hosted gene sets routes, plus a few bug fixes (#2155)

* first cut at hosted gs routes

* lint

* update tests to match csv parser changes

* update tests to new API

* update gene set name validation rules to match requirements

* add path mapping from dataset to geneset

* add test cases for geneset GET route

* fix test assertion

* remove debugging code

* update gene set uri mapping function

* fix error message

* allow extra user-specified headers in gene set csv file

* clarify comment

backend/czi_hosted/app/app.py

@@ -318,6 +318,25 @@ class LayoutObsAPI(DatasetResource):
         return common_rest.layout_obs_put(request, data_adaptor)
 
 
+class GenesetsAPI(DatasetResource):
+    @cache_control(public=True, max_age=ONE_WEEK)
+    @rest_get_data_adaptor
+    def get(self, data_adaptor):
+        return common_rest.genesets_get(request, data_adaptor)
+
+
+class SummarizeVarAPI(DatasetResource):
+    @rest_get_data_adaptor
+    @cache_control(public=True, max_age=ONE_WEEK)
+    def get(self, data_adaptor):
+        return common_rest.summarize_var_get(request, data_adaptor)
+
+    @rest_get_data_adaptor
+    @cache_control(no_store=True)
+    def post(self, data_adaptor):
+        return common_rest.summarize_var_post(request, data_adaptor)
+
+
 def get_api_base_resources(bp_base):
     """Add resources that are accessed from the api_base_url"""
     api = Api(bp_base)
@@ -343,6 +362,8 @@ def get_api_dataroot_resources(bp_dataroot, url_dataroot=None):
     add_resource(AnnotationsObsAPI, "/annotations/obs")
     add_resource(AnnotationsVarAPI, "/annotations/var")
     add_resource(DataVarAPI, "/data/var")
+    add_resource(GenesetsAPI, "/genesets")
+    add_resource(SummarizeVarAPI, "/summarize/var")
     # Display routes
     add_resource(ColorsAPI, "/colors")
     # Computation routes

backend/czi_hosted/common/annotations/annotations.py

@@ -1,21 +1,33 @@
-from abc import ABCMeta, abstractmethod
-
 import fastobo
 import fsspec
+import os
 
-from backend.common.errors import OntologyLoadFailure
+from flask import current_app, has_request_context
+
+from backend.common.errors import OntologyLoadFailure, DisabledFeatureError
 from backend.common.utils.type_conversion_utils import get_schema_type_hint_of_array
+from backend.common.genesets import write_gene_sets_tidycsv, read_gene_sets_tidycsv, validate_gene_sets
+from backend.common.utils.data_locator import DataLocator
+from backend.common.utils.utils import path_join
 
 
-class Annotations(metaclass=ABCMeta):
-    """ baseclass for annotations, including ontologies"""
+class Annotations:
+    """ baseclass for annotations, including ontologies and genesets """
 
     """ our default ontology is the PURL for the Cell Ontology.
     See http://www.obofoundry.org/ontology/cl.html """
     DefaultOnotology = "http://purl.obolibrary.org/obo/cl.obo"
 
-    def __init__(self):
+    def __init__(self, config={}):
         self.ontology_data = None
+        self.config = config
+
+    def user_annotations_enabled(self):
+        return self.config.get("user-annotations", False)
+
+    def check_user_annotations_enabled(self):
+        if not self.user_annotations_enabled():
+            raise DisabledFeatureError("User annotations are disabled.")
 
     def load_ontology(self, path):
         """Load and parse ontologies - currently support OBO files only."""
@@ -49,22 +61,78 @@ class Annotations(metaclass=ABCMeta):
 
         return schema
 
-    @abstractmethod
     def set_collection(self, name):
         """set or create a new annotation collection"""
-        pass
+        raise NotImplementedError
 
-    @abstractmethod
     def read_labels(self, data_adaptor):
         """Return the labels as a pandas.DataFrame"""
-        pass
+        raise NotImplementedError
 
