Statistical evaluation of transcriptomic data generated using the Affymetrix one-cycle, two-cycle and IVT-Express RNA...

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Microarrays enable global RNA expression profiling.
• Small tissue samples or isolated cells pose challenges for RNA extraction and labeling.
• RNA amplification is crucial for transcriptomic analyses from limited biological material.

Purpose of the Study:

• To evaluate potential gene representation bias in transcriptomic data generated by different RNA amplification protocols.
• To compare the standard one-cycle, two-cycle (small-sample), and IVT-Express protocols on the Affymetrix ATH1 array.
• To assess the reliability of two-cycle amplification for Arabidopsis root samples.

Main Methods:

• Comparative analysis of transcriptomic data from Arabidopsis root samples.
• Utilized Affymetrix one-cycle, two-cycle, and IVT-Express labeling protocols.
• Validated findings using two independent publicly available datasets.

Main Results:

• Transcriptomic data from all tested protocols were broadly similar.
• Identified 35 probe sets (out of 22810) with misrepresented gene expression in two-cycle data.
• 33 of these probe sets were confirmed as mis-amplified across independent datasets.

Conclusions:

• Caution is advised when using two-cycle amplification data due to potential probe set misrepresentation.
• The Affymetrix IVT-Express labeling protocol demonstrates reduced bias compared to two-cycle amplification.
• Recommends the IVT-Express kit for new transcriptomic experiments involving small samples.