Metric: Moran's I

CHANGELOG.md

@@ -8,13 +8,15 @@
 
 * Added `spaTrack` method (PR #4).
 * Added `Spearman's correlation` metric (PR #5).
+* Added `Moran's I` metric (PR #6).
 
 ## MAJOR CHANGES
 
 * Updated `api` files and set the data processor (PR #1).
 
 ## MINOR CHANGES
 
+* Added package versions to the moran's I config (PR #6).
 
 ## BUGFIXES
 

src/metrics/morans_i/config.vsh.yaml

@@ -0,0 +1,84 @@
+__merge__: ../../api/comp_metric.yaml
+
+name: morans_i
+
+info:
+  metrics:
+    - name: morans_i
+      label: Moran's I
+      summary: "Measures the spatial autocorrelation of the inferred pseudotime across the spatial coordinates of the cells/spots."
+      description: |
+        Computes Moran's I statistic for the inferred pseudotime, using a k-nearest-neighbour
+        graph built on the spatial coordinates (`obsm['X_spatial']`) of the cells/spots.
+
+        Moran's I quantifies how similar the pseudotime values of spatially neighbouring
+        cells/spots are. Values close to 1 indicate a spatially smooth trajectory in which
+        neighbouring cells receive similar pseudotime values, values around 0 indicate a
+        spatially random assignment, and negative values indicate that neighbouring cells
+        receive dissimilar pseudotime values.
+
+        Note that this metric evaluates the spatial coherence of the prediction only and does
+        not compare it against the ground-truth pseudotime; a spatially smooth but incorrect
+        ordering can still score highly.
+
+      references:
+        doi:
+          - 10.2307/2332142
+        bibtex: |
+          @article{Moran_1950,
+           author  = {Moran, P. A. P.},
+           title   = {Notes on Continuous Stochastic Phenomena},
+           journal = {Biometrika},
+           volume  = {37},
+           number  = {1/2},
+           pages   = {17--23},
+           year    = {1950},
+           doi     = {10.2307/2332142}
+           }
+
+      links:
+
+        documentation: https://scanpy.readthedocs.io/en/stable/api/generated/scanpy.metrics.morans_i.html
+
+        repository: https://github.com/scverse/scanpy
+      min: -1
+      max: 1
+      maximize: true
+
+arguments:
+  - name: "--n_neighbors"
+    type: "integer"
+    default: 6
+    description: Number of spatial neighbours used to build the k-NN graph.
+
+resources:
+  - type: python_script
+    path: script.py
+
+
+engines:
+  - type: docker
+    image: python:3.11-slim
+    setup:
+      - type: apt
+        packages:
+          - procps
+          - git
+      - type: python
+        packages:
+          - anndata~=0.10.9
+          - scanpy~=1.10.4
+          - numpy~=2.4.6
+          - pandas~=3.0.5
+          - scipy~=1.17.1
+          - pyyaml~=6.0.3
+          - requests~=2.34.2
+          - jsonschema~=4.26.0
+        github:
+          - "openproblems-bio/core#subdirectory=packages/python/openproblems"
+
+runners:
+  - type: executable
+  - type: nextflow
+    directives:
+      label: [midtime,midmem,midcpu]

src/metrics/morans_i/script.py

@@ -0,0 +1,78 @@
+import anndata as ad
+import numpy as np
+import pandas as pd
+import scanpy as sc
+from scanpy.metrics import morans_i
+
+## VIASH START
+# Note: this section is auto-generated by viash at runtime. To edit it, make changes
+# in config.vsh.yaml and then run `viash config inject config.vsh.yaml`.
+par = {
+    'input_solution': 'resources_test/task_spatial_trajectory_inference/dlpfc_151673/solution.h5ad',
+    'input_prediction': 'resources_test/task_spatial_trajectory_inference/dlpfc_151673/prediction.h5ad',
+    'output': 'output.h5ad',
+    'n_neighbors': 6,
+}
+meta = {
+  'name': 'morans_i'
+}
+## VIASH END
+
+
+def calc_morans_i(adata, pt_col, coords_key, n_neighbors=6):
+    """Spatial autocorrelation of pseudotime via Moran's I."""
+    try:
+        sc.pp.neighbors(adata, use_rep=coords_key, n_neighbors=n_neighbors, key_added="spatial_neighbors")
+        pt = pd.to_numeric(adata.obs[pt_col], errors="coerce").values
+
+        return float(morans_i(adata.obsp["spatial_neighbors_connectivities"], pt))
+    except Exception as e:
+        print(f"Moran's I skipped for {pt_col}: {e}")
+        return np.nan
+
+# read input data
+print('Reading input files', flush=True)
+input_solution = ad.read_h5ad(par['input_solution'])
+input_prediction = ad.read_h5ad(par['input_prediction'])
+
+assert (input_prediction.obs_names == input_solution.obs_names).all(), "obs_names not the same in prediction and solution inputs"
+
+# inferred pseudotime and spatial coordinates
+INFERRED_COL = "pseudotime_inferred"
+COORDS_KEY = "X_spatial"
+
+# spatial coordinates in the solution, the pseudotime in the prediction
+adata = ad.AnnData(
+    obs=pd.DataFrame(
+        {INFERRED_COL: input_prediction.obs[INFERRED_COL].values},
+        index=input_solution.obs_names,
+    ),
+    obsm={COORDS_KEY: np.asarray(input_solution.obsm[COORDS_KEY])},
+)
+
+# generate results
+print('Compute metrics', flush=True)
+# metric_ids and metric_values can have length > 1
+# but should be of equal length
+
+score = calc_morans_i(adata, INFERRED_COL, COORDS_KEY, n_neighbors=par['n_neighbors'])
+
+uns_metric_ids = [ 'morans_i' ]
+uns_metric_values = [ score ]
+
+# Write output data to file
+print("Write output AnnData to file...", flush=True)
+
+output = ad.AnnData(
+    obs=pd.DataFrame(index=pd.Index(np.array([], dtype=str))),
+    var=pd.DataFrame(index=pd.Index(np.array([], dtype=str))),
+    uns={
+        'dataset_id': input_solution.uns.get('dataset_id', 'unknown'),
+        'normalization_id': input_solution.uns.get('normalization_id', 'unknown'),
+        'method_id': input_prediction.uns.get('method_id', 'unknown'),
+        'metric_ids': uns_metric_ids,
+        'metric_values': uns_metric_values,
+    }
+)
+
+output.write_h5ad(par['output'], compression='gzip')