Reptile

Reptile: k-mer Spectrum-Based Error Correction for Short-Read NGS Data

Reptile performs error correction of short-read sequencing data from next-generation sequencing (NGS) technologies using k-mer spectrum analysis to identify and correct sequencing errors.

Key Features:

• k-mer Spectrum Analysis: Detects erroneous k-mers by analyzing frequency distributions within input reads.
• Hamming Distance Correction: Identifies and corrects candidate erroneous k-mers based on Hamming distance.
• Contextual Neighbor Evaluation: Examines adjacent k-mers within the same read to ensure sequence context consistency.
• Quality Score Integration: Incorporates sequencing quality scores to refine base correction decisions.
• Memory-Efficient Processing: Processes datasets exceeding available RAM without storing all input reads in main memory.

Scientific Applications:

• Preprocessing of High-Throughput Sequencing Data: Improves base accuracy in large-scale genomic datasets prior to downstream analyses such as genome assembly and variant detection.

Methodology:

Reptile analyzes k-mer frequency spectra from short-read datasets, identifies low-frequency k-mers as potential errors, applies Hamming distance-based correction to candidate bases, and validates corrections using neighboring k-mer context within each read while optionally integrating sequencing quality scores.