubMed Central ID: PMC9019913
ISSN
2041-6520
Ausgabe der Veröffentlichung
16
Zeitschrift
Chem Sci
Sprache
eng
Country
England
Paginierung
4498 - 4511
PII
d2sc00116k
Datum der Veröffentlichung
2022
Status
Published online
Titel
Data-driven discovery of cardiolipin-selective small molecules by computational active learning.
Sub types
  • Journal Article
Ausgabe der Zeitschrift
13

Data source: PubMed

Beziehungen:
Property of

Tools