AEMDA

The software tool 'AEMDA' introduces a novel computational framework for identifying associations between microRNAs (miRNAs) and diseases. Leveraging a learning-based approach, AEMDA extracts dense, high-dimensional representations of diseases and miRNAs from integrated data, including disease semantic similarity, miRNA functional similarity, and heterogeneous related interaction data. A deep autoencoder within AEMDA efficiently captures underlying associations without the need for negative samples. The reconstruction error serves as a measure to predict disease-associated miRNAs.

Topic

Functional, regulatory and non-coding RNA;Pathology;Machine learning

Detail

  • Operation: miRNA target prediction;Data retrieval;miRNA expression analysis

  • Software interface: Library

  • Language: Python

  • License: -

  • Cost: Free

  • Version name: -

  • Credit: the National Natural Science Foundation of China.

  • Input: -

  • Output: -

  • Contact: Cunmei Ji jicm2015@mail.qfnu.edu.cn nijch@163.com zhengch99@126.com, Chunhou Zheng jicm2015@mail.qfnu.edu.cn

  • Collection: -

  • Maturity: -

Publications

  • AEMDA: inferring miRNA-disease associations based on deep autoencoder.
  • Ji C, et al. AEMDA: inferring miRNA-disease associations based on deep autoencoder. AEMDA: inferring miRNA-disease associations based on deep autoencoder. 2021; 37:66-72. doi: 10.1093/bioinformatics/btaa670
  • https://doi.org/10.1093/BIOINFORMATICS/BTAA670
  • PMID: 32726399
  • PMC: -

Download and documentation

< Back to DB search