Feature Metadata CSV-file

What is a feature metadata file?

A feature metadata file contains additional information about each molecule (feature) in your dataset – for example, annotations, categories, or molecular properties.

How to format the feature metadata file before uploading to Omics Studio

Your file must be a CSV file with:

• The first column containing recognized molecule IDs (UniProt, Ensembl, CHEBI, or InChI)
• Optional additional columns for metadata (e.g. name, description, annotation, etc.)

Example table

📄 Feature Metadata (.csv)

ID	Name	Category	Description	...
A0A1B0GTW7	Cytochrome P450 3A4	Enzyme	Involved in drug metabolism	...
H3BV60	Albumin	Protein	Major plasma protein	...
O14717	TP53	Transcription factor	Tumor suppressor	...
CHEBI:15377	Water	Metabolite	Universal solvent	...

Rules

Header (first row)

ID	Name	Category	Description	...

• The header is the very first line of the file.
• Each column must have a name (no empty cells in the header).
• ID must be the first column.

First column

• Must contain recognized molecule IDs (UniProt, Ensembl, CHEBI, or InChI).
• Each row is validated using pattern matching (regex) to detect which ID type is used.
• No empty cells are allowed in this column.
• Only one type of ID is allowed per file.

ID
A0A1B0GTW7
H3BV60
O14717
CHEBI:15377

Additional columns

• Can include any metadata describing the molecules (e.g. name, function, category, annotation, etc.).
• Empty values are allowed in these columns.
• These columns are not validated.

Name	Category	Description
Cytochrome P450 3A4	Enzyme	Involved in drug metabolism
Albumin	Protein	Major plasma protein
TP53	Transcription factor	Tumor suppressor
Water	Metabolite

💡 Good to know

✅ The first column must contain valid molecule IDs (UniProt, Ensembl, CHEBI, or InChI).
✅ The column name for IDs does not matter — detection is automatic.
❌ Empty headers are not allowed.
❌ Empty values in the first column will cause an error.
✅ Empty values are allowed in all other columns.