auroraomics 0.1.0.dev2__tar.gz → 0.1.0.dev4__tar.gz

{auroraomics-0.1.0.dev2/src/auroraomics.egg-info → auroraomics-0.1.0.dev4}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: auroraomics
-Version: 0.1.0.dev2
-Summary: Virtual spatial transcriptomics from H&E histology: patch QC, tile packing and the .h5ad result contract.
+Version: 0.1.0.dev4
+Summary: Prepare H&E slides on your own machine, submit them to Aurora, and read the spatial gene expression it predicts.
 Author: Kalin Nonchev
 License-Expression: PolyForm-Noncommercial-1.0.0
 Keywords: histopathology,spatial-transcriptomics,h5ad,anndata,whole-slide-image
@@ -23,16 +23,22 @@ Requires-Dist: pydantic>=2
 Requires-Dist: anndata>=0.10
 Requires-Dist: tifffile>=2023.7.10
 Requires-Dist: imagecodecs>=2023.3.16
+Requires-Dist: scipy>1.8
 Provides-Extra: embed
 Requires-Dist: torch>=2.0; extra == "embed"
 Requires-Dist: timm>=1.0; extra == "embed"
 Requires-Dist: transformers>=4.40; extra == "embed"
 Requires-Dist: huggingface-hub>=0.23; extra == "embed"
+Provides-Extra: spatialdata
+Requires-Dist: spatialdata>=0.2; extra == "spatialdata"
+Requires-Dist: setuptools<81; python_version < "3.12" and extra == "spatialdata"
 Provides-Extra: mcp
 Requires-Dist: mcp<3,>=2; extra == "mcp"
 Provides-Extra: dev
 Requires-Dist: auroraomics[mcp]; extra == "dev"
+Requires-Dist: auroraomics[spatialdata]; extra == "dev"
 Requires-Dist: pytest>=7; extra == "dev"
+Requires-Dist: packaging>=22; extra == "dev"
 Requires-Dist: pytest-cov>=5; extra == "dev"
 Requires-Dist: setuptools>=77; extra == "dev"
 Requires-Dist: wheel; extra == "dev"
@@ -43,10 +49,14 @@ Dynamic: license-file
 
 # auroraomics
 
-Virtual spatial transcriptomics from H&E histology.
+Virtual spatial transcriptomics from H&E images.
+
+This package is the client of the **Aurora API**, the hosted service that runs
+the prediction: from Python, from a shell with the `auroraomics` command, or
+from an agent through its MCP server.
 
 This package holds the pieces of that pipeline that are pure Python, so the
-same code runs on your laptop, in a GPU container and on the service:
+same code runs on your machine and on the service:
 
 - **`auroraomics.qc`** — the three patch-quality predicates (foreground, blur,
   stained tissue) applied to every candidate tile before a model sees it, and
@@ -58,16 +68,19 @@ same code runs on your laptop, in a GPU container and on the service:
   batch rather than the whole matrix.
 - **`auroraomics.postprocess`** — apply a model card's `postprocess` entries
   to a result file: a registry of entries, one runner, `h5py` alone, streaming
-  one chunk at a time. Ships the unmeasured-gene mask, which states per gene
-  how many training datasets measured it and blanks the ones none did.
+  one chunk at a time. Ships the unmeasured-gene mask, which blanks the genes
+  the model never measured.
 - **`auroraomics.subsample`** — pick the densest contiguous square of spots
-  when a slide yields more tiles than a run is allowed to spend.
-- **`auroraomics.genes`** — the model family's gene symbols mapped to Ensembl
-  stable gene ids, which is what every result's `var` index is keyed by. Ships
-  as one committed table, resolved in the same order the service resolves it.
+  when a slide yields more tiles than one submission takes.
+- **`auroraomics.genes`** — Ensembl stable gene ids, which is what every
+  result's `var` index is keyed by, and the shape a gene the service resolved
+  comes back in.
 - **`auroraomics.contracts`** — the shared contract values (container layout,
   input-kind caps, result layout) as data, so nothing here re-types a number
   the service also reads.
+- **`auroraomics.export`** — write a result as a SpatialData Zarr store
+  *(extra)* or as a folder Seurat loads, with the same matrix, coordinates and
+  gene names as the `.h5ad`.
 - **`auroraomics.embed`** *(extra)* — run a pinned image encoder over your
   tiles locally and write the embeddings container, so a prediction can be made
   from numbers instead of pixels.
@@ -88,16 +101,19 @@ pip install auroraomics
 
