biomaRt query Error "no upstream flank found"

Question

[shatabdisen40]

Upstream_flank is an attribute in biomart,how can i use my python script to retrieve the upstream_flank with a value of 1000.

I generated the following xml query:

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE Query>
<Query  virtualSchemaName = "zome_mart" formatter = "FASTA"
header = "0" uniqueRows = "1" count = ""
datasetConfigVersion = "0.6" >
    <Dataset name = "phytozome" interface = "default" >
    <Filter name = "organism_id" value = "447"/>
    <Filter name = "upstream_flank" value = "500"/>
    <Attribute name = "gene_name1" />
    <Attribute name = "transcript_name1" />
    <Attribute name = "transcript_exon_intron" />
    <Attribute name = "organism_name" />
    <Attribute name = "chr_name1" />
    <Attribute name = "gene_chrom_start" />
    <Attribute name = "gene_chrom_end" />
    <Attribute name = "gene_chrom_strand" />
    <Attribute name = "transcript_id" />
    <Attribute name = "transcript_chrom_start" />
    <Attribute name = "transcript_chrom_end" />
    <Attribute name = "peptide_name" />
    <Attribute name = "exon_chrom_start" />

</Dataset>

</query>

but this query is not working and its throwing biomaRt query Error "no upstream flank found" .

I know the issue can be solved by using Biomart package in R but I have written all my code in python and hence want to resolve the issue in python.

Kindly assist me in solving this error.

[shatabdisen40]

Thank you for the clarification, I will try to look into this in detail.

Answer 1

[Mike Smith]

I think this is a problem with the current version of Ensembl BioMart, and I would recommending testing your code against the most recent archive version. There have been similar reports when using biomaRt (e.g. https://support.bioconductor.org/p/120429/), and currenty I don't think the flanking sequence retrieval is working. AFAIK the Ensembl team are looking into it.

---

Actually, on closer look, I'm not sure you can combine the transcript_exon_intron sequence attribute with the upstream_flank filter. I don't know the Mart well enough to be sure, but for human genes it would need to be something like gene_flank or coding_gene_flank. transcript_exon_intron is limited to Exon/intron regions and the flanking filter is meaningless for it.