Related Experiment Video
Updated: Dec 25, 2025
G2-seq: A High Throughput Sequencing-based Technique for Identifying Late Replicating Regions of the Genome
Published on: March 22, 2018
Evaluating Structural Variation Detection Tools for Long-Read Sequencing Datasets in Saccharomyces cerevisiae.
Mei-Wei Luan1, Xiao-Ming Zhang2, Zi-Bin Zhu1
1Key Laboratory of Genetics and Germplasm Innovation of Tropical Special Forest Trees and Ornamental Plants (Ministry of Education), Hainan Key Laboratory for Biology of Tropical Ornamental Plant Germplasm, College of Forestry, Hainan University, Haikou, China.
This study evaluated structural variation (SV) detection tools for long-read sequencing data in yeast. PBHoney demonstrated the highest accuracy, while Picky performed the lowest among the tested callers.
Area of Science:
- Genomics
- Bioinformatics
Background:
- Structural variations (SVs) are key genetic variations influencing phenotypes and diseases.
- Long-read sequencing advances SV identification, but tool performance comparison is lacking.
Purpose of the Study:
- To evaluate and compare the accuracy of different structural variation detection tools using long-read sequencing data.
- To establish a robust analysis workflow for assessing SV detection performance.
Main Methods:
- Developed an analysis workflow combining NGMLR and minimap2 aligners with five SV callers: Sniffles, Picky, smartie-sv, PBHoney, and NanoSV.
- Utilized six Saccharomyces cerevisiae datasets for comprehensive evaluation.
- Validated SV region accuracy through re-alignment with diverse tools and conditions.
Main Results:
- PBHoney achieved the highest average accuracy (89.04%), while Picky had the lowest (35.85%).
- NanoSV, Sniffles, and smartie-sv showed accuracies of 68.67%, 60.47%, and 57.67%, respectively.
- PacBio sequencing yielded significantly more SVs than ONT (p = 0.000173).
Conclusions:
- Tool performance in SV detection varies significantly, with PBHoney being the most accurate.
- The choice of SV detection tool and sequencing platform impacts detection yield and accuracy.
