scholarly journals Neuraldecipher - Reverse-Engineering ECFP Fingerprints to Their Molecular Structures

2020 ◽  
Author(s):  
Tuan Le ◽  
Robin Winter ◽  
Frank Noé ◽  
Djork-Arné Clevert
Keyword(s):  
Molecular Structure ◽  
Reverse Engineering ◽  
Quantitative Structure Activity Relationship ◽  
Molecular Structures ◽  
Pharmaceutical Companies ◽  
Molecular Fingerprints ◽  
Structure Activity ◽  
Great Relevance ◽  
Validation Set ◽  
Private Associations

<p>Protecting molecular structures from disclosure against external parties is of great relevance for industrial and private associations, such as pharmaceutical companies. Within the framework of external collaborations, it is common to exchange datasets by encoding the molecular structures into descriptors. Molecular fingerprints such as the extended-connectivity fingerprints are frequently used for such an exchange, because they typically perform well on quantitative structure-activity relationship tasks. </p><p>ECFPs are often considered to be non-invertible due to the way they are computed.</p><p>In this paper, we present a reverse-engineering method to deduce the molecular structure given revealed ECFPs. Our method includes the <i>Neuraldecipher</i>, a neural network model that predicts a compact vector representation of compounds, given ECFPs. We then utilize another pre-trained model to retrieve the molecular structure as SMILES representation. We demonstrate that our method is able to reconstruct molecular structures to some extent, and improves, when ECFPs with larger fingerprint sizes are revealed. For example, given ECFP count vectors of length 4096, we are able to correctly deduce around 60% of molecular structures on a validation set (112K unique samples) with our method.</p>

scholarly journals Neuraldecipher - Reverse-Engineering ECFP Fingerprints to Their Molecular Structures

2020 ◽  
Author(s):  
Tuan Le ◽  
Robin Winter ◽  
Frank Noé ◽  
Djork-Arné Clevert
Keyword(s):  
Molecular Structure ◽  
Reverse Engineering ◽  
Quantitative Structure Activity Relationship ◽  
Molecular Structures ◽  
Pharmaceutical Companies ◽  
Molecular Fingerprints ◽  
Structure Activity ◽  
Great Relevance ◽  
Validation Set ◽  
Private Associations

<p>Protecting molecular structures from disclosure against external parties is of great relevance for industrial and private associations, such as pharmaceutical companies. Within the framework of external collaborations, it is common to exchange datasets by encoding the molecular structures into descriptors. Molecular fingerprints such as the extended-connectivity fingerprints are frequently used for such an exchange, because they typically perform well on quantitative structure-activity relationship tasks. </p><p>ECFPs are often considered to be non-invertible due to the way they are computed.</p><p>In this paper, we present a reverse-engineering method to deduce the molecular structure given revealed ECFPs. Our method includes the <i>Neuraldecipher</i>, a neural network model that predicts a compact vector representation of compounds, given ECFPs. We then utilize another pre-trained model to retrieve the molecular structure as SMILES representation. We demonstrate that our method is able to reconstruct molecular structures to some extent, and improves, when ECFPs with larger fingerprint sizes are revealed. For example, given ECFP count vectors of length 4096, we are able to correctly deduce around 60% of molecular structures on a validation set (112K unique samples) with our method.</p>

scholarly journals Neuraldecipher - Reverse-Engineering ECFP Fingerprints to Their Molecular Structures

2020 ◽  
Author(s):  
Tuan Le ◽  
Robin Winter ◽  
Frank Noé ◽  
Djork-Arné Clevert
Keyword(s):  
Molecular Structure ◽  
Reverse Engineering ◽  
Quantitative Structure Activity Relationship ◽  
Molecular Structures ◽  
Pharmaceutical Companies ◽  
Molecular Fingerprints ◽  
Structure Activity ◽  
Great Relevance ◽  
Validation Set ◽  
Private Associations

