Related Experiment Video
Updated: Aug 21, 2025
10:10
Three Differential Expression Analysis Methods for RNA Sequencing: limma, EdgeR, DESeq2
Published on: September 18, 2021
38.2K
Modeling and cleaning RNA-seq data significantly improve detection of differentially expressed genes
Igor V Deyneko1, Orkhan N Mustafaev2, Alexander А Tyurin3
1Laboratory of Functional Genomics, К.А. Timiryazev Institute of Plant Physiology RAS, Moscow, Russia, 127276. igor.deyneko@inbox.ru.
BMC Bioinformatics
|November 17, 2022
Summary
This study introduces RNAdeNoise, a method to clean RNA-sequencing data by removing technical noise. This improves the detection of differentially expressed genes, especially those with low to moderate expression levels.
Area of Science:
- Genomics
- Bioinformatics
- Molecular Biology
Background:
- RNA-sequencing (RNA-seq) is a standard for mRNA quantification.
- Biological and technical noise can weaken the detection of differences in rare mRNAs.
Purpose of the Study:
- To develop a method for cleaning RNA-seq data to improve gene expression detection.
- To specifically enhance the identification of differentially expressed genes with low to moderate transcription levels.
Main Methods:
- A data modeling approach was used to identify parameters of random mRNA counts.
- Reads likely originating from technical noise were identified and removed.
Main Results:
- The developed method significantly increases the number of detected differentially expressed genes.
- Improved statistical significance (p-values) was observed without bias towards low-count genes.
- The method demonstrated effectiveness across different organisms and sequencing technologies (Illumina, BGI).
Conclusions:
- RNAdeNoise effectively removes technical noise from RNA-seq data.
- The method enhances the detection of differentially expressed genes, particularly those with low to moderate expression.
- RNAdeNoise offers an objective alternative to subjective threshold settings for minimal RNA counts.