 That is the whole documented path: open a slide, judge and pack its tiles,
 submit, and read the result. One extra adds local embedding extraction
-(`embed`), because a tensor runtime is gigabytes and specific to the machine it
-was built for. No extra brings the model: predicting gene expression runs on
-the service, not here.
+(`embed`), and one writes a result as a SpatialData store (`spatialdata`). No
+extra brings the model: predicting gene expression runs on the service, not
+here.
 
 ## Pack tiles, then look at the report
 
 ```python
 import auroraomics as ao
 
-report = ao.pack_tiles(tiles, "sample.zip", mpp=0.499, thumbnail=thumb)
+with ao.Client() as client:
+    thresholds = client.qc_thresholds()   # the service's quality floors, no key needed
+
+report = ao.pack_tiles(tiles, "sample.zip", mpp=0.499, thresholds=thresholds, thumbnail=thumb)
 print(report.written, "tiles kept,", report.rejected, "dropped")
 print(report.rejected_by_reason)          # {'foreground_ratio': 12, ...}
 ```
@@ -126,7 +142,8 @@ members do not match the container contract.
 ao.write_result(
     "result.h5ad",
     obs=obs,                    # per-spot columns, as plain arrays
-    var=var,                    # per-gene columns, indexed by gene id
+    var=var,                    # per-gene columns
+    var_index=gene_ids,         # the Ensembl gene id of each column
     spatial=coords,             # (n_spots, 2) array -> obsm["spatial"]
     x=batches,                  # an array, or an iterable of row batches
     uns={"model": {"id": "..."}},
@@ -149,71 +166,69 @@ pip install "auroraomics[embed]"
 import auroraomics as ao
 from auroraomics.embed import available_encoders, describe_encoders
 
-print(available_encoders())        # ('deepspot-h', 'dinov2-b14', 'midnight')
-
 with ao.Client() as client:
     # Both are the service's, so the file records the encoder and the
-    # thresholds the model it is submitted to was served with.
-    encoder = ao.resolve_encoder(client.encoders())
+    # thresholds the model it is submitted to was served with. There is no
+    # list of encoders inside this package: every name comes from the registry
+    # the service publishes, which is why these calls take one.
+    registry = client.encoders()
     thresholds = client.qc_thresholds()
+    # The tile size the model's card states, for the model the file is for.
+    patch_um = client.model_card(model_id)["input_spec"]["patch_um"]
+
+print(available_encoders(registry))   # the names this package will run today
+encoder = ao.resolve_encoder(registry)
 
