mirnaQC: a webserver for comparative quality control of miRNA-seq data

Area of Science:

• Bioinformatics
• Molecular Biology
• Genomics

Background:

• MicroRNA sequencing (miRNA-seq) is widely used but often lacks comprehensive quality control.
• Current quality control methods are limited, often relying solely on PhredScore, failing to detect issues like low yield or contaminants.
• Assessing critical quality aspects such as microRNA yield, degradation products (e.g., rRNA fragments), and adapter-dimer percentages is challenging with absolute thresholds.

Purpose of the Study:

• To present mirnaQC, a novel webserver designed for robust miRNA-seq quality control.
• To provide a user-friendly tool that assesses multiple quality parameters for improved data interpretation.
• To establish a benchmark for miRNA-seq quality using a large reference dataset.

Main Methods:

• Development of the mirnaQC webserver, incorporating 34 distinct quality control parameters.
• Utilizing a reference distribution derived from over 36,000 public miRNA-seq datasets for parameter ranking.
• Accepting FASTQ files and SRA accessions as input, providing detailed results with visualizations like PCA and heatmaps.

Main Results:

• The mirnaQC webserver effectively assesses various quality attributes crucial for miRNA-seq data.
• Results are presented with clear categorizations for potential technical artifacts in library preparation, sequencing, contamination, and yield.
• Comparative analysis of datasets using PCA and heatmaps aids in identifying underlying quality issues.

Conclusions:

• mirnaQC offers a significant advancement in miRNA-seq quality control beyond basic PhredScore analysis.
• The tool enhances the interpretability of quality metrics by comparing them against a large reference dataset.
• Demonstrated utility in identifying diverse quality issues in publicly available miRNA-seq datasets, improving data reliability.