Leveraging long-read assemblies and machine learning to enhance short-read transposable element detection and genotyping
GENETICS · 03.06.2026 · 233 (2)
DOI: 10.1093/genetics/iyag101 ↗Kurz gefasst
Fachartikel in der Zeitschrift GENETICS, online erschienen am 17.04.2026. Verfasst von Austin Daigle und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Genetics, University of North Carolina, Chapel Hill, NC 27599. Hauptthemen laut MEDLINE: DNA Transposable Elements, Machine Learning, Genotyping Techniques, Sequence Analysis, DNA. Die Arbeit nennt 61 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS, NIH HHS, National Institute of General Medical Sciences.
Abstract (Originalsprache)
Transposable elements (TEs) are parasitic genomic elements that are ubiquitous across the tree of life and play a crucial role in genome evolution. Advances in long-read sequencing have allowed highly accurate TE detection, though at a higher cost than short-read sequencing. Recent studies using long reads have shown that existing short-read TE detection methods perform inadequately when applied to real data. In this study, we use a machine learning approach (called TEforest) to discover and genotype TE insertions and deletions with short-read data by using TEs detected from Drosophila melanogaster long-read genome assemblies as training data. Our method first uses a highly sensitive algorithm to discover potential TE insertion or deletion sites in the genome, extracting relevant features from short-read alignments. To discriminate between true and false TE insertions, we train a gradient-boosted decision tree model with a labeled ground-truth dataset for which we have calculated the same set of short-read features. We conduct a comprehensive benchmark of TEforest and traditional TE detection methods using real data from D. melanogaster and humans, finding that TEforest identifies more true positives and fewer false positives across datasets with different read lengths and coverages, while also accurately inferring genotypes and the precise breakpoints of insertions. By learning short-read signatures of TEs previously only discoverable using long reads, our approach bridges the gap between large-scale population genetic studies and the accuracy of long-read assemblies. This work provides a user-friendly tool to study the prevalence and phenotypic effects of TE insertions across the genome.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- Department of Genetics, University of North Carolina, Chapel Hill, NC 27599.
- Curriculum in Bioinformatics and Computational Biology, University of North Carolina, Chapel Hill, NC 27599.
- Department of Ecology and Evolutionary Biology, University of California, Irvine, CA 92697.
ORCID-Profile: Austin Daigle, J J Emerson, Daniel R Schrider
Förderung
- NIGMS NIH HHS
- NIH HHS
- National Institute of General Medical Sciences
- National Institutes of Health
Zitieren
APA
Daigle, A., Whitehouse, L. S., Zhao, R., Emerson, J. J., & Schrider, D. R. (2026). Leveraging long-read assemblies and machine learning to enhance short-read transposable element detection and genotyping. GENETICS, 233(2). https://doi.org/10.1093/genetics/iyag101
Vancouver
Daigle A, Whitehouse LS, Zhao R, Emerson JJ, Schrider DR. Leveraging long-read assemblies and machine learning to enhance short-read transposable element detection and genotyping. GENETICS. 2026;233(2). doi:10.1093/genetics/iyag101
BibTeX
@article{Daigle2026,
author = {Daigle, Austin and Whitehouse, Logan S and Zhao, Roy and Emerson, J J and Schrider, Daniel R},
title = {Leveraging long-read assemblies and machine learning to enhance short-read transposable element detection and genotyping},
journal = {GENETICS},
year = {2026},
volume = {233},
number = {2},
doi = {10.1093/genetics/iyag101}
}
EUVIMED