scholarly journals Modeling Physico-Chemical ADMET Endpoints with Multitask Graph Convolutional Networks

Molecules ◽  
2019 ◽  
Vol 25 (1) ◽  
pp. 44 ◽  
Author(s):  
Floriane Montanari ◽  
Lara Kuhnke ◽  
Antonius Ter Laak ◽  
Djork-Arné Clevert
Keyword(s):  
Chemical Properties ◽  
Predictive Performance ◽  
Machine Learning Techniques ◽  
Convolutional Networks ◽  
Modeling Methods ◽  
Learning Techniques ◽  
Physico Chemical ◽  
Solubility Equation ◽  
Fully Connected ◽  
New Compounds

Simple physico-chemical properties, like logD, solubility, or melting point, can reveal a great deal about how a compound under development might later behave. These data are typically measured for most compounds in drug discovery projects in a medium throughput fashion. Collecting and assembling all the Bayer in-house data related to these properties allowed us to apply powerful machine learning techniques to predict the outcome of those assays for new compounds. In this paper, we report our finding that, especially for predicting physicochemical ADMET endpoints, a multitask graph convolutional approach appears a highly competitive choice. For seven endpoints of interest, we compared the performance of that approach to fully connected neural networks and different single task models. The new model shows increased predictive performance compared to previous modeling methods and will allow early prioritization of compounds even before they are synthesized. In addition, our model follows the generalized solubility equation without being explicitly trained under this constraint.

Modeling Physico-Chemical ADMET Endpoints With Multitask Graph Convolutional Networks

2019 ◽  
Author(s):  
Floriane Montanari ◽  
Lara Kuhnke ◽  
Antonius ter Laak ◽  
Djork-Arné Clevert
Keyword(s):  
Chemical Properties ◽  
Predictive Performance ◽  
Machine Learning Techniques ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Modeling Methods ◽  
Applied Machine Learning ◽  
Learning Techniques ◽  
Physico Chemical ◽  
New Compounds

Simple physico-chemical properties like logD, solubility or serum albumin binding have a direct impact on the likelihood of success of compounds in clinical trials. Here, we collected all the Bayer in house data related to these properties and applied machine learning techniques to predict them for new compounds. We report that, for the endpoints studied here, a multitask graph convolutional network appears a highly competitive choice. The new model shows increased predictive performance on all endpoints compared to previous modeling methods.<br>

Modeling Physico-Chemical ADMET Endpoints With Multitask Graph Convolutional Networks

2019 ◽  
Author(s):  
Floriane Montanari ◽  
Lara Kuhnke ◽  
Antonius ter Laak ◽  
Djork-Arné Clevert
Keyword(s):  
Chemical Properties ◽  
Predictive Performance ◽  
Machine Learning Techniques ◽  
Convolutional Network ◽  
Convolutional Networks ◽  
Modeling Methods ◽  
Applied Machine Learning ◽  
Learning Techniques ◽  
Physico Chemical ◽  
New Compounds

Simple physico-chemical properties like logD, solubility or serum albumin binding have a direct impact on the likelihood of success of compounds in clinical trials. Here, we collected all the Bayer in house data related to these properties and applied machine learning techniques to predict them for new compounds. We report that, for the endpoints studied here, a multitask graph convolutional network appears a highly competitive choice. The new model shows increased predictive performance on all endpoints compared to previous modeling methods.<br>

scholarly journals Prediction of antimicrobial peptides toxicity based on their physico-chemical properties using machine learning techniques

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Hossein Khabbaz ◽  
Mohammad Hossein Karimi-Jafari ◽  
Ali Akbar Saboury ◽  
Bagher BabaAli
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Antimicrobial Peptides ◽  
Mammalian Cells ◽  
Chemical Properties ◽  
Machine Learning Techniques ◽  
Learning Techniques ◽  
Physico Chemical ◽  
Machine Learning Model ◽  
Low Toxicity