 report = ao.embed_tiles(
-    slide, "sample.npz", mpp=0.499, patch_um=55, encoder=encoder, thresholds=thresholds
+    slide, "sample.npz", mpp=0.499, patch_um=patch_um, encoder=encoder, thresholds=thresholds
 )
 print(report.rows, "rows of", report.dim, "numbers from", report.encoder.name)
 ```
 
 `slide` is anything with a `(height, width, 3)` RGB `uint8` shape that can be
 sliced — an array, or a memory-mapped or lazily-read one, so a slide larger than
-memory works: only one crop exists at a time. Crops are taken at `patch_um`
-micrometres using the `mpp` you give, quality-checked with the same three
-predicates as `pack_tiles`, and pooled over the encoder's patch tokens.
-
-Every encoder is pinned to a repository **and** a revision sha, with its licence
-and access gate recorded beside it. `describe_encoders()` returns the whole
-registry, including the encoders this package refuses to run — a refusal names
-the licence or the approval you would have to accept, and which encoders remain.
-
-The written file carries one extra vector, `calibration`: the embedding of a
-fixed image shipped inside this package, through the same encoder, revision and
-preprocessing as your rows. Comparing that one vector against a known reference
-tells a reader whether the file was produced by the model it claims, before they
-look at a single row.
-
-The weights are downloaded from their publisher on first use, into the ordinary
-model cache. Pass `allow_download=False` to guarantee no request is made: for
-the duration of that load, name resolution and internet sockets are refused in
-this process, so the guarantee does not rest on a model library reading an
-environment variable it may already have read.
-
-An encoder runs at one input size: the size the served model's embeddings were
-computed at, which the registry pins and which is not always what the encoder's
-own repository declares. The DINOv2 releases carry a 518-px position grid; the
-model was trained on 224-px crops fed to that grid through an interpolated
-position embedding, so that is what this package does too. `resize_px` is
-checked against the pin rather than obeyed, because a vector computed at another
-size is a different measurement that nobody else's numbers can be compared with.
+memory works: only one crop exists at a time. Crops are taken at the tile size
+the model's card states, using the `mpp` you give, and quality-checked with the
+same three predicates as `pack_tiles`.
+
+Every encoder is pinned to an exact revision.
+`describe_encoders(registry)` returns the whole registry, including the
+encoders this package refuses to run: a refusal says why, and which encoders
+remain.
+
+The written file carries one extra embedding, `calibration`: computed from a
+fixed image shipped inside this package, through the same encoder and revision
+as your rows. Comparing that one embedding against a known
+reference tells a reader whether the file was produced by the model it claims,
+before they look at a single row.
+
+The weights are downloaded on first use. Pass `allow_download=False` to
+guarantee no request is made: for the duration of that load, name resolution
+and internet sockets are refused in this process.
+
+An encoder runs at the input size the registry pins for it. `resize_px` is
+checked against that pin rather than obeyed, because an embedding computed at
+another size cannot be compared with the served model's.
+
 ## Where a prediction runs
 
-Predicting gene expression runs on the service, and `predict_local` refuses —
-by design, and the design is the product rather than a limitation.
+Predicting gene expression runs on the service, and `predict_local` refuses.
 
-This is split inference at the encoder. An open-weight pathology foundation
-model turns your tiles into patch embeddings **on your machine**, so your slides
-never leave it; our gene decoder turns embeddings into expression **on ours**, so
-its weights never leave it. What crosses between us is a vector from a model
-neither side owns.
+**DeepSpot-H**, the foundation model for H&E images, turns each tile of your
+slide into one embedding. It can run on your machine or on ours. **DeepSpot-M**
+turns those embeddings into spatial gene expression. It only ever runs on ours.
 
 ```python
 from auroraomics.runtimes.deepspotm import check_archive, describe_models
 
-cards = client.models()                                 # the catalogue is the service's
-describe_models(cards)                                  # what it will run, and on what
-check_archive("sample.zip", model=MODEL, cards=cards)   # are my tiles the right size?
+with ao.Client() as client:
+    cards = client.models()                                # the catalogue is the service's
+describe_models(cards)                                     # what it will run, and on what
+check_archive("sample.zip", model=model_id, cards=cards)   # are my tiles the right size?
 ```
 
 Then submit with the API client and read the result back — the same `.h5ad`
-contract whichever lane ran it.
+contract whichever runtime ran it.
 
 ## Post-process a result
 
@@ -237,17 +252,39 @@ skipped: a card is the description of a file, and a reader promised a layer
 cannot tell a missing one from a model that predicted zeros.
 
 The entry this version ships states what the model measured. `var` gains
-`n_train_datasets` — the number of training datasets each gene was measured in,
-because a gene seen in one and a gene seen in twenty are not the same claim —
-and `measured_in_training`, derived from that count. A gene the model never
-measured becomes `nan` in `X` and in every layer, not zero: zero is a
-measurement, and an unmeasured gene left as one averages, correlates and
+`measured_in_training`, which is `False` for a gene the model never saw
+measured. Such a gene becomes `nan` in `X` and in every layer, not zero: zero
+is a measurement, and an unmeasured gene left as one averages, correlates and
 colours a heat map exactly like a prediction.
 
+## Read a result in SpatialData or Seurat
+
+Every result is an `.h5ad`, which is an HDF5 file: `anndata` reads it, and so
+does any HDF5 library. For tools that read other formats, write it again:
+
+```
+auroraomics export result.h5ad --format seurat    # result_seurat/, for Read10X
+auroraomics export result.h5ad --format zarr      # result.zarr, a SpatialData store
+```
+
+```python
+from auroraomics.export import write_seurat_dir, write_zarr
+
+write_seurat_dir("result.h5ad", "result_seurat")
+write_zarr("result.h5ad", "result.zarr")    # pip install "auroraomics[spatialdata]"
+```
+
+Both keep the result's own values: a gene the model never measured stays blank
+(`nan`) in either format, never zero.
+
 ## Prepare a bulk RNA profile
 
