Probe Region Expression Estimation for RNA-Seq Data for Improved Microarray Comparability.

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• Public gene expression databases offer vast data for human biology and disease research.
• Integrating data from diverse measurement platforms like microarrays and RNA-sequencing (RNA-seq) is challenging.
• The volume of new microarray studies submitted to public databases currently exceeds that of RNA-seq studies.

Purpose of the Study:

• To develop a novel method for processing RNA-sequencing (RNA-seq) data.
• To enhance the comparability of gene expression data across different platforms, particularly between RNA-seq and microarrays.
• To facilitate the integration of diverse gene expression datasets for more comprehensive biological insights.

Main Methods:

• Proposed a new RNA-seq data processing method named PREBS (Probe Region-based Expression Estimation).
• PREBS estimates gene expression from RNA-seq reads that overlap microarray probe regions.
• Standard microarray summarization algorithms are then applied to these estimates.

Main Results:

• PREBS generates gene expression estimates from RNA-seq that are highly similar to those derived from microarrays.
• Utilizing paired microarray and RNA-seq samples from the TCGA LAML dataset, PREBS demonstrated superior similarity compared to other RNA-seq processing methods.
• Gene signatures derived using PREBS estimates showed significantly higher accuracy in retrieving paired microarray samples via an RNA-seq query.

Conclusions:

• PREBS effectively bridges the gap between RNA-seq and microarray data, significantly improving cross-platform comparability.
• The method enables novel applications, including estimating expression for microarray probe sets using RNA-seq data.
• An implementation of PREBS is available as a Bioconductor package, promoting its adoption in the research community.