UMI-tools

UMI-tools implements network-based methods to process Unique Molecular Identifiers (UMIs) in high-throughput sequencing (HTS) and next-generation sequencing (NGS) data to identify PCR duplicates and improve molecular quantification by accounting for UMI sequencing errors.

Key Features:

• UMI handling: Treats UMIs as random oligonucleotide barcodes to distinguish identical sequences originating from different molecules versus PCR duplicates.
• Sequencing error correction: Accounts for sequencing errors within UMI sequences when resolving duplicate reads.
• Network-based deduplication: Implements a network-based methodology to group related UMIs and identify PCR duplicates.
• Quantification improvement: Improves molecular quantification accuracy by resolving UMI-derived ambiguities.
• Validation on simulated data: Performance has been assessed using simulated datasets.
• Validation on real datasets: Performance has been demonstrated on real datasets including iCLIP and single-cell RNA-seq.
• Reproducibility and clustering benefits: Application of the method yields improved reproducibility between iCLIP replicates and improved clustering in single-cell RNA-seq analyses.

Scientific Applications:

• iCLIP analysis: Enhances reproducibility between iCLIP replicates by accounting for UMI sequence errors during duplicate removal.
• Single-cell RNA sequencing: Improves clustering and molecular quantification in single-cell RNA-seq experiments by resolving UMI-related biases.
• NGS quantification: Reduces PCR duplicate–derived bias in next-generation sequencing experiments to enable more accurate molecule counts.

Methodology:

Applies a network-based method that accounts for UMI sequencing errors when grouping UMIs and identifying PCR duplicates.