Repository navigation
Custom Nodes
A custom node is a step you write yourself: a Nextflow process (or a named workflow) that becomes a reusable node in the menu. Use one for a tool nf-core doesn't have, a script of your own, or a changed version of an existing node.
- From scratch: + (Add node) → Custom process.
-
From any node: open its Code tab and click Convert to custom node. The node is
replaced in place by a custom node with a copy of its code, settings such as
ext.argsincluded; connections to ports that still exist are kept.
In the editor, give the node a Name, an Icon and a Description, and write or paste the Nextflow process code. N-WAVE reads the process and fills in:
| Section | What it is |
|---|---|
| File Inputs | The process's path inputs, which become input ports. At least one is needed. |
| Settings | Its val inputs, shown in the node's panel: Text, Integer, Float, Boolean, Selection (a comma-separated list) or Groovy expression, each with a default. |
| Outputs | Its outputs, which become output ports. At least one is needed. |
Check the labels and file patterns, then Save Node. The node appears in the menu under Custom.
In the node menu, use the Edit or Delete button next to a custom node (or Edit custom node on its Code tab). Changes apply to the copies already placed in workflows. Deleting a custom node also removes it from every saved workflow.
On an installation, custom nodes are stored by the server (in results/.nwave/custom-nodes
with the Docker setup), so everyone using it sees them. In the
browser demo, they are kept in this browser.
Put custom nodes into a node pack to share them as a file or through the community index: Node packs.
- Give the process a
containerwith the tools it needs, so it runs the same everywhere. - Write exactly the files the
output:block names, or the run fails with Missing output files. - For channel logic between nodes (join, mix, branch), use a Channel Operator instead (Node reference).
- See also Nextflow's process documentation.