A computational method to improve analysis of single cell RNA-sequencing

A computational method to improve analysis of single cell RNA-sequencing hfabryova Fri, 09/06/2024 - 11:56 June 21, 2024 BASIC NIAMS IRP Vittorio Sartorelli, M.D. Laboratory of Muscle Stem Cells and Gene Regulation Description Single-cell RNA sequencing (scRNA-seq) provides high-resolution details of cell-to-cell variation in gene expression, which offers insights into differences between cell types , how they change over time, differentiating dynamics, and disease mechanisms. However, challenges such as low capture rates and dropout events can introduce noise in data analysis. In this report, researchers described a deep neural generative framework, the dynamic batching adversarial autoencoder (DB-AAE), which excels at denoising scRNA-seq datasets. DB-AAE directly captures optimal features from input data and enhances feature preservation, including cell type-specific gene expression patterns. What is exciting about this article? The authors showed that their approach surpasses the performance of other methods in terms of denoising accuracy and preservation of biological signal. These results not only validate the effectiveness of the DB-AAE approach but also emphasize its potential as a valuable tool for improving the quality and reliability of downstream analyses in scRNA-seq analysis. Grant Support ZIA AR041126 Computational Biology Genetics and Genomics A deep learning framework for denoising and ordering scRNA-seq data using adversarial autoencoder with dynamic batching. Ko KD, Sartorelli V STAR Protoc. 2024 Jun 21; 5 (2). doi: 10.1016/j.xpro.2024.103067 PMID: 38748883 Research reported in this publication was supported by the Intramural Research Program of the NIHʼs National Institute of Arthritis and Musculoskeletal and Skin Diseases.