HQAlign: aligning nanopore reads for SV detection using current-level modeling.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Detecting structural variants (SVs) is crucial for understanding human diseases.
• Accurate alignment of long DNA reads is essential for identifying novel SVs.
• Nanopore sequencing offers long reads but presents alignment challenges due to high error rates.

Purpose of the Study:

• To design and evaluate HQAlign, a novel aligner for SV detection using nanopore sequencing reads.
• To leverage nanopore sequencing physics to improve alignment accuracy for SV detection.
• To enhance existing alignment pipelines for more robust SV identification.

Main Methods:

• HQAlign utilizes base-called nanopore reads and incorporates nanopore sequencing physics to improve alignments.
• The aligner integrates SV-specific modifications into the alignment pipeline.
• HQAlign adapts these improvements into the state-of-the-art long-read aligner, minimap2.

Main Results:

• HQAlign identifies 4%-6% more SVs missed by minimap2, with comparable standalone performance on real nanopore data.
• Breakpoint accuracy for common SV calls is improved by 10%-50% with HQAlign.
• Alignment rates are improved for nanopore reads against both the CHM13 and GRCh37 human genome assemblies.

Conclusions:

• HQAlign offers a significant advancement in SV detection using nanopore long-read data.
• The tool enhances accuracy and recall for SV identification, addressing limitations of current aligners.
• HQAlign provides a valuable resource for genomic research, particularly in disease-related structural variation studies.