<p>Protecting molecular structures from disclosure against external parties is of great relevance for industrial and private associations, such as pharmaceutical companies. Within the framework of external collaborations, it is common to exchange datasets by encoding the molecular structures into descriptors. Molecular fingerprints such as the extended-connectivity fingerprints are frequently used for such an exchange, because they typically perform well on quantitative structure-activity relationship tasks. </p><p>ECFPs are often considered to be non-invertible due to the way they are computed.</p><p>In this paper, we present a reverse-engineering method to deduce the molecular structure given revealed ECFPs. Our method includes the <i>Neuraldecipher</i>, a neural network model that predicts a compact vector representation of compounds, given ECFPs. We then utilize another pre-trained model to retrieve the molecular structure as SMILES representation. We demonstrate that our method is able to reconstruct molecular structures to some extent, and improves, when ECFPs with larger fingerprint sizes are revealed. For example, given ECFP count vectors of length 4096, we are able to correctly deduce around 60% of molecular structures on a validation set (112K unique samples) with our method.</p>

scholarly journals Neuraldecipher – reverse-engineering extended-connectivity fingerprints (ECFPs) to their molecular structures

2020 ◽  
Vol 11 (38) ◽  
pp. 10378-10389
Author(s):  
Tuan Le ◽  
Robin Winter ◽  
Frank Noé ◽  
Djork-Arné Clevert
Keyword(s):  
Reverse Engineering ◽  
Molecular Structures ◽  
Pharmaceutical Companies ◽  
Great Relevance ◽  
Private Associations ◽  
Extended Connectivity

Protecting molecular structures from disclosure against external parties is of great relevance for industrial and private associations, such as pharmaceutical companies.

scholarly journals Modeling the Antileukemia Activity of Ellipticine-Related Compounds: QSAR and Molecular Docking Study

Molecules ◽  
2019 ◽  
Vol 25 (1) ◽  
pp. 24 ◽  
Author(s):  
Edgar Márquez ◽  
José R. Mora ◽  
Virginia Flores-Morales ◽  
Daniel Insuasty ◽  
Luis Calle
Keyword(s):  
Molecular Structure ◽  
Binding Free Energy ◽  
Docking Study ◽  
Quantitative Structure Activity Relationship ◽  
Quantitative Structure ◽  
Molecular Docking Study ◽  
Modeling Simulation ◽  
Structure Activity ◽  
Related Compounds ◽  
Cancer Activity

The antileukemia cancer activity of organic compounds analogous to ellipticine representes a critical endpoint in the understanding of this dramatic disease. A molecular modeling simulation on a dataset of 23 compounds, all of which comply with Lipinski’s rules and have a structure analogous to ellipticine, was performed using the quantitative structure activity relationship (QSAR) technique, followed by a detailed docking study on three different proteins significantly involved in this disease (PDB IDs: SYK, PI3K and BTK). As a result, a model with only four descriptors (HOMO, softness, AC1RABAMBID, and TS1KFABMID) was found to be robust enough for prediction of the antileukemia activity of the compounds studied in this work, with an R2 of 0.899 and Q2 of 0.730. A favorable interaction between the compounds and their target proteins was found in all cases; in particular, compounds 9 and 22 showed high activity and binding free energy values of around −10 kcal/mol. Theses compounds were evaluated in detail based on their molecular structure, and some modifications are suggested herein to enhance their biological activity. In particular, compounds 22_1, 22_2, 9_1, and 9_2 are indicated as possible new, potent ellipticine derivatives to be synthesized and biologically tested.