-    @abstractmethod
     def write_labels(self, df, data_adaptor):
         """Write the labels (df) to a persistent storage such that it can later be read"""
-        pass
+        raise NotImplementedError
 
-    @abstractmethod
     def update_parameters(self, parameters, data_adaptor):
         """Update configuration parameters that describe information about the annotations feature"""
-        pass
+        params = {}
+        params["annotations_genesets_readonly"] = True
+        params["annotations_genesets_name_is_read_only"] = True
+        parameters.update(params)
+
+    @staticmethod
+    def gene_sets_to_csv(genesets):
+        """
+        Convert the internal genesets format (returned by read_gene_set) into
+        the simple Tidy CSV.
+        """
+        from io import StringIO
+
+        if isinstance(genesets, dict):
+            genesets = genesets.values()
+
+        with StringIO() as sio:
+            write_gene_sets_tidycsv(sio, genesets)
+            return sio.getvalue()
+
+    @staticmethod
+    def gene_sets_to_response(genesets):
+        """
+        Convert the internal genesets format (returned by read_gene_set) into
+        the dict expected by the JSON REST API
+        """
+        return list(genesets.values())
+
+    def read_gene_sets(self, data_adaptor, context=None):
+        if has_request_context():
+            if not current_app.auth.is_user_authenticated():
+                return ({}, 0)
+
+        gene_sets_uri_or_path = dataset_uri_to_geneset_uri(data_adaptor.data_locator.uri_or_path)
+
+        server_config = data_adaptor.server_config
+        region_name = None if server_config is None else server_config.data_locator__s3__region_name
+        gene_sets_locator = DataLocator(gene_sets_uri_or_path, region_name=region_name)
+        if not gene_sets_locator.exists():
+            return ({}, 0)
+
+        gene_sets = read_gene_sets_tidycsv(gene_sets_locator, context)
+        schema = data_adaptor.get_schema()
+        var_index = schema["annotations"]["var"].get("index", "index")
+        var_names = set(data_adaptor.query_var_array(var_index))
+
+        gene_sets = validate_gene_sets(gene_sets, var_names)
+        return (gene_sets, 0)
+
+
+def dataset_uri_to_geneset_uri(data_uri_or_path):
+    """ given a dataset URI, return the associated gene set URI """
+    data_basename = os.path.basename(data_uri_or_path)
+    base, ext = os.path.splitext(data_basename)
+    if ext is not None:  # strip extension, if any
+        data_basename = base
+
+    genesets_basename = f"{data_basename}-genesets.csv"
+    gene_sets_uri_or_path = path_join(data_uri_or_path, "..", genesets_basename)
+
+    return gene_sets_uri_or_path

backend/czi_hosted/common/annotations/hosted_tiledb.py

@@ -17,8 +17,8 @@ from backend.czi_hosted.db.cellxgene_orm import Annotation
 class AnnotationsHostedTileDB(Annotations):
     CXG_ANNO_COLLECTION = "cxg_anno_collection"
 
-    def __init__(self, directory_path, db):
-        super().__init__()
+    def __init__(self, config, directory_path, db):
+        super().__init__(config)
         self.db = db
         if directory_path[-1] == "/":
             self.directory_path = directory_path
@@ -158,6 +158,8 @@ class AnnotationsHostedTileDB(Annotations):
         self.db.session.commit()
 
     def update_parameters(self, parameters, data_adaptor):
+        super().update_parameters(parameters, data_adaptor)
+
         params = {}
         params["annotations"] = True
         params["user_annotation_collection_name_enabled"] = False

backend/czi_hosted/common/annotations/local_file_csv.py

@@ -16,8 +16,8 @@ from backend.common.errors import AnnotationsError
 class AnnotationsLocalFile(Annotations):
     CXG_ANNO_COLLECTION = "cxg_anno_collection"
 