 ```python
-report = ao.bulk_rna("sample.star_gene_counts.tsv", "sample.bulk.tsv", units="counts")
+with ao.Client() as client:
+    report = ao.bulk_rna(
+        "sample.star_gene_counts.tsv", "sample.bulk.tsv", units="counts",
+        card=client.model_card(model_id), resolve=client.resolve_keys,
+    )
 print(report.rows, "genes;", f"{report.coverage:.0%} of", report.coverage_of)
 ```
 
@@ -267,13 +304,6 @@ your project.
 
 ## Licence
 
-The code in this package is licensed under
-[PolyForm Noncommercial 1.0.0](https://polyformproject.org/licenses/noncommercial/1.0.0),
-which permits use for any purpose that is not commercial. It is the same licence the
-model package this client is built for carries, so installing both puts you under one
-rule rather than two.
-
-The model weights are licensed separately by whoever publishes them, and access to them
-may be gated. Read those terms before you use a model: they are not this licence, and a
-permission granted here is not a permission granted there.
+The code in this package is licensed for non-commercial use; its terms are in the
+`LICENSE` file it ships with. Commercial evaluation and use are governed by a written agreement with Aurora.
 

{auroraomics-0.1.0.dev2 → auroraomics-0.1.0.dev4}/README.md

@@ -1,9 +1,13 @@
 # auroraomics
 
-Virtual spatial transcriptomics from H&E histology.
+Virtual spatial transcriptomics from H&E images.
+
+This package is the client of the **Aurora API**, the hosted service that runs
+the prediction: from Python, from a shell with the `auroraomics` command, or
+from an agent through its MCP server.
 
 This package holds the pieces of that pipeline that are pure Python, so the
-same code runs on your laptop, in a GPU container and on the service:
+same code runs on your machine and on the service:
 
 - **`auroraomics.qc`** — the three patch-quality predicates (foreground, blur,
   stained tissue) applied to every candidate tile before a model sees it, and
@@ -15,16 +19,19 @@ same code runs on your laptop, in a GPU container and on the service:
   batch rather than the whole matrix.
 - **`auroraomics.postprocess`** — apply a model card's `postprocess` entries
   to a result file: a registry of entries, one runner, `h5py` alone, streaming
-  one chunk at a time. Ships the unmeasured-gene mask, which states per gene
-  how many training datasets measured it and blanks the ones none did.
+  one chunk at a time. Ships the unmeasured-gene mask, which blanks the genes
+  the model never measured.
 - **`auroraomics.subsample`** — pick the densest contiguous square of spots
-  when a slide yields more tiles than a run is allowed to spend.
-- **`auroraomics.genes`** — the model family's gene symbols mapped to Ensembl
-  stable gene ids, which is what every result's `var` index is keyed by. Ships
-  as one committed table, resolved in the same order the service resolves it.
+  when a slide yields more tiles than one submission takes.
+- **`auroraomics.genes`** — Ensembl stable gene ids, which is what every
+  result's `var` index is keyed by, and the shape a gene the service resolved
+  comes back in.
 - **`auroraomics.contracts`** — the shared contract values (container layout,
   input-kind caps, result layout) as data, so nothing here re-types a number
   the service also reads.
+- **`auroraomics.export`** — write a result as a SpatialData Zarr store
+  *(extra)* or as a folder Seurat loads, with the same matrix, coordinates and
+  gene names as the `.h5ad`.
 - **`auroraomics.embed`** *(extra)* — run a pinned image encoder over your
   tiles locally and write the embeddings container, so a prediction can be made
   from numbers instead of pixels.
@@ -45,16 +52,19 @@ pip install auroraomics
 
 That is the whole documented path: open a slide, judge and pack its tiles,
 submit, and read the result. One extra adds local embedding extraction
-(`embed`), because a tensor runtime is gigabytes and specific to the machine it
-was built for. No extra brings the model: predicting gene expression runs on
-the service, not here.
+(`embed`), and one writes a result as a SpatialData store (`spatialdata`). No
+extra brings the model: predicting gene expression runs on the service, not
+here.
 
 ## Pack tiles, then look at the report
 
