scholarly journals Coloring Molecules with Explainable Artificial Intelligence for Preclinical Relevance Assessment

2020 ◽  
Author(s):  
Jose Jimenez-Luna ◽  
Miha Skalic ◽  
Nils Weskamp ◽  
Gisbert Schneider
Keyword(s):  
Artificial Intelligence ◽  
De Novo ◽  
Molecular Design ◽  
Network Models ◽  
Neural Network Models ◽  
Molecular Features ◽  
Channel Inhibition ◽  
Explainable Artificial Intelligence ◽  
Rational Molecular Design ◽  
Graph Neural Networks

Graph neural networks are able to solve certain drug discovery tasks such as molecular property prediction and de novo molecule generation. However, these models are considered 'black-box' and 'hard-to-debug'. This study aimed to improve modeling transparency for rational molecular design by applying the integrated gradients explainable artificial intelligence (XAI) approach for graph neural network models. Models were trained for predicting plasma protein binding, cardiac potassium channel inhibition, passive permeability, and cytochrome P450 inhibition. The proposed methodology highlighted molecular features and structural elements that are in agreement with known pharmacophore motifs, correctly identified property cliffs, and provided insights into unspecific ligand-target interactions. The developed XAI approach is fully open-sourced and can be used by practitioners to train new models on other clinically-relevant endpoints.

scholarly journals Coloring Molecules with Explainable Artificial Intelligence for Preclinical Relevance Assessment

2020 ◽  
Author(s):  
Jose Jimenez-Luna ◽  
Miha Skalic ◽  
Nils Weskamp ◽  
Gisbert Schneider
Keyword(s):  
Artificial Intelligence ◽  
De Novo ◽  
Molecular Design ◽  
Network Models ◽  
Neural Network Models ◽  
Molecular Features ◽  
Channel Inhibition ◽  
Explainable Artificial Intelligence ◽  
Rational Molecular Design ◽  
Graph Neural Networks

Graph neural networks are able to solve certain drug discovery tasks such as molecular property prediction and de novo molecule generation. However, these models are considered 'black-box' and 'hard-to-debug'. This study aimed to improve modeling transparency for rational molecular design by applying the integrated gradients explainable artificial intelligence (XAI) approach for graph neural network models. Models were trained for predicting plasma protein binding, cardiac potassium channel inhibition, passive permeability, and cytochrome P450 inhibition. The proposed methodology highlighted molecular features and structural elements that are in agreement with known pharmacophore motifs, correctly identified property cliffs, and provided insights into unspecific ligand-target interactions. The developed XAI approach is fully open-sourced and can be used by practitioners to train new models on other clinically-relevant endpoints.

scholarly journals Towards Robust Representations of Spatial Networks Using Graph Neural Networks

2021 ◽  
Vol 11 (15) ◽  
pp. 6918
Author(s):  
Chidubem Iddianozie ◽  
Gavin McArdle
Keyword(s):  
Machine Learning ◽  
Model Performance ◽  
Network Models ◽  
Data Representation ◽  
Spatial Networks ◽  
Neural Network Models ◽  
Improve Model ◽  
Graph Neural Networks ◽  
Spatial Entities ◽  
Improve Model Performance

The effectiveness of a machine learning model is impacted by the data representation used. Consequently, it is crucial to investigate robust representations for efficient machine learning methods. In this paper, we explore the link between data representations and model performance for inference tasks on spatial networks. We argue that representations which explicitly encode the relations between spatial entities would improve model performance. Specifically, we consider homogeneous and heterogeneous representations of spatial networks. We recognise that the expressive nature of the heterogeneous representation may benefit spatial networks and could improve model performance on certain tasks. Thus, we carry out an empirical study using Graph Neural Network models for two inference tasks on spatial networks. Our results demonstrate that heterogeneous representations improves model performance for down-stream inference tasks on spatial networks.

scholarly journals Demand Modelling in Telecommunications

10.14311/1121 ◽  
2009 ◽  
Vol 49 (2) ◽  
Author(s):  
M. Chvalina
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Neural Networks ◽  
Network Models ◽  
Short Term ◽  
Neural Network Models ◽  
Term Forecast ◽  
Artificial Intelligence Methods ◽  
Demand Modelling

This article analyses the existing possibilities for using Standard Statistical Methods and Artificial Intelligence Methods for a short-term forecast and simulation of demand in the field of telecommunications. The most widespread methods are based on Time Series Analysis. Nowadays, approaches based on Artificial Intelligence Methods, including Neural Networks, are booming. Separate approaches will be used in the study of Demand Modelling in Telecommunications, and the results of these models will be compared with actual guaranteed values. Then we will examine the quality of Neural Network models. 

scholarly journals Combining Generative Artificial Intelligence and On-Chip Synthesis for De Novo Drug Design