Biological Activity, Physical Properties, and Toxicity

2021 ◽  
Vol 6 (3) ◽  
pp. 25-34
Author(s):  
Ranita Pal ◽  
Pratim Kumar Chattaraj
Keyword(s):  
Molecular Structure ◽  
Biological Activity ◽  
Molecular Biology ◽  
Drug Design ◽  
Physical Properties ◽  
Quantitative Structure Activity Relationship ◽  
Molecular Properties ◽  
Quantitative Structure ◽  
Qsar Analysis ◽  
Structure Activity

In the current pandemic-stricken world, quantitative structure-activity relationship (QSAR) analysis has become a necessity in the domain of molecular biology and drug design, realizing that it helps estimate properties and activities of a compound, without actually having to spend time and resources to synthesize it in the laboratory. Correlating the molecular structure of a compound with its activity depends on the choice of the descriptors, which becomes a difficult and confusing task when we have so many to choose from. In this mini-review, the authors delineate the importance of very simple and easy to compute descriptors in estimating various molecular properties/toxicity.

scholarly journals Useful Irregularity Indices in QSPR Study for Bismuth Tri-Iodide

2019 ◽  
Vol 2019 ◽  
pp. 1-17
Author(s):  
Abaid ur Rehman Virk ◽  
M. A. Rehman ◽  
Ce Shi ◽  
Waqas Nazeer
Keyword(s):  
Chemical Compound ◽  
Chemical Compounds ◽  
Quantitative Structure Activity Relationship ◽  
Molecular Structures ◽  
Mathematical Language ◽  
Quantitative Structure ◽  
Boiling Points ◽  
Structure Activity ◽  
Qspr Study ◽  
Single Number

Topological indices give us a mathematical language to study molecular structures. They convert a chemical compound into a single number which foresees properties, for example, boiling points, viscosity, and the radius of gyrations. Drugs and other chemical compounds are often modeled as various polygonal shapes, trees, and graphs. In this paper, we will compute some irregularity indices for bismuth tri-iodide chain and sheet that are useful in the quantitative structure-activity relationship.

Multilayer Perceptron Model for Predicting Acute Toxicity of Fungicides on Rats

2018 ◽  
Vol 3 (1) ◽  
pp. 100-118 ◽  
Author(s):  
Mabrouk Hamadache ◽  
Abdeltif Amrane ◽  
Salah Hanini ◽  
Othmane Benkortbi
Keyword(s):  
Molecular Descriptors ◽  
External Validation ◽  
Qsar Model ◽  
Quantitative Structure Activity Relationship ◽  
Acute Oral Toxicity ◽  
Oral Toxicity ◽  
Structure Activity ◽  
Marquardt Algorithm ◽  
Qsar Models ◽  
Validation Set

Quantitative Structure Activity Relationship (QSAR) models are expected to play an important role in the risk assessment of chemicals on humans and the environment. In this study, a QSAR model based on 10 molecular descriptors to predict acute oral toxicity of 91 fungicides to rats was developed and validated. Good results (PRESS/SSY = 0.085 and VIF < 5) were obtained, showing the validation of descriptors in the obtained model. The best results were obtained with a 10/11/1 Artificial Neural Network model trained with the Levenberg-Marquardt algorithm. The prediction accuracy for the external validation set was estimated by the Q2ext which was equal to 0.960. Accordingly, the model developed in this study provided excellent predictions and can be used to predict the acute oral toxicity of fungicides, particularly for those that have not been tested as well as new fungicides.

Quantitative structure–activity relationship correlation between molecular structure and the Rayleigh enantiomeric enrichment factor

2015 ◽  
Vol 17 (8) ◽  
pp. 1370-1376 ◽  
Author(s):  
S. Jammer ◽  
D. Rizkov ◽  
F. Gelman ◽  
O. Lev
Keyword(s):  
Molecular Structure ◽  
Enrichment Factor ◽  
Homologous Series ◽  
Quantitative Structure Activity Relationship ◽  
Activity Relationship ◽  
Enantioselective Hydrolysis ◽  
Quantitative Structure ◽  
Enantiomeric Enrichment ◽  
Molecular Features ◽  
Structure Activity

The enantiomeric enrichment caused by enzymatic enantioselective hydrolysis is studied for a homologous series, revealing a correlation between substrate molecular features and the Rayleigh enantiomeric enrichment factor,εER.

Sign in / Sign up

Export Citation Format

Share Document