---
title: Interpreting cytometry outputs
description: >-
  FlowSOM cluster assignments, segmentation masks, and spatial interaction
  scores.
tags:
  - cytometry
  - outputs
---

> **Retired tools**

> `flow_gate`, `flow_cluster` and `flow_differential` are retired and their outputs are no longer produced. For FCS clustering use `flowsom`. There is no hosted tool for automated gating or differential abundance; ask Ivy to run those in Python in the chat's [code workspace](/docs/sandbox).

Each tool returns discovered clusters, segmentation masks, or spatial interaction scores. Use them to map cell populations and cell neighborhoods, and to export the results for downstream analysis.

> **Documented output fields go beyond the implemented tools**

> The IMC field tables below describe the intended result shape; the registered IMC backends currently return a much smaller set of fields.

## Interpreting cytometry outputs

## FlowSOM output

`flowsom` returns:

- SOM information: the grid dimensions (`xdim`, `ydim`) and training time.
- A metacluster summary: the size and fraction of events in each metacluster, and which SOM nodes belong to it.
- Per-metacluster median marker intensities (MFIs), and per-node MFIs.
- Per-event cluster labels as a CSV.
- An optional star plot PNG of the SOM grid (`generate_plot`).

## IMC Segment output

| Field | Type | Description |
|-------|------|-------------|
| `total_cells` | number | Total cells segmented |
| `method` | string | Segmentation method used |
| `image_dimensions` | object | Image width and height in pixels |
| `pixel_size` | number | Pixel size in microns |
| `cell_mask` | file | Cell mask image (TIFF, each cell labeled with unique integer) |
| `cell_data` | file | CSV with per-cell features |
| `segmentation_stats` | object | Quality metrics |

### Cell data fields

| Field | Type | Description |
|-------|------|-------------|
| `cell_id` | integer | Unique cell identifier (matches mask label) |
| `x_centroid` | number | X coordinate of cell centroid (microns) |
| `y_centroid` | number | Y coordinate of cell centroid (microns) |
| `area` | number | Cell area in square microns |
| `eccentricity` | number | Cell shape eccentricity (0 = circle, 1 = line) |
| `solidity` | number | Cell boundary regularity (0-1, higher = smoother) |
| `marker_*` | number | Mean intensity of each marker within the cell mask |

### Segmentation quality metrics

| Field | Type | Description |
|-------|------|-------------|
| `mean_cell_area` | number | Mean cell area in square microns |
| `median_cell_area` | number | Median cell area |
| `std_cell_area` | number | Standard deviation of cell area |
| `oversegmentation_rate` | number | Estimated rate of split cells (0-1) |
| `undersegmentation_rate` | number | Estimated rate of merged cells (0-1) |
| `cells_per_mm2` | number | Cell density |

## IMC Spatial output

| Field | Type | Description |
|-------|------|-------------|
| `analysis_type` | string | Analysis type performed |
| `total_cells` | number | Total cells analyzed |
| `cell_types_analyzed` | string[] | Cell types included |
| `radius` | number | Neighborhood radius (microns) |
| `permutations` | number | Number of permutations for significance |
| `results` | object | Analysis-specific results |

### Neighborhood analysis fields

| Field | Type | Description |
|-------|------|-------------|
| `interactions` | array | Pairwise cell type interaction results |

### Interaction fields

| Field | Type | Description |
|-------|------|-------------|
| `cell_type_a` | string | First cell type |
| `cell_type_b` | string | Second cell type |
| `observed_frequency` | number | Observed interaction frequency |
| `expected_frequency` | number | Expected frequency under random spatial distribution |
| `interaction_score` | number | Ratio of observed to expected (>1 = attraction, <1 = avoidance) |
| `z_score` | number | Z-score from permutation testing |
| `p_value` | number | Permutation p-value |
| `relationship` | string | `"attraction"`, `"avoidance"`, or `"random"` |

### Interaction score interpretation

| Score | Z-score | Interpretation |
|-------|---------|----------------|
| > 2.0 | > 3.0 | Strong spatial attraction |
| 1.5 - 2.0 | 2.0 - 3.0 | Moderate attraction |
| 0.7 - 1.5 | -2.0 - 2.0 | Random spatial distribution |
| 0.3 - 0.7 | -3.0 - -2.0 | Moderate avoidance |
| < 0.3 | < -3.0 | Strong spatial avoidance |

### Spatial statistics fields

| Field | Type | Description |
|-------|------|-------------|
| `ripleys_k` | object | Ripley's K function values at multiple radii |
| `ripleys_l` | object | Ripley's L function (variance-stabilized K) |
| `nearest_neighbor` | object | Nearest-neighbor distance distributions per cell type |
| `morans_i` | object | Moran's I spatial autocorrelation per marker |

### Community detection fields

| Field | Type | Description |
|-------|------|-------------|
| `n_communities` | number | Number of spatial communities identified |
| `communities` | array | Community descriptions |

### Community fields

| Field | Type | Description |
|-------|------|-------------|
| `community_id` | string | Community identifier |
| `n_cells` | number | Number of cells in this community |
| `dominant_types` | object | Cell type composition (type -> frequency) |
| `spatial_extent` | number | Approximate diameter in microns |
| `description` | string | Suggested tissue microenvironment label |

## Read next