2020 ◽  
Author(s):  
Francesca Grisoni ◽  
Berend Huisman ◽  
Alexander Button ◽  
Michael Moret ◽  
Kenneth Atz ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Drug Design ◽  
De Novo ◽  
Molecular Design ◽  
Chemical Space ◽  
Amide Bond ◽  
Reaction Products ◽  
De Novo Drug Design ◽  
On Chip

<p>Automation of the molecular design-make-test-analyze cycle speeds up the identification of hit and lead compounds for drug discovery. Using deep learning for computational molecular design and a customized microfluidics platform for on-chip compound synthesis, liver X receptor (LXR) agonists were generated from scratch. The computational pipeline was tuned to explore the chemical space defined by known LXRα agonists, and to suggest structural analogs of known ligands and novel molecular cores. To further the design of lead-like molecules and ensure compatibility with automated on-chip synthesis, this chemical space was confined to the set of virtual products obtainable from 17 different one-step reactions. Overall, 25 <i>de novo</i> generated compounds were successfully synthesized in flow via formation of sulfonamide, amide bond, and ester bond. First-pass <i>in vitro</i> activity screening of the crude reaction products in hybrid Gal4 reporter gene assays revealed 17 (68%) hits, with up to 60-fold LXR activation. The batch re-synthesis, purification, and re-testing of 14 of these compounds confirmed that 12 of them were potent LXRα or LXRβ agonists. These results support the utilization of the proposed design-make-test-analyze framework as a blueprint for automated drug design with artificial intelligence and miniaturized bench-top synthesis.<b></b></p>

scholarly journals Artificial Intelligence Guided De Novo Molecular Design Targeting COVID-19

2020 ◽  
Author(s):  
Srilok Srinivasan ◽  
Rohit Batra ◽  
Henry Chan ◽  
Ganesh Kamath ◽  
Mathew J. Cherukara ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
In Silico ◽  
High Throughput Screening ◽  
De Novo ◽  
Molecular Design ◽  
Chemical Space ◽  
Computational Cost ◽  
Ligand Design ◽  
Therapeutic Agents ◽  
Docking Simulations

An extensive search for active therapeutic agents against the SARS-CoV-2 is being conducted across the globe. Computational docking simulations have traditionally been used for <i>in silico</i> ligand design and remain popular method of choice for high-throughput screening of therapeutic agents in the fight against COVID-19. Despite the vast chemical space (millions to billions of biomolecules) that can be potentially explored as therapeutic agents, we remain severely limited in the search of candidate compounds owing to the high computational cost of these ensemble docking simulations employed in traditional <i>in silico</i> ligand design. Here, we present a <i>de novo</i> molecular design strategy that leverages artificial intelligence to discover new therapeutic biomolecules against SARS-CoV-2. A Monte Carlo Tree Search algorithm combined with a multi-task neural network (MTNN) surrogate model for expensive docking simulations and recurrent neural networks (RNN) for rollouts, is used to sample the exhaustive SMILES space of candidate biomolecules. Using Vina scores as target objective to measure binding of therapeutic molecules to either the isolated spike protein (S-protein) of SARS-CoV-2 at its host receptor region or to the S-protein:Angiotensin converting enzyme 2 (ACE2) receptor interface, we generate several (~100's) new biomolecules that outperform FDA (~1000’s) and non-FDA biomolecules (~million) from existing databases. A transfer learning strategy is deployed to retrain the MTNN surrogate as new candidate molecules are identified - this iterative search and retrain strategy is shown to accelerate the discovery of desired candidates. We perform detailed analysis using Lipinski's rules and also analyze the structural similarities between the various top performing candidates. We spilt the molecules using a molecular fragmenting algorithm and identify the common chemical fragments and patterns – such information is important to identify moieties that are responsible for improved performance. Although we focus on therapeutic biomolecules, our AI strategy is broadly applicable for accelerated design and discovery of any chemical molecules with user-desired functionality.

scholarly journals Energy and Policy Considerations for Modern Deep Learning Research

2020 ◽  
Vol 34 (09) ◽  
pp. 13693-13696
Author(s):  
Emma Strubell ◽  
Ananya Ganesh ◽  
Andrew McCallum
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Machine Learning ◽  
Network Models ◽  
Intelligence Community ◽  
Environmental Costs ◽  
Neural Network Models ◽  
Learning Research ◽  
Specialized Hardware ◽  
Compute Time

