Sample View Parameters¶

Sample view provides dynamic parameters for expressions in its child specifications and data sources. These parameters react to the current layout and sample-collection state.

Sample height¶

The height parameter contains the height in pixels of one sample row. It is available in the repeated child spec. The height depends on the number of samples and the height of the sample view. It also changes when the end user peeks samples.

To adapt the maximum size of "point" marks to the sample height, use a dynamic scale range for the size channel. This example uses expressions and height to adjust the point size:

"encoding": {
  "size": {
    "field": "VAF",
    "type": "quantitative",
    "scale": {
      "domain": [0, 1],
      "range": [
        { "expr": "0" },
        { "expr": "pow(clamp(height * 0.65, 2, 18), 2)" }
      ]
    }
  },
  ...
}

Multiplying height by 0.65 leaves padding above and below the points. The clamp function limits the diameter to 2–18 pixels. Because the size channel encodes area rather than diameter, the expression squares the value. The PARPiCL example uses this technique.

Visible samples¶

visibleSamples contains the identifiers of samples currently visible in the sample hierarchy. It reflects filtering and grouping state, but not transient viewport closeup state.

{
  "expr": "visibleSamples"
}

visibleSampleMetadata provides metadata values for the currently visible samples. Use bracket access for full metadata paths or dot access for simple hierarchical names:

{
  "expr": "visibleSampleMetadata['Clinical/patientId']"
}

{
  "expr": "visibleSampleMetadata.Clinical.patientId"
}

These parameters are useful with URL templates and multiple files to load only the per-sample, per-patient, or per-cohort-partition files relevant to the current sample set.

Aggregate tracks also receive a sampleCount parameter. See Aggregating Samples.