 ```python
 import auroraomics as ao
 
-report = ao.pack_tiles(tiles, "sample.zip", mpp=0.499, thumbnail=thumb)
+with ao.Client() as client:
+    thresholds = client.qc_thresholds()   # the service's quality floors, no key needed
+
+report = ao.pack_tiles(tiles, "sample.zip", mpp=0.499, thresholds=thresholds, thumbnail=thumb)
 print(report.written, "tiles kept,", report.rejected, "dropped")
 print(report.rejected_by_reason)          # {'foreground_ratio': 12, ...}
 ```
@@ -83,7 +93,8 @@ members do not match the container contract.
 ao.write_result(
     "result.h5ad",
     obs=obs,                    # per-spot columns, as plain arrays
-    var=var,                    # per-gene columns, indexed by gene id
+    var=var,                    # per-gene columns
+    var_index=gene_ids,         # the Ensembl gene id of each column
     spatial=coords,             # (n_spots, 2) array -> obsm["spatial"]
     x=batches,                  # an array, or an iterable of row batches
     uns={"model": {"id": "..."}},
@@ -106,71 +117,69 @@ pip install "auroraomics[embed]"
 import auroraomics as ao
 from auroraomics.embed import available_encoders, describe_encoders
 
-print(available_encoders())        # ('deepspot-h', 'dinov2-b14', 'midnight')
-
 with ao.Client() as client:
     # Both are the service's, so the file records the encoder and the
-    # thresholds the model it is submitted to was served with.
-    encoder = ao.resolve_encoder(client.encoders())
+    # thresholds the model it is submitted to was served with. There is no
+    # list of encoders inside this package: every name comes from the registry
+    # the service publishes, which is why these calls take one.
+    registry = client.encoders()
     thresholds = client.qc_thresholds()
+    # The tile size the model's card states, for the model the file is for.
+    patch_um = client.model_card(model_id)["input_spec"]["patch_um"]
+
+print(available_encoders(registry))   # the names this package will run today
+encoder = ao.resolve_encoder(registry)
 
 report = ao.embed_tiles(
-    slide, "sample.npz", mpp=0.499, patch_um=55, encoder=encoder, thresholds=thresholds
+    slide, "sample.npz", mpp=0.499, patch_um=patch_um, encoder=encoder, thresholds=thresholds
 )
 print(report.rows, "rows of", report.dim, "numbers from", report.encoder.name)
 ```
 
