ChimeraMiner: An Improved Chimeric Read Detection Pipeline and Its Application in Single Cell Sequencing

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• Multiple displacement amplification (MDA) is a widely used whole genome amplification technique for single-cell studies.
• MDA, while effective, generates chimeric reads that disrupt downstream analyses.
• Accurate detection of these chimeric sequences is crucial for reliable single-cell genomics.

Purpose of the Study:

• To develop and evaluate ChimeraMiner, an improved pipeline for detecting and classifying chimeric reads from MDA sequencing data.
• To compare the efficiency and performance of ChimeraMiner against existing methods.
• To assess the impact of chimera removal on structural variation detection in single-cell datasets.

Main Methods:

• Construction of the ChimeraMiner pipeline for analyzing MDA sequencing data.
• Classification of chimeric sequences using ChimeraMiner.
• Evaluation using two MDA datasets (MDA1 and MDA2) and comparison with a previous pipeline.
• Application to single-cell datasets to assess structural variation detection.

Main Results:

• ChimeraMiner demonstrated significantly improved processing efficiency, using only 43.4% of the time compared to the previous pipeline.
• The pipeline accurately classified millions of chimeric read pairs, identifying a substantial number of novel chimeras missed by prior methods.
• Application of ChimeraMiner effectively removed 83.8% of false positive structural variations in single-cell datasets.

Conclusions:

• ChimeraMiner offers enhanced efficiency and accuracy in detecting and classifying chimeric reads from MDA-based single-cell sequencing.
• The pipeline effectively mitigates false positives in structural variation detection, improving data reliability.
• ChimeraMiner is a promising tool for widespread adoption in single-cell sequencing research.