Evidence files

File types, document categories, what is checked, and where files are kept.

Reference

Evidence files are the photographs, chromatograms, documents, genetic fingerprints and lab notebooks that support a record. They are stored in BioStor, the platform's file store, as objects linked to a record by its Process ID, and shown on the record through signed links. This page explains what kinds of file the workbench accepts, how each is classified, what is checked on the way in and where the files live. For how to upload them, see Uploading an evidence package and Attaching a file to one record.

Profiles

BioStor gives every object a profile: the kind of evidence it is, which decides what metadata it carries and how the record displays it. Eleven profiles are reachable from the workbench.

Profile	File types	Where it appears on the record
`specimen_image`	`.jpg`, `.jpeg`	Images
`specimen_documentation` and six other document categories (below)	`.pdf`	Documents
`chromatogram`	`.ab1`, `.abi`, `.scf`	Traces
`genetic_fingerprint`	`.fsa`	Genotype files
`lab_notebook`	`.xlsx`	Spreadsheets

.abi is treated as .ab1. A FASTQ profile exists in BioStor but cannot be reached from the workbench, so supporting reads are not uploaded here.

The seven PDF profiles are the document categories. They describe what a document is, so that a reviewer or a statement can find the right one:

Category	Profile name
Specimen Documentation (the default)	`specimen_documentation`
Chain of Custody	`chain_of_custody`
Collection Statement	`collection_statement`
Permit	`permit`
Taxonomic Identification	`taxonomic_identification`
Lab Worksheet	`lab_worksheet`
Scientific Paper	`scientific_paper`

On the Enrich record form you choose the category from a list; in a document package you put the profile name in the Document Type column. The full profile definitions are in File types.

What is checked on upload

The checks are about the file's name, its link to a record and its metadata, not its contents.

File type. Enrich record accepts only the extensions listed for the chosen profile and rejects anything else before it is sent. Package files are not filtered by type in the workbench; each file is sent to BioStor under the package's profile, and a file BioStor will not take for that profile fails the submission at the biostor_upload stage.
Link to a record. Enrich record adds the Process ID of the open record automatically. Packages take the identifier from metadata.tsv: images and chromatograms accept a Sample ID and/or a Process ID; documents, fingerprints and lab notebooks require a Process ID.
Metadata. Columns in metadata.tsv that the workbench recognises are passed to BioStor with the file. Profile-specific fields on the Enrich form are optional.
Contents. Sequence-type files (.ab1, .scf, .fsa, and FASTA or FASTQ where BioStor receives them) are not opened and checked for valid format on upload. A checksum of each file is recorded when it is stored, but it is not re-verified afterwards. If a chromatogram or fingerprint file is corrupt, you will find out when a view tries to render it, not at upload.

Where files are kept

Files are objects in BioStor, with their metadata, profile and the Process ID they belong to. The record store keeps a count of each kind of evidence per record, which is refreshed after every successful upload and drives the counts on the recordset console. The workbench never links to a file directly: every image, document or trace you open is streamed through a signed proxy link that is tied to your session and that one object. See Maps, images, alignments and traces for the viewers and Downloading records, sequences and files for bulk download.

Access

Good to know: Evidence uploads are not restricted by your access to the record. Any signed-in user can attach a file to any Process ID, whether through a package or Enrich record, and the upload is made with the workbench's own BioStor credentials rather than yours. The platform relies on you to name the right record. Check the Process ID in your metadata.tsv before uploading, and remember that an activity-log entry records who uploaded what.

Good to know

Good to know: The side menu and the upload modal use different names for the same thing: Specimen Documentation is "Document", Chromatograms is "Trace", Genetic Fingerprints is "Genotype Data" and Lab Notebooks is "Spreadsheet".

Good to know: There is no documented file-size limit. Very large files are sent in pieces by the package route; Enrich record sends the file in one request.