 `slide` is anything with a `(height, width, 3)` RGB `uint8` shape that can be
 sliced — an array, or a memory-mapped or lazily-read one, so a slide larger than
-memory works: only one crop exists at a time. Crops are taken at `patch_um`
-micrometres using the `mpp` you give, quality-checked with the same three
-predicates as `pack_tiles`, and pooled over the encoder's patch tokens.
-
-Every encoder is pinned to a repository **and** a revision sha, with its licence
-and access gate recorded beside it. `describe_encoders()` returns the whole
-registry, including the encoders this package refuses to run — a refusal names
-the licence or the approval you would have to accept, and which encoders remain.
-
-The written file carries one extra vector, `calibration`: the embedding of a
-fixed image shipped inside this package, through the same encoder, revision and
-preprocessing as your rows. Comparing that one vector against a known reference
-tells a reader whether the file was produced by the model it claims, before they
-look at a single row.
-
-The weights are downloaded from their publisher on first use, into the ordinary
-model cache. Pass `allow_download=False` to guarantee no request is made: for
-the duration of that load, name resolution and internet sockets are refused in
-this process, so the guarantee does not rest on a model library reading an
-environment variable it may already have read.
-
-An encoder runs at one input size: the size the served model's embeddings were
-computed at, which the registry pins and which is not always what the encoder's
-own repository declares. The DINOv2 releases carry a 518-px position grid; the
-model was trained on 224-px crops fed to that grid through an interpolated
-position embedding, so that is what this package does too. `resize_px` is
-checked against the pin rather than obeyed, because a vector computed at another
-size is a different measurement that nobody else's numbers can be compared with.
+memory works: only one crop exists at a time. Crops are taken at the tile size
+the model's card states, using the `mpp` you give, and quality-checked with the
+same three predicates as `pack_tiles`.
+
+Every encoder is pinned to an exact revision.
+`describe_encoders(registry)` returns the whole registry, including the
+encoders this package refuses to run: a refusal says why, and which encoders
+remain.
+
+The written file carries one extra embedding, `calibration`: computed from a
+fixed image shipped inside this package, through the same encoder and revision
+as your rows. Comparing that one embedding against a known
+reference tells a reader whether the file was produced by the model it claims,
+before they look at a single row.
+
+The weights are downloaded on first use. Pass `allow_download=False` to
+guarantee no request is made: for the duration of that load, name resolution
+and internet sockets are refused in this process.
+
+An encoder runs at the input size the registry pins for it. `resize_px` is
+checked against that pin rather than obeyed, because an embedding computed at
+another size cannot be compared with the served model's.
+
 ## Where a prediction runs
 
-Predicting gene expression runs on the service, and `predict_local` refuses —
-by design, and the design is the product rather than a limitation.
+Predicting gene expression runs on the service, and `predict_local` refuses.
 
-This is split inference at the encoder. An open-weight pathology foundation
-model turns your tiles into patch embeddings **on your machine**, so your slides
-never leave it; our gene decoder turns embeddings into expression **on ours**, so
-its weights never leave it. What crosses between us is a vector from a model
-neither side owns.
+**DeepSpot-H**, the foundation model for H&E images, turns each tile of your
+slide into one embedding. It can run on your machine or on ours. **DeepSpot-M**
+turns those embeddings into spatial gene expression. It only ever runs on ours.
 
 ```python
 from auroraomics.runtimes.deepspotm import check_archive, describe_models
 
-cards = client.models()                                 # the catalogue is the service's
-describe_models(cards)                                  # what it will run, and on what
-check_archive("sample.zip", model=MODEL, cards=cards)   # are my tiles the right size?
+with ao.Client() as client:
+    cards = client.models()                                # the catalogue is the service's
+describe_models(cards)                                     # what it will run, and on what
+check_archive("sample.zip", model=model_id, cards=cards)   # are my tiles the right size?
 ```
 
 Then submit with the API client and read the result back — the same `.h5ad`
-contract whichever lane ran it.
+contract whichever runtime ran it.
 
 ## Post-process a result
 
@@ -194,17 +203,39 @@ skipped: a card is the description of a file, and a reader promised a layer
 cannot tell a missing one from a model that predicted zeros.
 
