Artificial Neural Network for Drug Design, Delivery and Disposition 2016
DOI: 10.1016/b978-0-12-801559-9.00002-8
|View full text |Cite
|
Sign up to set email alerts
|

The Role of Artificial Neural Networks on Target Validation in Drug Discovery and Development

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
6
0

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(6 citation statements)
references
References 57 publications
0
6
0
Order By: Relevance
“…[ 34 ] It is considered an important nonlinear modeling technique and widely used for QSAR studies. [ 35,36 ]…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…[ 34 ] It is considered an important nonlinear modeling technique and widely used for QSAR studies. [ 35,36 ]…”
Section: Methodsmentioning
confidence: 99%
“…[34] It is considered an important nonlinear modeling technique and widely used for QSAR studies. [35,36] The most common neural architecture used in QSAR research is the back-propagation (BP) method (multilayer feed forward BP). [37] Their principle is based on adapting the weights parameters, WK, to decrease the error between the output given by the network (calculated activity) and the desired output (experimental activity).…”
Section: Artificial Neural Networkmentioning
confidence: 99%
“…Recurrent Neural Networks (RNNs) represent an advanced subclass of ANNs tailored for processing sequential data, making them instrumental in fields such as natural language processing, genomics, and notably, drug discovery. [135,147] Unlike traditional ANNs, RNNs are equipped with loops in their architecture, allowing the network to retain information from one step and pass it to the next, facilitating the processing of variable-length input sequences. This capacity enables RNNs to capture dynamic temporal behaviors crucial for analyzing timeseries data, genomic sequences, protein structures, and Simplified Molecular Input Line Entry System (SMILES) strings.…”
Section: Recurrent Neural Networkmentioning
confidence: 99%
“…These networks are composed of interconnected units or artificial neurons that collaboratively process information, mirroring the functioning of biological neural networks [134] . Central to an ANN′s architecture are multiple layers: the input layer, which receives the initial data; one or more hidden layers, which transform the data through complex relationships; and the output layer, where the final predictions or classifications are made [135136] . This layered structure allows ANNs to excel in learning from data, identifying patterns, and making informed predictions across various domains [137,138] …”
Section: Artificial Neural Networkmentioning
confidence: 99%
“…Due to their simplicity and easy implementation, some computer-based design of experiments (DoE) methods have been proposed. RSMs have a high sensitivity to independent variable composition and constitute one of the most widely used methods [78,107,108]. On one hand, an RSM is based on first-, second-or third-order low dimensional polynomial equations [109], leading to its lack of fitting and optimization abilities for a multiple response model based on limited variables.…”
Section: Selection and Optimization Of Formulationsmentioning
confidence: 99%