Deep interactome learning for generative drug design

Experimental Data Snapshot

  • Resolution: 1.85 Å
  • R-Value Free: 0.213 
  • R-Value Work: 0.171 

Starting Model: experimental
View more details

wwPDB Validation   3D Report Full Report

Ligand Structure Quality Assessment 

This is version 1.0 of the entry. See complete history


Prospective de novo drug design with deep interactome learning.

Atz, K.Cotos, L.Isert, C.Hakansson, M.Focht, D.Hilleke, M.Nippa, D.F.Iff, M.Ledergerber, J.Schiebroek, C.C.G.Romeo, V.Hiss, J.A.Merk, D.Schneider, P.Kuhn, B.Grether, U.Schneider, G.

(2024) Nat Commun 15: 3408-3408

  • DOI: https://doi.org/10.1038/s41467-024-47613-w
  • Primary Citation of Related Structures:  

  • PubMed Abstract: 

    De novo drug design aims to generate molecules from scratch that possess specific chemical and pharmacological properties. We present a computational approach utilizing interactome-based deep learning for ligand- and structure-based generation of drug-like molecules. This method capitalizes on the unique strengths of both graph neural networks and chemical language models, offering an alternative to the need for application-specific reinforcement, transfer, or few-shot learning. It enables the "zero-shot" construction of compound libraries tailored to possess specific bioactivity, synthesizability, and structural novelty. In order to proactively evaluate the deep interactome learning framework for protein structure-based drug design, potential new ligands targeting the binding site of the human peroxisome proliferator-activated receptor (PPAR) subtype gamma are generated. The top-ranking designs are chemically synthesized and computationally, biophysically, and biochemically characterized. Potent PPAR partial agonists are identified, demonstrating favorable activity and the desired selectivity profiles for both nuclear receptors and off-target interactions. Crystal structure determination of the ligand-receptor complex confirms the anticipated binding mode. This successful outcome positively advocates interactome-based de novo design for application in bioorganic and medicinal chemistry, enabling the creation of innovative bioactive molecules.

  • Organizational Affiliation

    ETH Zurich, Department of Chemistry and Applied Biosciences, Vladimir-Prelog-Weg 4, 8093, Zurich, Switzerland.

Find similar proteins by:  (by identity cutoff)  |  3D Structure
Entity ID: 1
MoleculeChains Sequence LengthOrganismDetailsImage
Peroxisome proliferator-activated receptor gamma
A, B
275Homo sapiensMutation(s): 0 
Gene Names: PPARGNR1C3
UniProt & NIH Common Fund Data Resources
Find proteins for P37231 (Homo sapiens)
Explore P37231 
Go to UniProtKB:  P37231
GTEx:  ENSG00000132170 
Entity Groups  
Sequence Clusters30% Identity50% Identity70% Identity90% Identity95% Identity100% Identity
UniProt GroupP37231
Sequence Annotations
  • Reference Sequence
Small Molecules
Ligands 3 Unique
IDChains Name / Formula / InChI Key2D Diagram3D Interactions
Y5I (Subject of Investigation/LOI)
Query on Y5I

Download Ideal Coordinates CCD File 
C [auth A]3-[2-fluoranyl-4-[3-[2-fluoranyl-4-(5-methyl-1,3,4-thiadiazol-2-yl)phenoxy]propoxy]phenyl]propanoic acid
C21 H20 F2 N2 O4 S
Query on SO4

Download Ideal Coordinates CCD File 
G [auth A],
H [auth A],
I [auth A],
J [auth A],
L [auth B]
O4 S
Query on GOL

Download Ideal Coordinates CCD File 
D [auth A],
E [auth A],
F [auth A],
K [auth B]
C3 H8 O3
Experimental Data & Validation

Experimental Data

  • Resolution: 1.85 Å
  • R-Value Free: 0.213 
  • R-Value Work: 0.171 
  • Space Group: C 1 2 1
Unit Cell:
Length ( Å )Angle ( ˚ )
a = 93.413α = 90
b = 60.611β = 102.859
c = 117.945γ = 90
Software Package:
Software NamePurpose
XDSdata reduction
XDSdata scaling

Structure Validation

View Full Validation Report

Ligand Structure Quality Assessment 

Entry History & Funding Information

Deposition Data

Funding OrganizationLocationGrant Number
Swiss National Science FoundationSwitzerland205321_182176

Revision History  (Full details and data files)

  • Version 1.0: 2024-05-15
    Type: Initial release