File annotation_data.zip has the following structure: Directory Overview The dataset is organized into two primary directories: annotations/: Contains the annotation files for various plant species. input_assemblies/: Contains the genome assembly files for each species. reference_genomes/: Contains standard annotations for each species and parsed tested annotations. Scripts for evaluation are available in https://github.com/kavonrtep/dante_ltr_evaluation Annotations Directory Within the annotations/ directory, there is a subdirectory for each plant species analyzed. Each subdirectory contains compressed GFF3 and BED files generated by the annotation tools DANTE, DANTE_LTR, EDTA, and Inpactor2. The files are named according to the tool and the type of annotation: DANTE.gff3.gz: GFF3 files with annotations from the DANTE tool. DANTE_LTR.gff3.gz: GFF3 files with annotations from the DANTE_LTR tool, focusing on LTR retrotransposons. EDTA.LTR.intact.gff3.gz: GFF3 files from EDTA, specifically annotated LTR retrotransposons considered to be intact. Inpactor2.tab.gz: Original tabular file provided by Inpactor2. Inpactor2.bed.gz: BED files created from tabular file Input Assemblies Directory Directory input_assemblies/ contains the genomic sequences used for annotation, organized by species. Each species subdirectory contains the genomic FASTA files, individually compressed with gzip. The files are named according to their accession numbers.