 The entry this version ships states what the model measured. `var` gains
-`n_train_datasets` — the number of training datasets each gene was measured in,
-because a gene seen in one and a gene seen in twenty are not the same claim —
-and `measured_in_training`, derived from that count. A gene the model never
-measured becomes `nan` in `X` and in every layer, not zero: zero is a
-measurement, and an unmeasured gene left as one averages, correlates and
+`measured_in_training`, which is `False` for a gene the model never saw
+measured. Such a gene becomes `nan` in `X` and in every layer, not zero: zero
+is a measurement, and an unmeasured gene left as one averages, correlates and
 colours a heat map exactly like a prediction.
 
+## Read a result in SpatialData or Seurat
+
+Every result is an `.h5ad`, which is an HDF5 file: `anndata` reads it, and so
+does any HDF5 library. For tools that read other formats, write it again:
+
+```
+auroraomics export result.h5ad --format seurat    # result_seurat/, for Read10X
+auroraomics export result.h5ad --format zarr      # result.zarr, a SpatialData store
+```
+
+```python
+from auroraomics.export import write_seurat_dir, write_zarr
+
+write_seurat_dir("result.h5ad", "result_seurat")
+write_zarr("result.h5ad", "result.zarr")    # pip install "auroraomics[spatialdata]"
+```
+
+Both keep the result's own values: a gene the model never measured stays blank
+(`nan`) in either format, never zero.
+
 ## Prepare a bulk RNA profile
 
 ```python
-report = ao.bulk_rna("sample.star_gene_counts.tsv", "sample.bulk.tsv", units="counts")
+with ao.Client() as client:
+    report = ao.bulk_rna(
+        "sample.star_gene_counts.tsv", "sample.bulk.tsv", units="counts",
+        card=client.model_card(model_id), resolve=client.resolve_keys,
+    )
 print(report.rows, "genes;", f"{report.coverage:.0%} of", report.coverage_of)
 ```
 
@@ -224,13 +255,6 @@ your project.
 
 ## Licence
 
-The code in this package is licensed under
-[PolyForm Noncommercial 1.0.0](https://polyformproject.org/licenses/noncommercial/1.0.0),
-which permits use for any purpose that is not commercial. It is the same licence the
-model package this client is built for carries, so installing both puts you under one
-rule rather than two.
-
-The model weights are licensed separately by whoever publishes them, and access to them
-may be gated. Read those terms before you use a model: they are not this licence, and a
-permission granted here is not a permission granted there.
+The code in this package is licensed for non-commercial use; its terms are in the
+`LICENSE` file it ships with. Commercial evaluation and use are governed by a written agreement with Aurora.
 

{auroraomics-0.1.0.dev2 → auroraomics-0.1.0.dev4}/pyproject.toml

@@ -4,14 +4,12 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "auroraomics"
-version = "0.1.0.dev2"
-description = "Virtual spatial transcriptomics from H&E histology: patch QC, tile packing and the .h5ad result contract."
+version = "0.1.0.dev4"
+description = "Prepare H&E slides on your own machine, submit them to Aurora, and read the spatial gene expression it predicts."
 readme = "README.md"
 requires-python = ">=3.10"
-# Non-commercial, and the same licence as the model package this client is built
-# for, so a user installing both faces one rule rather than two. PEP 639 forbids
-# a licence CLASSIFIER beside an expression, so this line is the whole
-# declaration; the model weights carry their own separate terms.
+# Non-commercial; the terms are in LICENSE. PEP 639 forbids a licence
+# CLASSIFIER beside an expression, so this line is the whole declaration.
 license = "PolyForm-Noncommercial-1.0.0"
 license-files = ["LICENSE"]
 authors = [{ name = "Kalin Nonchev" }]
@@ -47,9 +45,9 @@ classifiers = [
 # all of it. That is the trade, taken deliberately: one install line that works,
 # over a smaller one that cannot reach the end of the page describing it.
 #
-# What is still NOT here is the name that matters to the GPU pipeline images —
-# no torch. They install this package with `--no-deps`, so what they resolve
-# does not change either way; what protects them is the encoder line below.
+# What is still NOT here is torch. An environment that installs this package
+# with `--no-deps` resolves the same either way; what protects it is the
+# encoder line below.
 dependencies = [
   "numpy>=1.23",
   "h5py>=3.9",
@@ -66,12 +64,18 @@ dependencies = [
   # rather than being handed tiles they cut themselves.
   "tifffile>=2023.7.10",
   "imagecodecs>=2023.3.16",
+  # Writing a result as the folder Seurat loads (`auroraomics.export`): the
+  # matrix is written by scipy's Matrix Market writer. scipy already arrives
+  # behind anndata, so a plain install resolves nothing new for it; it is named
+  # because that module imports it directly, at anndata's own lower bound.
+  "scipy>1.8",
 ]
 
