Reproducibility, bioinformatic analysis and power of the SAGE method to evaluate changes in transcriptome

Area of Science:

• Molecular Biology
• Genomics
• Bioinformatics

Background:

• Serial Analysis of Gene Expression (SAGE) is a powerful technique for global gene expression profiling.
• Assessing the reproducibility of SAGE is crucial for reliable experimental outcomes.
• Previous assessments of SAGE reproducibility have been limited.

Purpose of the Study:

• To rigorously evaluate the reproducibility of the SAGE method.
• To identify key factors influencing SAGE reproducibility.
• To provide recommendations for enhancing SAGE data accuracy.

Main Methods:

• Statistical analysis of SAGE data, including determination coefficient (R2).
• Identification and analysis of factors affecting reproducibility, such as concatemer/ditag replication and PCR amplification.
• Bioinformatic analysis to correct for artifacts and improve data quality.

Main Results:

• SAGE demonstrated a high determination coefficient (R2 = 0.96) for reproducibility.
• Factors like concatemer/ditag replication, sequencing depth, and PCR amplification significantly impact reproducibility.
• Bioinformatic analysis effectively eliminated artifacts.
• Increased sequencing tag numbers enhanced the method's power to detect transcripts and their regulation.

Conclusions:

• The SAGE method is highly reproducible when key factors are controlled.
• Bioinformatic analysis and optimization of sequencing depth are essential for accurate SAGE results.
• This study provides a framework for improving the reliability and sensitivity of SAGE experiments.