Automatic curation of LTR retrotransposon libraries from plant genomes through machine learning

Abstract

Transposable elements are mobile sequences that can move and insert themselves into chromosomes, activating under internal or external stimuli, giving the organism the ability to adapt to the environment. Annotating transposable elements in genomic data is currently considered a crucial task to understand key aspects of organisms such as phenotype variability, species evolution, and genome size, among others. Because of the way they replicate, LTR retrotransposons are the most common transposable elements in plants, accounting in some cases for up to 80% of all DNA information. To annotate these elements, a reference library is usually created, a curation process is performed, eliminating TE fragments and false positives and then annotated in the genome using the homology method. However, the curation process can take weeks, requires extensive manual work and the execution of multiple time-consuming bioinformatics software. Here, we propose a machine learning-based approach to perform this process automatically on plant genomes, obtaining up to 91.18% F1-score. This approach was tested with four plant species, obtaining up to 93.6% F1-score (Oryza granulata) in only 22.61 s, where bioinformatics methods took approximately 6 h. This acceleration demonstrates that the ML-based approach is efficient and could be used in massive sequencing projects.

MCHelper automatically curates transposable element libraries across eukaryotic species

“…Recently, a pipeline to de novo annotate TEs included multiple steps to deal with some specific limitations of raw libraries, in particular the presence of redundant and fragmented consensus (Baril et al 2022). Machine learning approaches have also been implemented to automatically curate particular TE orders in plants (Orozco-Arias et al 2021, 2022). Thus, previous attempts at automation were partial and still required substantial time investment by the researcher.…”

“…M. Storer et al, 2021, 2022). Additionally, some specific tools to aid in the manual curation have also been produced (Baril et al, 2022; Goubert et al, 2022; Orozco-Arias, et al, 2021; 2022). Still, manual curation of TE libraries is time-consuming and requires acquiring knowledge on the biology of TEs present in the genomes of interest.…”

Detection and classification of long terminal repeat sequences in plant LTR-retrotransposons and their analysis using explainable machine learning

“…Genomic sequence analysis can thus provide important additional information about the composition, classification, and function of LTRs in LTR retrotransposons and even in their evolutionarily contrasting element subtypes (superfamilies and families, see Wicker et al (2007)). This approach has shown some success when applied to full-length LTR retrotransposon sequences in plants (Arango-López et al 2017), including machine learning approaches (Orozco-Arias et al 2022), but has not been applied specifically to LTRs whose structure is more loosely defined than the structure of internal coding regions of the retrotransposons. This imprecise characterization complicates the analysis of plant LTR sequences with traditional methods.…”