go-v.orgdoi

Cmelo_NLRome_AMES20443_ME01355_combined_v1.gff3

Javier Belinchon-Moreno, Nathalie Boissot, Patricia Faivre Rampant · Recherche Data Gouv France · 2026

  • DOI 10.57745/bpiipe
  • Publisher OpenAlex
  • Type dataset
  • Issued 2026

Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.

Recherche Data Gouv France · OpenAlex · dataset · 2026

Abstract

This dataset contains genome assembly and annotation data of the complete set of Nucleotide-binding-site-Leucine-rich-Repeat (NLR) regions (NLRome) from 143 melon accessions. This data was generated during the PhD project of Javier BELINCHON-MORENO at INRAE-EPGV and INRAE-GAFL. NLR genes encode the most diverse family of plant resistance proteins, playing a central role in the so-called effector triggered immunity of plants by recognizing specific pathogen effectors. NLR genes are often arranged in clusters that include a large number of repetitive elements.

This fact, together with the repetitive intra-structure of NLR genes, makes them prone to major evolutionary structural changes like duplication or transposition. For these reasons, NLR clusters commonly present a high level of presence/absence (PAV) polymorphisms. NLR clusters of Cucumis melo were obtained using the long-read targeted sequencing method "adaptive sampling" (NAS) from Oxford Nanopore Technologies. Nanopore adaptive sampling (NAS) is a cost and labor-effective long-reads targeted sequencing method that only requires a reference sequence to target as input. We identified 15 NLR clusters using NLGenomeSweeper v.

1.2.1 in the recently assembled whole-genome sequence (WGS) of ANSO-77 (NCBI BioProject PRJNA662717), which served as the reference for the NAS approach. Targeted assemblies were first performed using SMARTdenovo v. 2018.2.19. When assembly quality or contiguity requirements were not achieved for a given NLR region of a given accession, we attempted its assembly with Canu v. 2.2, Flye v. 2.9.1, Shasta v. 0.11.1, or NextDenovo v. 2.5.0. For those NLR regions showing a certain level of heterozygosity, we generated pseudo-haplotypes by correcting NAS reads with HERRO followed by assembly with Hifiasm v. 0.19.9. Later, using NLR annotation tools like NLRAnnotator v.

2.1b, and analysing the NLR set of previously released Cucumis melo assemblies, we discovered that six isolated NLR genes were missed in the reference used for adaptive sampling. To include these genes within the assembled NLRomes, we used the rejected NAS reads along with previously available WGS short reads for each accession, performing reference-guided assemblies with the software Geneious Prime v. 2022.1.1 and using the ANSO-77 assembly as reference. Each FASTA file corresponds to the assembly of the complete set of NLR regions of one accession.

In the FASTA file, sequence headers present the following structure: accesionnumber#haplotype#contigname_clusterposition, where accesionnumber is the code of the accession; haplotype is the haplotype strand -1 or 2- (normally just 1 appears as sequenced accessions are diploid and mainly homozygous; 2 appears when two or more pseudo-haplotypes were assembled for a given NLR region); contigname is the name of the assembled contig (contigs starting by "utg", "tig", "ctg", "contig", number, and "guided" were assembled with SMARTdenovo, Canu, NextDenovo, Flye, Shasta, and Geneious, respectively); and clusterposition is the genomic position of the assembled NLR region within the reference genome, ANSO-77 (as established in the file "clusters_positions_ANSO-77.txt").

Structural gene annotations were performed with Helixer v. 0.3.3, and this annotation was complemented with an expert manual annotation of Vat genes between the M5 and M4 markers on cluster Chr05.3 (Chovelon et al., 2021). Final annotations (GFF files) are a combination of manual annotation between M5 and M4 markers, and Helixer annotation for the rest of NLR zones. We used four different tools to identify and classify NLRs: NLGenomeSweeper, NLR-Annotator v. 2.1b, RGAugury v. 2.0 and Resistify v. 0.5.1. We combined the four annotations to generate a consensus. We lifted the annotations from NLGenomeSweeper and NLR-Annotator on the structural gene annotation from Helixer.

We discarded nucleotide-based annotations not supported by any protein-based annotator, though these were usually very few. We selected the NLR domain composition based on majority vote method. When no NLR classification was predominant, we prioritized Resistify, followed by RGAugury, NLRAnnotator, and NLGenomeSweeper.

FILE DESCRIPTION - Positions of targeted-sequenced NLR regions in ANSO-77 genome assembly : file clusters_positions_ANSO-77.txt - Targeted assemblies of NLR regions : files NLRome_assemblies/Cmelo_NLRome_NAME_MECODE_VERSION.fasta.gz - Structural annotation of targeted assemblies of NLR regions : files NLRome_structural_annotations/Cmelo_NLRome_NAME_MECODE_combined_VERSION.gff3 - Selected NLR annotation of targeted assemblies of NLR regions : files NLRome_selected_NLR_annotations/final_selected_NLRannotations_NAME_MECODE_VERSION.gff3 - Discarded NLR annotation of targeted assemblies of NLR regions (only supported by NLAnnotator or NLGenomeSweeper) : files fNLRome_discarded_NLR_annotations/inal_discarded_NLRannotations_NAME_MECODE_VERSION.bed Where NAME is the common accession name, MECODE is the ME code of the accession, and VERSION corresponds to the file version. - Updated annotations for Fom-1, Fom-2 and the candidate gene 1 for flaccid necrosis produced by Zucchini yellow mosaic virus (ZYMV) : files updated_annotations_genes/Fom-1_updated_annotation.txt updated_annotations_genes/Fom-2_updated_annotation.txt and updated_annotations_genes/Fn_candidategene1_updated_annotation.txt METADATA - Sample-related metadata is available in the attached file: metadata_samples.txt - Sequencing-related metadata is available in the attached file: metadata_sequencing.txt - Assembly-related metadata is available in the attached file: metadata_assembly.txt - Detailed explanation about the sequencing and assembly of each accession is available in the attached file : details_sequencing_assembly.txt -Updated annotations for Fom-1, Fom-2 and the candidate gene 1 for flaccid necrosis produced by Zucchini yellow mosaic virus (ZYMV): Multi-alignments of these alleles revealed erroneous Helixer intron-exon annotations in 12 Fom-2 alleles, all Fom-1 alleles, and 6 alleles for the ZYMV-flaccid necrosis candidate gene 1.

Therefore, we used Liftoff v1.6.3 with default parameters to retrieve the correct annotations. We used as reference the genes ME00791#1#utg22_Chr11.1_000033 (gene Fom-2 from Isabelle); Fom-1 gene (splice variant A) from MR-1, retrieved from GenBank accession JX295632.2; and ME00509#1#utg45_Chr05.3_000021 (candidate 1 for ZYMV-flaccid necrosis from ANSO-77). We confirmed the accuracy of new annotations by protein multi-alignments. AVAILABILITY OF SCRIPTS The repository containing all the scripts used for NLRome-targeted genome assemblies, NLRome annotation and data analysis is indicated in file scripts_availability.txt

Open the resolved locator · 10.57745/bpiipe

Metadata licence: OpenAlex metadata (CC0)

Registry
OpenAlex DOI index + API
Fetched
2026-11-19 20:48 UTC
Cache state
not cached yet (bulk-index+openalex)
Locator
https://entrepot.recherche.data.gouv.fr/file.xhtml
Metadata licence
OpenAlex metadata (CC0)
Scope
Metadata as returned by the resolver; abstracts are never generated here.