Research on the Application of Deep Learning in Text Generation

In this paper we propose the generation of synthetic small and more sophisticated molecule structures that optimize the binding affinity to a target (ASYNT-GAN). To achieve this we leverage on three important achievements in A.I.: Attention, Deep Learning on Graphs and Generative Adversarial Networks. Similar to text generation based on parts of text we are able to generate a molecule architecture based on an existing target. By adopting this approach, we propose a novel way of searching for existing compounds that are suitable candidates. Similar to question and answer Natural Language solutions we are able to find drugs with highest relevance to a target. We are able to identify substructures of the molecular structure that are the most suitable for binding. In addition, we are proposing a novel way of generating the molecule in 3D space in such a way that the binding is optimized. We show that we are able to generate compound structures and protein structures that are optimised for binding to a target.

Download Full-text

De Novo Drug Design using Artificial Intelligence ASYNT-GAN

10.20944/preprints202010.0196.v1 ◽

2020 ◽

Author(s):

Ivan Jacobs ◽

Manolis Maragoudakis

Keyword(s):

Molecular Structure ◽

Artificial Intelligence ◽

Deep Learning ◽

De Novo ◽

Protein Structures ◽

Generative Adversarial Networks ◽

Text Generation ◽

De Novo Drug Design ◽

3D Space ◽

Adversarial Networks

In this paper we propose the generation of synthetic small and more sophisticated molecule structures that optimize the binding affinity to a target (ASYNT-GAN). To achieve this we leverage on three important achievements in A.I.: Attention, Deep Learning on Graphs and Generative Adversarial Networks. Similar to text generation based on parts of text we are able to generate a molecule architecture based on an existing target. By adopting this approach, we propose a novel way of searching for existing compounds that are suitable candidates. Similar to question and answer Natural Language solutions we are able to find drugs with highest relevance to a target. We are able to identify substructures of the molecular structure that are the most suitable for binding. In addition, we are proposing a novel way of generating the molecule in 3D space in such a way that the binding is optimized. We show that we are able to generate compound structures and protein structures that are optimised for binding to a target.

Download Full-text

A Research on Automatic Handwritten Devnagari Text Generation in Different Styles Using Recurrent Neural Network (Deep Learning) Especially for Marathi Script

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.b1154.0982s1119 ◽

2019 ◽

Vol 8 (2S11) ◽

pp. 938-942

Keyword(s):

Neural Network ◽

Deep Learning ◽

Recurrent Neural Network ◽

Text Generation ◽

Vector Machines ◽

Head Element ◽

Class Models

The point of handwritten numeral reputation (HNR) framework is to order input numeral all in all of k classifications. There are standard HNR frameworks have 2 elements: handwritten numeral popularity. In spotlight exam step, data relevant as an example classifier. the example arrangement step names the numeral by means of and large of k classifications exploitation the class models. in the course of the years, right savvy amount of labor has been allotted in the space of HNR. Fluctuated methods are arranged within the writing for characterization of composed numerals. those hold close Hough changes, visible diagram methods, head element research, and bolster vector machines, closest neighbor methods, neural figuring and fluffy essentially based totally methodologies

Download Full-text