Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Molecular Models

Molecular Models

Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.

Fischer Projections

Fischer Projections

Learning to draw Fischer projections of molecules and understanding their relevance plays a crucial role in the visual depiction of organic molecules. A Fischer projection is a two-dimensional projection on a planar surface to simplify the three-dimensional wedge–dash representation of molecules. This is especially helpful in the case of molecules with multiple chiral centers that can be difficult to draw. Here, all the bonds of interest are represented as horizontal or vertical lines. While...

Newman Projections

Newman Projections

Different notations are used to represent the three-dimensional structure of molecules on two-dimensional surfaces. One of the most commonly used representations is the dash-wedge formula. The dashed wedges, solid wedges, and the plane lines indicate the groups situated behind the plane, coming out of the plane, and in the plane, respectively.
The organic molecules rotate across the single bonds leading to numerous temporary three-dimensional structures of varying energy known as...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

FLOWR.ROOT - A flow matching-based foundation model for joint multi-purpose structure-aware 3D ligand generation and affinity prediction.

Nature communications·2026

Same author

Scalable Boltzmann generators for equilibrium sampling of large-scale materials.

Nature communications·2026

Same author

FLOWR: flow matching for structure-aware de novo, interaction- and fragment-based ligand generation.

Nature computational science·2026

Same author

Assessing generative modeling approaches for free energy estimates in condensed matter.

The Journal of chemical physics·2026

Same author

Learning data-efficient coarse-grained molecular dynamics from forces and noise.

Nature communications·2026

Same author

Extending the range of graph neural networks with global encodings.

Nature communications·2026

Search research articles

Related Experiment Video

Updated: Nov 2, 2025

Author Spotlight: Exploring Cellular Processes by Modeling Ligands in Cryo-EM Maps

Author Spotlight: Exploring Cellular Processes by Modeling Ligands in Cryo-EM Maps

Published on: July 19, 2024

Neuraldecipher - reverse-engineering extended-connectivity fingerprints (ECFPs) to their molecular structures.

Tuan Le^1,2, Robin Winter^1,2, Frank Noé²

¹Department of Digital Technologies, Bayer AG Berlin Germany tuan.le2@bayer.com djork-arne.clevert@bayer.com.

Chemical Science

|June 7, 2021

Summary

Researchers developed a fast reverse-engineering method to reconstruct molecular structures from extended-connectivity fingerprints (ECFPs). This method successfully deduces up to 69% of molecular structures using larger fingerprint sizes.

More Related Videos

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Published on: December 16, 2013

High Precision FRET at Single-molecule Level for Biomolecule Structure Determination

High Precision FRET at Single-molecule Level for Biomolecule Structure Determination

Published on: May 13, 2017

Related Experiment Videos

Last Updated: Nov 2, 2025

Author Spotlight: Exploring Cellular Processes by Modeling Ligands in Cryo-EM Maps

Author Spotlight: Exploring Cellular Processes by Modeling Ligands in Cryo-EM Maps

Published on: July 19, 2024

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Structure and Coordination Determination of Peptide-metal Complexes Using 1D and 2D 1H NMR

Published on: December 16, 2013

High Precision FRET at Single-molecule Level for Biomolecule Structure Determination

High Precision FRET at Single-molecule Level for Biomolecule Structure Determination

Published on: May 13, 2017

Area of Science:

Computational chemistry
Cheminformatics
Machine learning in drug discovery

Background:

Protecting proprietary molecular structures is crucial for pharmaceutical companies during collaborations.
Molecular fingerprints, like extended-connectivity fingerprints (ECFPs), are used to encode structures for data exchange.
ECFPs are often considered non-invertible due to their computation method.

Purpose of the Study:

To present a novel, fast reverse-engineering method for deducing molecular structures from ECFPs.
To assess the effectiveness of the proposed method in reconstructing molecular information.
To explore the impact of ECFP size on reconstruction accuracy.

Main Methods:

Developed a neural network model, Neuraldecipher, to predict compact vector representations from ECFPs.
Utilized a pre-trained model to retrieve molecular structures in SMILES format from the predicted vectors.
Evaluated the method's performance on a large validation set (112K unique samples).

Main Results:

The proposed method can reconstruct molecular structures to a certain extent.
Reconstruction accuracy improves with larger fingerprint sizes for ECFPs.
Achieved up to 69% accuracy in deducing molecular structures using ECFP count vectors of length 4096.

Conclusions:

The developed reverse-engineering technique offers a viable approach to potentially recover molecular structures from ECFPs.
Larger ECFP sizes enhance the success rate of molecular structure reconstruction.
This method has implications for data security and information recovery in cheminformatics.