-    def __init__(self, output_dir, output_file):
-        super().__init__()
+    def __init__(self, config, output_dir, output_file):
+        super().__init__(config)
         self.output_dir = output_dir
         self.output_file = output_file
         # lock used to protect label file write ops
@@ -169,6 +169,8 @@ class AnnotationsLocalFile(Annotations):
                 os.remove(os.path.join(backup_dir, bu))
 
     def update_parameters(self, parameters, data_adaptor):
+        super().update_parameters(parameters, data_adaptor)
+
         params = {}
         params["annotations"] = True
         params["user_annotation_collection_name_enabled"] = True
@@ -190,7 +192,7 @@ class AnnotationsLocalFile(Annotations):
             collection = self.get_collection()
             if current_app.auth.is_user_authenticated():
                 params["annotations-user-data-idhash"] = self._get_userdata_idhash(data_adaptor)
-                params["annotations-data-collection-is-read-only"] = False
+                params["annotations-data-collection-is-read-only"] = not self.user_annotations_enabled()
                 params["annotations-data-collection-name"] = collection
 
         parameters.update(params)

backend/czi_hosted/common/config/client_config.py

@@ -44,6 +44,9 @@ def get_client_config(app_config, data_adaptor):
         "annotations": False,
         "annotations_file": None,
         "annotations_dir": None,
+        "annotations_genesets": True,  # feature flag
+        "annotations_genesets_readonly": True,
+        "annotations_genesets_summary_methods": ["mean"],
         "annotations_cell_ontology_enabled": False,
         "annotations_cell_ontology_obopath": None,
         "annotations_cell_ontology_terms": None,

backend/czi_hosted/common/config/dataset_config.py

@@ -1,6 +1,7 @@
 import os
 from os.path import splitext, isdir
 
+from backend.czi_hosted.common.annotations.annotations import Annotations
 from backend.czi_hosted.common.annotations.hosted_tiledb import AnnotationsHostedTileDB
 from backend.czi_hosted.common.annotations.local_file_csv import AnnotationsLocalFile
 from backend.czi_hosted.common.config.base_config import BaseConfig
@@ -53,8 +54,10 @@ class DatasetConfig(BaseConfig):
         except KeyError as e:
             raise ConfigurationError(f"Unexpected config: {str(e)}")
 
-        # The annotation object is created during complete_config and stored here.
-        self.user_annotations = None
+        # Create the default annotation, which supports gene set reading without
+        # further configuration. Depending on configuration options, `complete_config` 
+        # may create a more specialized annotation object and replace this default.
+        self.user_annotations = Annotations()
 
     def complete_config(self, context):
         self.handle_app()
@@ -147,7 +150,11 @@ class DatasetConfig(BaseConfig):
             except OSError:
                 raise ConfigurationError("Unable to create directory specified by --annotations-dir")
 
-        self.user_annotations = AnnotationsLocalFile(dirname, filename)
+        anno_config = {
+            "user-annotations": self.user_annotations__enable,
+            "genesets-save": False,
+        }
+        self.user_annotations = AnnotationsLocalFile(anno_config, dirname, filename)
 
         # if the user has specified a fixed label file, go ahead and validate it
         # so that we can remove errors early in the process.
@@ -163,7 +170,12 @@ class DatasetConfig(BaseConfig):
         self.validate_correct_type_of_configuration_attribute(
             "user_annotations__hosted_tiledb_array__hosted_file_directory", str
         )
+        anno_config = {
+            "user-annotations": self.user_annotations__enable,
+            "genesets-save": False,
+        }
         self.user_annotations = AnnotationsHostedTileDB(
+            anno_config,
             directory_path=self.user_annotations__hosted_tiledb_array__hosted_file_directory,
             db=DbUtils(self.user_annotations__hosted_tiledb_array__db_uri),
         )

backend/czi_hosted/common/rest.py

@@ -3,6 +3,7 @@ import logging
 import sys
 from http import HTTPStatus
 import zlib
+import json
 
 from flask import make_response, jsonify, current_app, abort
 from werkzeug.urls import url_unquote
@@ -17,9 +18,10 @@ from backend.common.errors import (
     ExceedsLimitError,
     DatasetAccessError,
     ColorFormatException,
+    AnnotationsError,
+    UnsupportedSummaryMethod,
 )
-
-import json
+from backend.common.genesets import summarizeQueryHash
 from backend.common.fbs.matrix import decode_matrix_fbs
 
