getGenome fails to download an available genome from RefSeq

[ybukhman]

I got an error when trying to download the elk genome, taxid 1574408, from RefSeq.

R command:
getGenome(db = "refseq", organism = 1574408, reference = TRUE, path = "refseq_fasta_files")

Error:

-> Starting genome retrieval of '1574408' from refseq ...


----------> No reference genome or representative genome was found for '1574408'. Thus, download for this species has been omitted. Have you tried to specify getGenome(db = 'refseq', organism = '1574408' , reference = FALSE) ? Alternatively, you can retrieve genome assemblies using the NCBI accession ID or NCBI Taxonomy ID. See '?'is.genome.available' for examples.
Error in assembly$ftp_path : $ operator is invalid for atomic vectors

However, a reference genome is, in fact available:

> is.genome.available(db = "refseq", organism = 1574408, details = TRUE) %>% select(refseq_category)
    refseq_category
             <char>
1: reference genome

The elk genome assembly page is here: https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_019320065.1/

I'm trying to download a few dozen genomes from an R script. The error in getGenome kills the process.

Thanks!

Yury

[HajkD]

Hi @ybukhman

Thank you for your question.

I had a look at this and found the following:

dplyr::glimpse(biomartr::is.genome.available(db = "refseq", organism = "Cervus canadensis", detail = TRUE))

Rows: 1
Columns: 38
$ assembly_accession        <chr> "GCF_019320065.1"
$ bioproject                <chr> "PRJNA758978"
$ biosample                 <chr> "SAMN15814149"
$ wgs_master                <chr> "JACTMD000000000.1"
$ refseq_category           <chr> "reference genome"
$ taxid                     <int> 1574408
$ species_taxid             <int> 1574408
$ organism_name             <chr> "Cervus canadensis"
$ infraspecific_name        <chr> "na"
$ isolate                   <chr> "Bull"
$ version_status            <chr> NA
$ assembly_level            <chr> NA
$ release_type              <chr> NA
$ genome_rep                <chr> NA
$ seq_rel_date              <IDate> NA
$ asm_name                  <chr> NA
$ submitter                 <chr> NA
$ gbrs_paired_asm           <chr> NA
$ paired_asm_comp           <chr> NA
$ ftp_path                  <chr> NA
$ excluded_from_refseq      <chr> NA
$ relation_to_type_material <chr> NA
$ asm_not_live_date         <chr> NA
$ assembly_type             <chr> NA
$ group                     <chr> NA
$ genome_size               <int> NA
$ genome_size_ungapped      <int> NA
$ gc_percent                <dbl> NA
$ replicon_count            <int> NA
$ scaffold_count            <int> NA
$ contig_count              <int> NA
$ annotation_provider       <chr> NA
$ annotation_name           <chr> NA
$ annotation_date           <lgl> NA
$ total_gene_count          <int> NA
$ protein_coding_gene_count <int> NA
$ non_coding_gene_count     <int> NA
$ pubmed_id                 <chr> NA

Since the information about the FTP path is not returned by the NCBI API ($ ftp_path <chr> NA), this is on the NCBI side and this information would need to be added from their end to enable automated downstream access.

I will now add a capture which tests if the ftp path is empty and then skips the download.

I hope this helps!