Medicament module pharmacokinetic property and toxicity predicting method based on capsule network

A technology of pharmacokinetics and drug molecules, which is applied in the field of pharmacokinetic properties and toxicity prediction of drug molecules based on capsule network, can solve the large dependence of training set size, loss of original information of molecular fingerprints and molecular descriptors, Problems such as poor prediction and classification

Inactive Publication Date: 2019-07-05
SICHUAN UNIV
View PDF5 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method is based on the molecular fingerprints and molecular descriptors of ligands, and uses deep learning capsule networks to establish the relationship between molecular fingerprints and molecular descriptors, pharmacokinetic properties and toxicity, and overcomes the i

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Medicament module pharmacokinetic property and toxicity predicting method based on capsule network
  • Medicament module pharmacokinetic property and toxicity predicting method based on capsule network
  • Medicament module pharmacokinetic property and toxicity predicting method based on capsule network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0138] Predict the active compound of the potassium ion channel encoded by hERG (human ether-a-go-go-related gene), and use the convolution operation as a feature extractor to obtain the low-level features of the molecule. The implementation process is as follows:

[0139] In the first step, data related to hERG activity are collected. Data related to hERG were obtained from the ChEMBL open source database (https: / / www.ebi.ac.uk / chembl / ). ChEMBL is a well-known biological activity database established by the European Institute of Bioinformatics. Anyone can download this public database from the website, so it is widely used by cheminformatics researchers. The workflow for establishing the initial ChEMBL-hERG dataset is as follows:

[0140] 1) According to the ID number of hERG in the data (ChEMBL 240), 17,952 compounds that were tested for hERG activity were extracted;

[0141] 2) Compounds identified as "Nonstandard unit for type", "Outside typical range" (outside typical...

Embodiment 2

[0152] Still predicting the active compounds of the potassium ion channel encoded by hERG (human ether-a-go-go-related gene), a restricted Boltzmann machine was used as a feature extractor.

[0153] The first and second steps are the same as in Example 1.

[0154] In the third step, the prediction model of hERG active / inactive molecules based on the capsule network was established with the ChEMBL-hERG training set. Initialize network weights randomly using a truncated normal distribution and set stddev to 0.01. The feature extractor uses a restricted Boltzmann machine. The probability distribution of the energy function is used as the activation function. To reduce internal covariate shift, the input distribution of each layer is normalized to a standard Gaussian distribution using a batch normalization method. Adam method is used for network optimization. By monitoring multiple evaluation indicators (accuracy, specificity, sensitivity and Matthews correlation coefficient,...

Embodiment 1、2

[0162] The present embodiment 1, 2 adopts the following formula to carry out evaluation verification:

[0163]

[0164]

[0165]

[0166] Where Q represents the overall prediction accuracy of the prediction model, SE represents sensitivity, which refers to the proportion of positive / active compounds correctly predicted by the prediction model, and SP represents specificity, which refers to the proportion of negative / inactive compounds correctly predicted by the prediction model .

[0167] When convolution and restricted Boltzmann machine are used as feature extractors, the overall prediction accuracy of the test set is about 90%, indicating that the established model also has a good predictive ability for compounds independent of the training set.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a medicament module pharmacokinetic property and toxicity predicting method based on a capsule network. After a comprehensive module fingerprint and a module descriptor are constructed and early-period preparing operation for establishing model is performed, a low-grade characteristic content of a molecule is extracted from an upper-layer low-grade characterized through convolutional or restricted Boltzmann machine operation; then a capsule network method is used for abstracting the high-grade characteristic of the molecule in a lower-layer high-grade characteristic; a relation between the high-grade characteristic and an active label is fit through a dynamic routing algorithm, thereby predicting the pharmacokinetic property and the toxicity class of an unknown smallmolecule. The method does not require collection of large scale datasets for training, optimization is performed on input through end-to-end and furthermore automatic dimension reduction is realized.A coupling coefficient is updated through iterating a dynamic routing process. The dynamic routing conveys all characteristics of an upper-layer capsule to a random lower-layer capsule, thereby greatly reserving a hierarchical position relation. The method realizes a better predicting effect than that of a traditional machine learning method.

Description

technical field [0001] The invention relates to the field of computer-aided drug molecule design, in particular to a method for predicting pharmacokinetic properties and toxicity of drug molecules based on a capsule network. Background technique [0002] The great success of a drug depends not only on its good efficacy, but also on its excellent pharmacokinetic properties and low toxicity. According to statistics, the poor absorption, distribution, metabolism, excretion and toxicity of candidate drugs account for more than 50% of the reasons for the failure of drug development. Therefore, it is possible to exclude and optimize compounds with poor pharmacokinetic properties and toxicity in the early stage of drug development Greatly improve the success rate of drug development. In recent years, although in vitro high-throughput screening methods can be used to measure the pharmacokinetic properties and toxicity of compounds, experimental-based assays are not only expensive a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G16C10/00G16C20/30G16C20/70
Inventor 杨胜勇王译伟邹俊黄磊姜斯文
Owner SICHUAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap