Downloading records, sequences and files

The Data Download menu on a recordset console (and on the Search page) gets data out of the platform in three ways: a specimen records download that produces a spreadsheet or JSON file, a sequence download that produces a FASTA file, and bulk downloads that package every file of one type in the recordset as a ZIP. The first two run as analysis jobs; the third is built while you wait.

How it works

Specimen and sequence downloads are analysis jobs. You choose the fields and options, the job is queued with the records selected in the records table, and the result arrives in the Analysis panel like any other job, as a package you download from there. This means a large export does not tie up your browser, and the export is kept with your other results. See Tracking jobs and reading reports.

Bulk file downloads are different. They are assembled on the spot from the file store and sent straight to your browser, and they always cover the whole recordset.

Good to know: Every row in the records table is selected when the console opens, so a download job covers the whole recordset unless you narrow the selection first. Bulk file downloads ignore the selection altogether.

Specimen records

1. Select the records you want (or leave all selected).
2. Open Data Download and choose Specimen Records.
3. Tick the fields to include from each group: specimen, taxonomy, collection, sequence, record organization, forensic extension, forensic extension advanced, individualization extension and feature rollup. The defaults are processid, sampleid, museumid, identification and country/ocean.
4. Choose the Format and submit.

The field groups are described in What a record holds and listed in full in the field dictionary.

Sequences

1. Select the records.
2. Open Data Download and choose Sequences.
3. Under Align Sequences, choose None (use sequences as provided) or MAFFT. With MAFFT, the job aligns the sequences before writing the file.
4. Choose the fields to include in each FASTA header, from the same groups as above. The defaults are record_id, identification, country/ocean and nuc_basecount.
5. Set the Sequence Length Filter and, if wanted, Exclude Records with Stop Codons and Exclude Records Flagged as Misidentifications or Errors.
6. Submit. The only output format is FASTA.

Use this job, rather than the alignment browser, when you need an aligned FASTA file; the alignment runs inside the job.

All files of one type

The remaining six items under Data Download package every file of that type in the recordset into one ZIP:

Choose an item and a Packaging notice appears while the ZIP is built. When it is ready, the browser saves it as <type>_<token>.zip. Alongside the files, the ZIP contains metadata.tsv, one row per file with the process ID it belongs to and the file's stored metadata, so each file can be traced back to its record. Files that share a name are suffixed _2, _3 and so on.

Limit: Bulk downloads cover the whole recordset, not the selection, and are refused for recordsets of more than 8,000 records. A file that cannot be fetched from the file store is left out of the ZIP without a warning; compare the number of files with the count on the console if completeness matters. The ZIP is built while you wait, so a large recordset takes a while and the notice stays open until it is done.

Other ways to get data out

• Each file preview in the record detail has its own Download button for that one file. See Maps, images, alignments and traces.
• The compliance report on the console can download its list of issues as a TSV file. See The recordset console.
• The Primers page can export selected primers as a FASTA file.
• Analysis results are downloaded from the Analysis panel as a result package.

Related

• The recordset console
• Running an analysis
• Tracking jobs and reading reports
• Field dictionary