AggloIndel

AggloIndel detects deletions and insertions by agglomeratively clustering short-read paired-end sequencing data to identify overlapping structural variations in mapped reads.

Key Features:

• Detection of deletions and insertions: Identifies deletions and insertions by clustering paired-end mappings from short-read Next-Generation Sequencing data.
• Overlapping deletion detection: Resolves potentially overlapping deletions to distinguish overlapping events from single indels.
• Assumption-free analysis: Operates without assumptions about sample number, sample heterogeneity, or polyploidy.
• Iterative agglomerative clustering: Employs agglomerative clustering that iteratively merges mappings using a similarity score that considers both putative indel location and size.
• Singleton-based error identification: Erroneous mappings typically appear as singleton clusters, enabling identification of mapping errors.
• Multi-sample paired-end analysis: Analyzes paired-end mapping data from multiple samples simultaneously to support comparison across samples.

Scientific Applications:

• Cancer genomics: Detects structural variations such as deletions relevant to tumor development and helps distinguish tumor-specific from patient-specific variations using multi-sample paired-end data.
• Complex structural variation analysis: Supports analysis of overlapping structural variants in complex, heterogeneous, or polyploid genomes.
• Mapping quality assessment: Facilitates data quality assessment by identifying singleton clusters as potential erroneous mappings.

Methodology:

Processes paired ends mapped to a reference genome and applies iterative agglomerative clustering that merges mappings based on a similarity score accounting for putative indel location and size; singleton clusters indicate erroneous mappings.