misFinder

misFinder identifies and corrects mis-assemblies in genome assemblies by integrating reference-genome comparison and paired-end read alignment to distinguish assembly errors from structural variations.

Key Features:

• Unbiased Error Identification: Integrates reference-genome comparison and aligned paired-end reads to distinguish mis-assemblies from true structural variations.
• High Accuracy in Error Detection: Leverages a reference genome (or closely related references) together with coverage data and insert-distance consistency features derived from paired-end reads to make high-confidence error calls.
• Reduction of False Positives/Negatives: Minimizes false positives and false negatives in mis-assembly detection to improve the reliability of downstream genomic analyses.
• Performance Superiority: Demonstrated superior detection of true mis-assemblies with fewer false calls than QUAST and REAPR on simulated and real paired-end read datasets.

Scientific Applications:

• Variant Detection: Improves accuracy of variant calling by correcting mis-assemblies that could confound variant detection.
• Gene Annotation: Supports reliable gene annotation by ensuring contiguous and correctly assembled gene loci.
• Comparative Genomics: Enables accurate comparative genomics by distinguishing structural variation from assembly artifacts.

Methodology:

Combines reference-genome comparison with paired-end read alignment and assesses coverage consistency and insert-distance features to identify and correct mis-assembled positions and to discriminate structural variations from assembly errors.