Impact of Cattle Breed in scRNA-Seq Reference on Muscle Fiber Type Deconvolution from Bulk RNA-Seq: A Comparison of...

Area of Science:

• Genomics
• Bioinformatics
• Livestock Science

Background:

• Bulk RNA sequencing offers transcriptome-wide data but lacks cell-type resolution.
• Single-cell RNA sequencing (scRNA-seq) provides cellular detail but is costly.
• Deconvolution methods estimate cell proportions from bulk data, but accuracy depends on reference data and software.

Purpose of the Study:

• To investigate the impact of cattle breed in reference scRNA-seq data on deconvolution analysis of muscle fiber type proportions.
• To compare the performance of four deconvolution software tools (CIBERSORTx, Cellanneal, DeconvR-NNLS, DeconvR-RLM).

Main Methods:

• Analysis of bulk RNA sequencing data from *Longissimus dorsi* muscle of Holstein Friesian cows.
• Utilized scRNA-seq reference data from Brahman and Wagyu cattle breeds.
• Employed Kruskal-Wallis and Dunn's tests to assess statistical significance.

Main Results:

• The breed of the reference scRNA-seq data significantly influenced the estimated proportions of type IIa and IIx muscle fibers.
• Differences in muscle fiber type proportions were observed across the four deconvolution methods.
• This study is the first to demonstrate breed-specific impacts on deconvolution outcomes in cattle.

Conclusions:

• Cattle breed selection for reference scRNA-seq datasets is a critical factor influencing deconvolution accuracy.
• Future deconvolution studies in livestock genomics must consider breed compatibility between reference and target datasets.
• Careful consideration of breed matching is essential for reliable cell type proportion estimation.