Gene ID Conversion with g:Profiler 

🔍 Introduction

When you merge datasets, run downstream analyses, or build visualizations, you almost always need one canonical identifier system (e.g., Ensembl IDs or Entrez IDs). The problem: the same gene may appear as HGNC symbols, Ensembl stable IDs, Entrez Gene IDs, UniProt accessions, microarray probes, and more.

g:Profiler – g:Convert solves this by providing a central, regularly updated service that translates gene identifiers across dozens of namespaces for hundreds of organisms. This task ( OmiX – Gene_ID_conversion ) wraps g:Convert so you can normalize IDs from a table in a simple, reproducible way—no manual lookups, no stale spreadsheets.

The list of supported databases (namespaces) in g:Profiler — useful for filling your target_namespace or numeric_namespace parameters:

• Namespaces list ▶ https://biit.cs.ut.ee/gprofiler/page/namespaces-list

the link for the list of supported organisms in g:Profiler:

• Organisms list ▶ https://biit.cs.ut.ee/gprofiler/api/util/organisms_list

🧰 Prerequisites

• Access to Constellab and a valid Digital Lab environment
• Installed bricks: gws_omix ≥ 0.11.12
• An input table (CSV or TSV), such as: a DEG results file, a one-column list of identifiers, or any table that contains a column with gene IDs.

🧪 What the task does (Steps)

This task performs identifier harmonization on your gene IDs using g:Profiler g:Convert:

1. Reads your table (CSV/TSV) and picks the ID columnIf you leave id_column empty, the first column is used automatically.
2. Queries g:Convert for your organism and target namespace Examples of target namespaces: ENSG, ENTREZGENE, ENSGACC, UNIPROT, etc.
3. Handles “bare numbers” correctly with the Numeric IDs treated as option If your inputs are just numbers (e.g., 2167, 2289, …) and you want them treated as Entrez Gene IDs, set: target_namespace = ENTREZGENE (what you want to obtain) numeric_namespace = ENTREZGENE_ACC (how to interpret the numeric inputs)
4. Writes a clean table with the mapping result: incoming, converted, name, description

📥 Inputs

• File: CSV or TSV containing your gene identifiers Mixed identifiers are fine; g:Profiler handles a lot of cases.

⚙️ Parameters

• organism_name Scientific name or g:Profiler code (e.g., Homo sapiens ).
• id_column (optional) Column to convert. Leave empty to use the first column in your file.
• target_namespace Target identifier system (e.g., ENSG, ENTREZGENE, ENSGACC, UNIPROT, …).
• numeric_namespace (optional; default auto) How to interpret bare numeric inputs. Choose a specific namespace (e.g., ENTREZGENE_ACC) when your input column is purely numbers. Use auto when your inputs are not purely numeric or already have prefixes (e.g., ENSG…).

📤 Outputs

• converted_table (Table) The conversion results; columns: incoming – the exact input string the service received converted – the mapped identifier in your target_namespacename – a convenient alias (often the preferred gene symbol) description – human-readable description if available