 
@@ -106,7 +108,7 @@ def schema_get_helper(data_adaptor):
 
     # add label obs annotations as needed
     annotations = data_adaptor.dataset_config.user_annotations
-    if annotations is not None:
+    if annotations.user_annotations_enabled():
         label_schema = annotations.get_schema(data_adaptor)
         schema["annotations"]["obs"]["columns"].extend(label_schema)
 
@@ -140,7 +142,7 @@ def annotations_obs_get(request, data_adaptor):
     try:
         labels = None
         annotations = data_adaptor.dataset_config.user_annotations
-        if annotations:
+        if annotations.user_annotations_enabled():
             labels = annotations.read_labels(data_adaptor)
         fbs = data_adaptor.annotation_to_fbs_matrix(Axis.OBS, fields, labels)
         return make_response(fbs, HTTPStatus.OK, {"Content-Type": "application/octet-stream"})
@@ -151,7 +153,7 @@ def annotations_obs_get(request, data_adaptor):
 def annotations_put_fbs_helper(data_adaptor, fbs):
     """helper function to write annotations from fbs"""
     annotations = data_adaptor.dataset_config.user_annotations
-    if annotations is None:
+    if not annotations.user_annotations_enabled():
         raise DisabledFeatureError("Writable annotations are not enabled")
 
     new_label_df = decode_matrix_fbs(fbs)
@@ -166,7 +168,7 @@ def inflate(data):
 
 def annotations_obs_put(request, data_adaptor):
     annotations = data_adaptor.dataset_config.user_annotations
-    if annotations is None:
+    if not annotations.user_annotations_enabled():
         return abort(HTTPStatus.NOT_IMPLEMENTED)
 
     anno_collection = request.args.get("annotation-collection-name", default=None)
@@ -197,7 +199,7 @@ def annotations_var_get(request, data_adaptor):
     try:
         labels = None
         annotations = data_adaptor.dataset_config.user_annotations
-        if annotations is not None:
+        if annotations.user_annotations_enabled():
             labels = annotations.read_labels(data_adaptor)
         return make_response(
             data_adaptor.annotation_to_fbs_matrix(Axis.VAR, fields, labels),
@@ -328,3 +330,70 @@ def layout_obs_put(request, data_adaptor):
         return abort_and_log(HTTPStatus.NOT_IMPLEMENTED, str(e))
     except (ValueError, DisabledFeatureError, FilterError) as e:
         return abort_and_log(HTTPStatus.BAD_REQUEST, str(e), include_exc_info=True)
+
+
+def genesets_get(request, data_adaptor):
+    preferred_mimetype = request.accept_mimetypes.best_match(["application/json", "text/csv"])
+    if preferred_mimetype not in ("application/json", "text/csv"):
+        return abort(HTTPStatus.NOT_ACCEPTABLE)
+
+    try:
+        annotations = data_adaptor.dataset_config.user_annotations
+        (genesets, tid) = annotations.read_gene_sets(data_adaptor)
+
+        if preferred_mimetype == "text/csv":
+            return make_response(
+                annotations.gene_sets_to_csv(genesets),
+                HTTPStatus.OK,
+                {
+                    "Content-Type": "text/csv",
+                    "Content-Disposition": "attachment; filename=genesets.csv",
+                },
+            )
+        else:
+            return make_response(
+                jsonify({"genesets": annotations.gene_sets_to_response(genesets), "tid": tid}), HTTPStatus.OK
+            )
+    except (ValueError, KeyError, AnnotationsError) as e:
+        return abort_and_log(HTTPStatus.BAD_REQUEST, str(e))
+
+
+def summarize_var_helper(request, data_adaptor, key, raw_query):
+    preferred_mimetype = request.accept_mimetypes.best_match(["application/octet-stream"])
+    if preferred_mimetype != "application/octet-stream":
+        return abort(HTTPStatus.NOT_ACCEPTABLE)
+
+    summary_method = request.values.get("method", default="mean")
+    query_hash = summarizeQueryHash(raw_query)
+    if key and query_hash != key:
+        return abort(HTTPStatus.BAD_REQUEST, description="query key did not match")
+
+    args_filter_only = request.values.copy()
+    args_filter_only.poplist("method")
+    args_filter_only.poplist("key")
+
+    try:
+        filter = _query_parameter_to_filter(args_filter_only)
+        return make_response(
+            data_adaptor.summarize_var(summary_method, filter, query_hash),
+            HTTPStatus.OK,
+            {"Content-Type": "application/octet-stream"},
+        )
+    except (ValueError) as e:
+        return abort(HTTPStatus.NOT_FOUND, description=str(e))
+    except (UnsupportedSummaryMethod, FilterError) as e:
+        return abort(HTTPStatus.BAD_REQUEST, description=str(e))
+
+
+def summarize_var_get(request, data_adaptor):
+    return summarize_var_helper(request, data_adaptor, None, request.query_string)
+
+
+def summarize_var_post(request, data_adaptor):
+    if not request.content_type or "application/x-www-form-urlencoded" not in request.content_type:
+        return abort(HTTPStatus.UNSUPPORTED_MEDIA_TYPE)
+    if request.content_length > 1_000_000:  # just a sanity check to avoid memory exhaustion
+        return abort(HTTPStatus.BAD_REQUEST)
+
+    key = request.args.get("key", default=None)
+    return summarize_var_helper(request, data_adaptor, key, request.get_data())