The field of artificial intelligence has experienced a dramatic methodological shift towards large neural networks trained on plentiful data. This shift has been fueled by recent advances in hardware and techniques enabling remarkable levels of computation, resulting in impressive advances in AI across many applications. However, the massive computation required to obtain these exciting results is costly both financially, due to the price of specialized hardware and electricity or cloud compute time, and to the environment, as a result of non-renewable energy used to fuel modern tensor processing hardware. In a paper published this year at ACL, we brought this issue to the attention of NLP researchers by quantifying the approximate financial and environmental costs of training and tuning neural network models for NLP (Strubell, Ganesh, and McCallum 2019). In this extended abstract, we briefly summarize our findings in NLP, incorporating updated estimates and broader information from recent related publications, and provide actionable recommendations to reduce costs and improve equity in the machine learning and artificial intelligence community.

scholarly journals System of providing sustainability of tower cranes from overturn in extreme wind loads

2018 ◽  
Vol 224 ◽  
pp. 02086
Author(s):  
Pavel Sorokin ◽  
Alexey Mishin ◽  
Vitaliy Antsev ◽  
Alexey Red’kin
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Network Models ◽  
Wind Loads ◽  
Neural Network Models ◽  
Quality Of Work ◽  
Development Stages ◽  
The Neural Network ◽  
Extreme Wind

The article is devoted to the issues of ensuring stability of tower cranes from overturn. The development stages of devices for ensuring tower cranes safety are examined and their shortcomings are revealed. The system consisting of subsystems and drives is proposed and their interaction is presented. The article deals with a subsystem based on artificial intelligence methods. The neural network models of forecasting wind parameters are developed. The quality of work of neural network models is estimated. The ways of further topic development are suggested.

De Novo Generation of Hit-like Molecules from Gene Expression Signatures Using Artificial Intelligence

2018 ◽  
Author(s):  
Oscar Mendez-Lucio ◽  
Benoit Baillif ◽  
Djork-Arné Clevert ◽  
David Rouquié ◽  
Joerg Wichard
Keyword(s):  
Gene Expression ◽  
Artificial Intelligence ◽  
De Novo ◽  
Molecular Design ◽  
Gene Expression Signature ◽  
Cellular Level ◽  
Generative Models ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Gene Expression Signatures

Finding new molecules with a desired biological activity is an extremely difficult task. In this context, artificial intelligence and generative models have been used for molecular <i>de novo</i> design and compound optimization. Herein, we report the first generative model that bridges systems biology and molecular design conditioning a generative adversarial network with transcriptomic data. By doing this we could generate molecules that have high probability to produce a desired biological effect at cellular level. We show that this model is able to design active-like molecules for desired targets without any previous target annotation of the training compounds as long as the gene expression signature of the desired state is provided. The molecules generated by this model are more similar to active compounds than the ones identified by similarity of gene expression signatures, which is the state-of-the-art method for navigating compound-induced gene expression data. Overall, this method represents a novel way to bridge chemistry and biology to advance in the long and difficult road of drug discovery.

De Novo Generation of Hit-like Molecules from Gene Expression Signatures Using Artificial Intelligence

2018 ◽  
Author(s):  
Oscar Mendez-Lucio ◽  
Benoit Baillif ◽  
Djork-Arné Clevert ◽  
David Rouquié ◽  
Joerg Wichard
Keyword(s):  
Gene Expression ◽  
Artificial Intelligence ◽  
De Novo ◽  
Molecular Design ◽  
Gene Expression Signature ◽  
Cellular Level ◽  
Generative Models ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Gene Expression Signatures

Finding new molecules with a desired biological activity is an extremely difficult task. In this context, artificial intelligence and generative models have been used for molecular <i>de novo</i> design and compound optimization. Herein, we report the first generative model that bridges systems biology and molecular design conditioning a generative adversarial network with transcriptomic data. By doing this we could generate molecules that have high probability to produce a desired biological effect at cellular level. We show that this model is able to design active-like molecules for desired targets without any previous target annotation of the training compounds as long as the gene expression signature of the desired state is provided. The molecules generated by this model are more similar to active compounds than the ones identified by similarity of gene expression signatures, which is the state-of-the-art method for navigating compound-induced gene expression data. Overall, this method represents a novel way to bridge chemistry and biology to advance in the long and difficult road of drug discovery.

Comprehensive Introduction to Neural Networks

2021 ◽  
pp. 24-48
Author(s):  
Rajesh Sai K. ◽  
Veneela Adapa ◽  
Hari Kishan Kondaveeti
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Neural Networks ◽  
Human Brain ◽  
Network Models ◽  
Neuron Models ◽  
Neural Network Models ◽  
The Neural Networks ◽  
Biological Neuron ◽  
Comprehensive Introduction

Unknowingly, artificial intelligence (AI) has become an inevitable part of our lives. In this chapter, the authors discuss how the neural networks, a sub-part of AI, changed the way we analyse things. In this chapter, the advent of neural networks, inspiration from the human brain, simplification models of biological neuron models are discussed. Later, a detailed overview of various neural network models, their strengths, limitations, applications, and challenges are presented in detail.

Sign in / Sign up

Export Citation Format

Share Document