Abstract Background Antimicrobial peptides are promising tools to fight against ever-growing antibiotic resistance. However, despite many advantages, their toxicity to mammalian cells is a critical obstacle in clinical application and needs to be addressed. Results In this study, by using an up-to-date dataset, a machine learning model has been trained successfully to predict the toxicity of antimicrobial peptides. The comprehensive set of features of both physico-chemical and linguistic-based with local and global essences have undergone feature selection to identify key properties behind toxicity of antimicrobial peptides. After feature selection, the hybrid model showed the best performance with a recall of 0. 876 and a F1 score of 0. 849. Conclusions The obtained model can be useful in extracting AMPs with low toxicity from AMP libraries in clinical applications. On the other hand, several properties with local nature including positions of strand forming and hydrophobic residues in final selected features show that these properties are critical definer of peptide properties and should be considered in developing models for activity prediction of peptides. The executable code is available at https://git.io/JRZaT.

Application of Machine Learning Techniques to Predict Binding Affinity for Drug Targets: A Study of Cyclin-Dependent Kinase 2

2020 ◽  
Vol 28 (2) ◽  
pp. 253-265 ◽  
Author(s):  
Gabriela Bitencourt-Ferreira ◽  
Amauri Duarte da Silva ◽  
Walter Filgueira de Azevedo
Keyword(s):  
Machine Learning ◽  
Binding Affinity ◽  
Predictive Performance ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
Scoring Functions ◽  
Cyclin Dependent Kinase ◽  
Learning Models ◽  
Learning Techniques ◽  
Machine Learning Models

Background: The elucidation of the structure of cyclin-dependent kinase 2 (CDK2) made it possible to develop targeted scoring functions for virtual screening aimed to identify new inhibitors for this enzyme. CDK2 is a protein target for the development of drugs intended to modulate cellcycle progression and control. Such drugs have potential anticancer activities. Objective: Our goal here is to review recent applications of machine learning methods to predict ligand- binding affinity for protein targets. To assess the predictive performance of classical scoring functions and targeted scoring functions, we focused our analysis on CDK2 structures. Methods: We have experimental structural data for hundreds of binary complexes of CDK2 with different ligands, many of them with inhibition constant information. We investigate here computational methods to calculate the binding affinity of CDK2 through classical scoring functions and machine- learning models. Results: Analysis of the predictive performance of classical scoring functions available in docking programs such as Molegro Virtual Docker, AutoDock4, and Autodock Vina indicated that these methods failed to predict binding affinity with significant correlation with experimental data. Targeted scoring functions developed through supervised machine learning techniques showed a significant correlation with experimental data. Conclusion: Here, we described the application of supervised machine learning techniques to generate a scoring function to predict binding affinity. Machine learning models showed superior predictive performance when compared with classical scoring functions. Analysis of the computational models obtained through machine learning could capture essential structural features responsible for binding affinity against CDK2.

scholarly journals Exploring the octanol–water partition coefficient dataset using deep learning techniques and data augmentation

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Nadin Ulrich ◽  
Kai-Uwe Goss ◽  
Andrea Ebert
Keyword(s):  
Partition Coefficient ◽  
Data Augmentation ◽  
Environmental Chemistry ◽  
Chemical Properties ◽  
Predictive Performance ◽  
Chemical Structures ◽  
Tautomeric Forms ◽  
Learning Techniques ◽  
Water Partition Coefficient ◽  
Using Data

AbstractToday more and more data are freely available. Based on these big datasets deep neural networks (DNNs) rapidly gain relevance in computational chemistry. Here, we explore the potential of DNNs to predict chemical properties from chemical structures. We have selected the octanol-water partition coefficient (log P) as an example, which plays an essential role in environmental chemistry and toxicology but also in chemical analysis. The predictive performance of the developed DNN is good with an rmse of 0.47 log units in the test dataset and an rmse of 0.33 for an external dataset from the SAMPL6 challenge. To this end, we trained the DNN using data augmentation considering all potential tautomeric forms of the chemicals. We further demonstrate how DNN models can help in the curation of the log P dataset by identifying potential errors, and address limitations of the dataset itself.