 [project.optional-dependencies]
-# TWO extras, and only one of them is a runtime: `embed` is the single opt-in
-# the documented path can need, and `mcp` is a protocol adapter for an agent.
-# Where that line is drawn is the whole of this comment.
+# THREE extras, and only one of them is a model runtime: `embed` is the single
+# opt-in the documented path can need, `mcp` is a protocol adapter for an agent,
+# and `spatialdata` writes a result in one more format. Where the first line is
+# drawn is the rest of this comment.
 #
 # There is no model-runtime extra, and that is the boundary rather than an
 # omission. Predicting gene expression runs on the service: the model definition
@@ -79,7 +83,7 @@ dependencies = [
 # resolves the model, or the adapter and checkpoint stack that would carry one,
 # and `predict_local` refuses with a message that says so. What runs on a user's
 # own machine is cutting, filtering and EMBEDDING tiles — their pixels stay put,
-# and a vector from the encoder is what crosses. So the line is at the
+# and an embedding from the encoder is what crosses. So the line is at the
 # ENCODER, not at the size of an install: `embed` below does resolve a tensor
 # runtime, on purpose, and everything on the CLIENT side of the encoder — the
 # HTTP client, the slide reader — sits in the core above, where nobody has to
@@ -111,6 +115,21 @@ embed = [
   "transformers>=4.40",
   "huggingface-hub>=0.23",
 ]
+# Writing a result as a SpatialData Zarr store (`auroraomics.export.write_zarr`,
+# `auroraomics export --format zarr`). An extra because it is large — some
+# seventy distributions and 700 MB on top of the core, measured on Python 3.10 —
+# and only someone who wants SpatialData's own format needs it; the Seurat
+# folder and the .h5ad need nothing from it.
+#
+# The setuptools bound is not a preference. Every SpatialData release before
+# 0.8 imports a schema library that reads `pkg_resources`, which setuptools 81
+# removed, so on a fresh environment the import fails with ModuleNotFoundError
+# however the rest resolves. 0.8 dropped that library and requires Python 3.12,
+# so the bound applies exactly where an older release is what pip can choose.
+spatialdata = [
+  "spatialdata>=0.2",
+  "setuptools<81; python_version < '3.12'",
+]
 # The MCP server: `auroraomics mcp` over stdio, so an agent reaches the same
 # client the CLI does. Only the protocol adapter is here; the tool table and
 # every handler are in the client, which is in the core now, so this extra names
@@ -137,7 +156,19 @@ dev = [
   # that spelled its own bound would be testing a dependency set no user can
   # install. The client needs no reference any more — it is the core.
   "auroraomics[mcp]",
+  # The SpatialData extra by reference, for the same reason: the export tests
+  # read every store back with SpatialData's own reader, so the environment
+  # that runs them installs what a user who asks for the format installs.
+  "auroraomics[spatialdata]",
   "pytest>=7",
+  # The release guard parses version strings with `packaging.version`, at module
+  # scope. It has always resolved, because more than one dependency in this list
+  # requires packaging — but arriving in a closure is not the same as being
+  # declared: the day whichever one carries it stops doing so, the check that
+  # decides whether a build may be uploaded stops being a check and becomes a
+  # collection error, which reports as one missing optional dependency rather
+  # than as a suite that no longer runs.
+  "packaging>=22",
   # Coverage is measured on every run, with a floor, because this package is
   # the one that SHIPS: a user installs it and calls it, so a branch nothing
   # here exercises is a branch discovered in the field. The floor sat at