backend/czi_hosted/data_common/data_adaptor.py

@@ -3,11 +3,12 @@ from os.path import basename, splitext
 
 import numpy as np
 import pandas as pd
+from scipy import sparse
 from server_timing import Timing as ServerTiming
 
 from backend.czi_hosted.common.config.app_config import AppConfig
 from backend.common.constants import Axis
-from backend.common.errors import FilterError, JSONEncodingValueError, ExceedsLimitError
+from backend.common.errors import FilterError, JSONEncodingValueError, ExceedsLimitError, UnsupportedSummaryMethod
 from backend.common.utils.utils import jsonify_numpy
 from backend.common.fbs.matrix import encode_matrix_fbs
 
@@ -338,7 +339,7 @@ class DataAdaptor(metaclass=ABCMeta):
     @staticmethod
     def normalize_embedding(embedding):
         """Normalize embedding layout to meet client assumptions.
-           Embedding is an ndarray, shape (n_obs, n)., where n is normally 2
+        Embedding is an ndarray, shape (n_obs, n)., where n is normally 2
         """
 
         # scale isotropically
@@ -394,3 +395,26 @@ class DataAdaptor(metaclass=ABCMeta):
         except RuntimeError:
             lastmod = None
         return lastmod
+
+    def summarize_var(self, method, filter, query_hash):
+        if method != "mean":
+            raise UnsupportedSummaryMethod("Unknown gene set summary method.")
+
+        obs_selector, var_selector = self._filter_to_mask(filter)
+        if obs_selector is not None:
+            raise FilterError("filtering on obs unsupported")
+
+        # if no filter, just return zeros.  We don't have a use case
+        # for summarizing the entire X without a filter, and it would
+        # potentially be quite compute / memory intensive.
+        if var_selector is None or np.count_nonzero(var_selector) == 0:
+            mean = np.zeros((self.get_shape()[0], 1), dtype=np.float32)
+        else:
+            X = self.get_X_array(obs_selector, var_selector)
+            if sparse.issparse(X):
+                mean = X.mean(axis=1)
+            else:
+                mean = X.mean(axis=1, keepdims=True)
+
+        col_idx = pd.Index([query_hash])
+        return encode_matrix_fbs(mean, col_idx=col_idx, row_idx=None)