SHREC

SHREC corrects errors in short-read data from second-generation sequencing technologies to improve the accuracy of de novo sequencing and downstream genomic analyses.

Key Features:

• Data structure: Uses a generalized suffix trie as the underlying data structure to facilitate error correction of short reads.
• Sensitivity and specificity: Achieves over 99% sensitivity and 96% specificity on simulated datasets with error rates up to 3%.
• Correction accuracy: Delivers error correction accuracy exceeding 80% on simulated data and surpassing 88% on real data.
• Sequencing focus: Tailored for short-read data from second-generation sequencing technologies and optimized for de novo short-read sequencing compared with methods tuned for shotgun sequencing.
• Scalability: Capable of processing up to 10 million short reads on standard workstations.
• Implementation: Implemented as a Java application.

Scientific Applications:

• De novo sequencing: Improves the reliability and accuracy of de novo sequence assembly by correcting sequencing errors in short reads.
• Short-read genomic analyses: Enhances downstream analyses that depend on high-quality short-read datasets generated by second-generation sequencing technologies.

Methodology:

SHREC employs a generalized suffix trie data structure and associated algorithms to identify and correct erroneous short reads.