Predicting of Drug-Disease Associations via Sparse Auto-Encoder-Based Rotation Forest

Han Jing Jiang, Zhu Hong You, Kai Zheng, Zhan Heng Chen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

Computational drug repositioning, designed to identify new indications for existing drugs, significantly reducing the cost and time involved in drug development. Confirming the association between drugs and disease is a critical process in drug development. At present, the most advanced method is to apply the recommendation system or matrix decomposition method to predict the similarity between drugs and diseases. In this paper, the association between drugs and diseases is integrated and a novel computational method based on sparse auto-encoder combined with rotation forest (SAEROF) is proposed to predict new drug indications. First, we constructed a drug-disease similarity based on drug-disease association and integrated it into sparse auto-encoder to obtain the final drug-disease similarity. Then, we adopt a rotation forest algorithm to predicted scores for unknown drug–disease pairs. Cross validation and independent test results show that this model is better than the existing model and has reliable prediction performance. In addition, the case study of two diseases further proves the practical value of this method, and the results obtained can be found in CTD database.

Original languageEnglish
Title of host publicationIntelligent Computing Methodologies - 15th International Conference, ICIC 2019, Proceedings
EditorsDe-Shuang Huang, Zhi-Kai Huang, Abir Hussain
PublisherSpringer Verlag
Pages369-380
Number of pages12
ISBN (Print)9783030267650
DOIs
StatePublished - 2019
Externally publishedYes
Event15th International Conference on Intelligent Computing, ICIC 2019 - Nanchang, China
Duration: 3 Aug 20196 Aug 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11645 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference15th International Conference on Intelligent Computing, ICIC 2019
Country/TerritoryChina
CityNanchang
Period3/08/196/08/19

Keywords

  • Drug-disease associations
  • Rotation forest
  • Sparse auto-encoder

Fingerprint

Dive into the research topics of 'Predicting of Drug-Disease Associations via Sparse Auto-Encoder-Based Rotation Forest'. Together they form a unique fingerprint.

Cite this