Fast Dual-Regularized Autoencoder for Sparse Biological Data

9 pages•Published: July 12, 2024

Abstract

Relationship inference from sparse data is an important task with applications ranging from product recommendation to drug discovery. A recently proposed linear model for sparse matrix completion has demonstrated surprising advantage in speed and accuracy over more sophisticated recommender systems algorithms. Here we extend the linear model to develop a shallow autoencoder for the dual neighborhood-regularized matrix completion problem. We demonstrate the speed and accuracy advantage of our approach over the existing state-of-the-art in predicting drug-target interactions and drug-disease associations.

Keyphrases: biological relationship inference, drug repurposing, drug target interactions, recommender systems, sparse matrix completion

In: Hisham Al-Mubaid, Tamer Aldwairi and Oliver Eulenstein (editors). Proceedings of the 16th International Conference on Bioinformatics and Computational Biology (BICOB-2024), vol 101, pages 113-121.

Links:	https://easychair.org/publications/paper/PcSL
	https://doi.org/10.29007/v896

BibTeX entry

@inproceedings{BICOB-2024:Fast_Dual_Regularized_Autoencoder,
  author    = {Aleksandar Poleksic},
  title     = {Fast Dual-Regularized Autoencoder for Sparse Biological Data},
  booktitle = {Proceedings of the 16th International Conference on Bioinformatics and Computational Biology (BICOB-2024)},
  editor    = {Hisham Al-Mubaid and Tamer Aldwairi and Oliver Eulenstein},
  series    = {EPiC Series in Computing},
  volume    = {101},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {/publications/paper/PcSL},
  doi       = {10.29007/v896},
  pages     = {113-121},
  year      = {2024}}

Download PDF Open PDF in browser