Configure compound representations¶

Feature list methods → Feature grouping → Configure compound representations

Controls how each compound row (produced by Compound grouping) exposes its per-raw-file intensity — area, height, normalized area, and normalized height — to downstream modules, exports, and visualizations.

Three modes are available: fall back to the representative member's feature, or synthesize a compound-level feature by summing across all members or only members that carry an Ion Identity.

The module is non-destructive: it does not alter member rows or grouping. Re-run any time to switch modes; the change takes effect immediately for any subsequent intensity read.

Warning

Run Compound grouping first — this module needs a populated compound list to act on.

Parameters¶

Feature lists¶

The feature list(s) whose compound rows should be reconfigured. Each list is processed independently.

Intensity representation¶

Selects how every compound row computes its per-raw-file feature values. One of:

Representative row (default) No compound-level features are stored. When something reads a compound row's feature for raw file X, the module returns the representative member row's feature for X (or null if the representative has no feature in that file). This matches the behavior before the configuration module existed and uses the least memory.
Sum of all members For each raw file, a synthesized compound feature is built whose area, height, normalized area, and normalized height are the sums across every member row's feature in that file. Members with no feature in a given file contribute zero.
Sum of members with ion identity Same as Sum of all members, but only members where hasIonIdentity() is true contribute to the sum. Correlation-only members (those joined by MS1 shape but without a resolved adduct) are excluded.

Tip

Stay on Representative row when you want classical per-feature quantification (e.g. the representative is the M+H and you do not want adduct co-summation to shift values).

Output and side effects¶

For Sum of all members and Sum of members with ion identity, a compound-level feature is materialized per raw file and stored on the compound row. These features back all subsequent intensity reads (table columns, exports, plots).
For Representative row, no compound-level features are created; reads delegate to the representative member.
An applied-method record is appended so the project replays consistently. Re-running the module overwrites the previous mode.

Robin Schmid