DEG-by-Index Ratio Transformation Normalization with Blood RNA-Seq Enhances Early and Consistent Detection of Mouse Tumorigenesis

TL;DR

A new normalization method called DiRT improves early detection of tumors in mice using blood RNA-Seq data.

Contribution

DiRT enables earlier and more consistent detection of tumor-related gene changes in mouse blood RNA-Seq data.

Findings

DiRT separates tumor and control samples as early as three days after tumor induction.

DiRT-derived genes are enriched in the platelet activation signaling pathway.

Standard methods detect changes later or inconsistently compared to DiRT.

Abstract

Identifying early signals of tumors from blood using RNA-Seq is challenging because differences between samples can hide important changes in gene activity. In this study, we applied a new method called DiRT to analyze mouse blood RNA-Seq data. DiRT was able to clearly separate tumor samples from healthy ones at the earliest stages and consistently track changes as the disease progressed. In contrast, standard methods often detected differences only later or inconsistently. These results show that DiRT improves the sensitivity and reliability of blood RNA-Seq analysis, making it easier to detect tumor-related signals early in the disease.

Variability in blood RNA-Seq data can obscure transcriptional changes that reflect tumor responses, and conventional normalization methods such as RLE/DESeq2 or TMM/edgeR often fail to capture these changes consistently. To address this challenge, we applied DiRT (DEG-by-index Ratio Transformation), a normalization and analysis strategy previously used in insect models, to 111 blood RNA-Seq datasets from mouse tumorigenesis models. DiRT achieved clearer separation between tumor and control samples as early as three days after tumor induction and maintained consistent marker signals across all stages of disease progression. In contrast, standard methods typically revealed differences only at later or scattered time points. KEGG pathway analysis further showed that DiRT-derived differentially expressed genes (DEGs) were enriched in the platelet activation signaling pathway, a pathway not identified using RLE/DESeq2 or TMM/edgeR. These findings demonstrate that DiRT enhances both sensitivity and reproducibility, enabling earlier and more consistent detection of transcriptional changes in blood during tumor development.

Figures