论文信息 - A Survey of Missing Value Imputation for Gene Expression Data Using Deep Learning Models

A Survey of Missing Value Imputation for Gene Expression Data Using Deep Learning Models

ion Gene expression analysis is essential for transcriptomics studies. However, due to unintended noise generation by the device or an insufficient number of mRNA molecules, the rate of missing values, which are false zero, is high. Therefore, a number of deep learning-based studies have been conducted to impute missing values. In this paper, we will overview the state-of-the-art deep learning-based models: DeepImpute, AutoImpute, Deep Count Autoencoder Network (DCA), scGAIN, scIGANs.

Jaeyoon Kim | J. Seok | Junhee Seok

[1] Jaeyoon Kim,et al. A Survey of Missing Data Imputation Using Generative Adversarial Networks , 2020, 2020 International Conference on Artificial Intelligence in Information and Communication (ICAIIC).

[2] Zhigang Zhang,et al. scIGANs: single-cell RNA-seq imputation using generative adversarial networks , 2020, bioRxiv.

[3] Soheil Feizi,et al. scGAIN: Single Cell RNA-seq Data Imputation using Generative Adversarial Networks , 2019, bioRxiv.

[4] Fabian J Theis,et al. Single-cell RNA-seq denoising using a deep count autoencoder , 2019, Nature Communications.

[5] A. Majumdar,et al. AutoImpute: Autoencoder based imputation of single-cell RNA-seq data , 2018, Scientific Reports.

[6] Lana X. Garmire,et al. DeepImpute: an accurate, fast, and scalable deep neural network method to impute single-cell RNA-seq data , 2018, Genome Biology.