scholarly journals Generalized Read-Across prediction using genra-py

2021 ◽  
Author(s):  
Imran Shah ◽  
Tia Tate ◽  
Grace Patlewicz
Keyword(s):  
Risk Assessment ◽  
Human Health Risk ◽  
Chemical Properties ◽  
Predictive Performance ◽  
Hazard Identification ◽  
Data Driven ◽  
Chemical Safety ◽  
Physico Chemical ◽  
Data Driven Approach ◽  
Assessment Problems

Abstract Motivation Generalized Read-Across (GenRA) is a data-driven approach to estimate physico-chemical, biological or eco-toxicological properties of chemicals by inference from analogues. GenRA attempts to mimic a human expert’s manual read-across reasoning for filling data gaps about new chemicals from known chemicals with an interpretable and automated approach based on nearest-neighbors. A key objective of GenRA is to systematically explore different choices of input data selection and neighborhood definition to objectively evaluate predictive performance of automated read-across estimates of chemical properties. Results We have implemented genra-py as a python package that can be freely used for chemical safety analysis and risk assessment applications. Automated read-across prediction in genra-py conforms to the scikit-learn machine learning library's estimator design pattern, making it easy to use and integrate in computational pipelines. We demonstrate the data-driven application of genra-py to address two key human health risk assessment problems namely: hazard identification and point of departure estimation. Availability and implementation The package is available from github.com/i-shah/genra-py.

scholarly journals Automated methods of coherence evaluation of Ukrainian texts using machine learning techniques

2020 ◽  
pp. 295-303
Author(s):  
A.A. Kramov ◽  
◽  
S.D. Pogorilyy ◽  
Keyword(s):  
Machine Learning ◽  
Semantic Similarity ◽  
Machine Learning Techniques ◽  
Vector Representation ◽  
Early Stopping ◽  
Experimental Examination ◽  
Convolutional Networks ◽  
Learning Techniques ◽  
Similarity Graph ◽  
Automated Methods

The main methods of coherence evaluation of texts with the usage of different machine learning techniques have been analyzed. The principles of methods with the usage of recurrent and convolutional neural networks have been described in details. The advantages of a semantic similarity graph method have been considered. Other approaches to perform the vector representation of sentences for the estimation of semantic similarity between the elements of a text have been suggested to use. The experimental examination of methods has been performed on the set of Ukrainian scientific articles. The training of recurrent and convolutional networks with the usage of early stopping has been performed. The accuracy of the solving of document discrimination and insertion tasks has been calculated. The comparative analysis of the results obtained has been performed.

scholarly journals Modern machine learning outperforms GLMs at predicting spikes

2017 ◽  
Author(s):  
Ari S. Benjamin ◽  
Hugo L. Fernandes ◽  
Tucker Tomlinson ◽  
Pavan Ramkumar ◽  
Chris VerSteeg ◽  
...  
Keyword(s):  
Machine Learning ◽  
Neural Activity ◽  
Linear Models ◽  
Feedforward Neural Networks ◽  
Predictive Performance ◽  
Machine Learning Techniques ◽  
Machine Learning Methods ◽  
Learning Techniques ◽  
Neural Spiking ◽  
Modern Machine

AbstractNeuroscience has long focused on finding encoding models that effectively ask “what predicts neural spiking?” and generalized linear models (GLMs) are a typical approach. It is often unknown how much of explainable neural activity is captured, or missed, when fitting a GLM. Here we compared the predictive performance of GLMs to three leading machine learning methods: feedforward neural networks, gradient boosted trees (using XGBoost), and stacked ensembles that combine the predictions of several methods. We predicted spike counts in macaque motor (M1) and somatosensory (S1) cortices from standard representations of reaching kinematics, and in rat hippocampal cells from open field location and orientation. In general, the modern methods (particularly XGBoost and the ensemble) produced more accurate spike predictions and were less sensitive to the preprocessing of features. This discrepancy in performance suggests that standard feature sets may often relate to neural activity in a nonlinear manner not captured by GLMs. Encoding models built with machine learning techniques, which can be largely automated, more accurately predict spikes and can offer meaningful benchmarks for simpler models.

Sign in / Sign up

Export Citation Format

Share Document