genesets route for local server (#2079)

* first cut at GET /genesets route

* update existing tests to match code changes

* more GET /genesets and initial tests

* add missing test fixture

* geneset validation accepts OTA format

* genesets route: better error handling, more tests

* lint

local_server/common/annotations/annotations.py

@@ -3,19 +3,34 @@ from abc import ABCMeta, abstractmethod
 import fastobo
 import fsspec
 
-from local_server.common.errors import OntologyLoadFailure
+from local_server.common.errors import OntologyLoadFailure, DisabledFeatureError
 from local_server.common.utils.type_conversion_utils import get_schema_type_hint_of_array
 
 
 class Annotations(metaclass=ABCMeta):
-    """ baseclass for annotations, including ontologies"""
+    """ 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 genesets_save_enabled(self):
+        return self.config.get("genesets-save", False)
+
+    def check_user_annotations_enabled(self):
+        if not self.user_annotations_enabled():
+            raise DisabledFeatureError("User annotations are disabled.")
+
+    def check_genesets_save_enabled(self):
+        if not self.genesets_save_enabled():
+            raise DisabledFeatureError("User genesets save is disabled.")
 
     def load_ontology(self, path):
         """Load and parse ontologies - currently support OBO files only."""
@@ -64,7 +79,66 @@ class Annotations(metaclass=ABCMeta):
         """Write the labels (df) to a persistent storage such that it can later be read"""
         pass
 
+    @abstractmethod
+    def read_genesets(self, data_adaptor):
+        """Return the genesets from persistent storage """
+        pass
+
+    @abstractmethod
+    def write_genesets(self, gs, data_adaptor):
+        """Write the genesets (gs) to a persistent storage such that it can later be read"""
+        pass
+
     @abstractmethod
     def update_parameters(self, parameters, data_adaptor):
         """Update configuration parameters that describe information about the annotations feature"""
         pass
+
+    Genesets_Header = [
+        "geneset_name",
+        "geneset_description",
+        "gene_symbol",
+        "gene_description",
+    ]
+
+    @staticmethod
+    def genesets_to_csv(genesets):
+        """
+        Convert the internal genesets format (returned by read_geneset) into
+        the simple Tidy CSV.
+        """
+        from io import StringIO
+        import csv
+
+        if type(genesets) == dict:
+            genesets = genesets.values()
+
+        with StringIO() as sio:
+            writer = csv.writer(sio, dialect='excel')
+            writer.writerow(Annotations.Genesets_Header)
+            for geneset in genesets:
+                # genes may be empty, in which case we skip the geneset entirely
+                genes = geneset["genes"]
+                if not genes:
+                    writer.writerow([geneset["geneset_name"], geneset.get("geneset_description", ""), "", ""])
+                else:
+                    writer.writerows(
+                        [
+                            [
+                                geneset["geneset_name"],
+                                geneset.get("geneset_description", ""),
+                                gene["gene_symbol"],
+                                gene.get("gene_description", ""),
+                            ]
+                            for gene in genes
+                        ]
+                    )
+            return sio.getvalue()
+
+    @staticmethod
+    def genesets_to_response(genesets):
+        """
+        Convert the internal genesets format (returned by read_geneset) into
+        the dict expected by the JSON REST API
+        """
+        return list(genesets.values())

local_server/common/annotations/local_file_csv.py

@@ -4,29 +4,35 @@ import re
 import threading
 from datetime import datetime
 from hashlib import blake2b
+import csv
 
 import pandas as pd
 from flask import session, has_request_context, current_app
 
 from local_server import __version__ as cellxgene_version
 from local_server.common.annotations.annotations import Annotations
-from local_server.common.errors import AnnotationsError
+from local_server.common.errors import AnnotationsError, ObsoleteRequest
 
 
 class AnnotationsLocalFile(Annotations):
     CXG_ANNO_COLLECTION = "cxg_anno_collection"
 
-    def __init__(self, output_dir, output_file):
-        super().__init__()
+    def __init__(self, config, output_dir, label_output_file, genesets_output_file):
+        super().__init__(config)
         self.output_dir = output_dir
-        self.output_file = output_file
+        self.label_output_file = label_output_file
+        self.genesets_output_file = genesets_output_file
         # lock used to protect label file write ops
         self.label_lock = threading.RLock()
+        self.genesets_lock = threading.RLock()
 
-        # cache the most recent annotations
+        # cache the most recent annotations.
         self.last_fname = None
         self.last_labels = None
 
+        # txn ID - used to de-dup geneset writes
+        self.last_geneset_tid = 0
+
     def is_safe_collection_name(self, name):
         """
         return true if this is a safe collection name
@@ -47,11 +53,13 @@ class AnnotationsLocalFile(Annotations):
         return session.get(self.CXG_ANNO_COLLECTION)
 
     def read_labels(self, data_adaptor):
+        self.check_user_annotations_enabled()  # raises
+
         if has_request_context():
             if not current_app.auth.is_user_authenticated():
                 return pd.DataFrame()
 
-        fname = self._get_filename(data_adaptor)
+        fname = self._get_celllabels_filename(data_adaptor)
         with self.label_lock:
             if fname is not None and os.path.exists(fname) and os.path.getsize(fname) > 0:
                 # returned the cached labels if possible, otherwise read them from the file
@@ -69,6 +77,8 @@ class AnnotationsLocalFile(Annotations):
                 return pd.DataFrame()
 
     def write_labels(self, df, data_adaptor):
+        self.check_user_annotations_enabled()  # raises
+
         # update our internal state and save it.  Multi-threading often enabled,
         # so treat this as a critical section.
         with self.label_lock:
@@ -81,7 +91,7 @@ class AnnotationsLocalFile(Annotations):
                 f"which was last modified on {lastmodstr}\n"
             )
 
-            fname = self._get_filename(data_adaptor)
+            fname = self._get_celllabels_filename(data_adaptor)
             self._backup(fname)
             if not df.empty:
                 with open(fname, "w", newline="") as f:
@@ -95,12 +105,56 @@ class AnnotationsLocalFile(Annotations):
             self.last_fname = fname
             self.last_labels = df
 
+    def read_genesets(self, data_adaptor, context=None):
+        if has_request_context():
+            if not current_app.auth.is_user_authenticated():
+                return ([], None)
+
+        fname = self._get_genesets_filename(data_adaptor)
+        genesets = {}
+        tid = None
+        with self.genesets_lock:
+            tid = self.last_geneset_tid  # inside the critical section
+            if fname is not None and os.path.exists(fname) and os.path.getsize(fname) > 0:
+                with open(fname, newline="") as f:
+                    genesets = read_geneset_tidycsv(f, context)
+
+        return (genesets, tid)
+
+    def write_genesets(self, genesets, tid, data_adaptor):
+        self.check_genesets_save_enabled()  # raises
+
+        if type(tid) != int or tid < 0:
+            raise ValueError("tid must be a positive integer")
+
+        with self.genesets_lock:
+            # skip if the request is stale
+            if tid is not None:
+                if tid <= self.last_geneset_tid:
+                    raise ObsoleteRequest("TID is stale.")
+                self.last_geneset_tid = tid
+
+            lastmod = data_adaptor.get_last_mod_time()
+            lastmodstr = "'unknown'" if lastmod is None else lastmod.isoformat(timespec="seconds")
+            header = (
+                f"# Geneset generated on {datetime.now().isoformat(timespec='seconds')} "
+                f"using cellxgene version {cellxgene_version}\n"
+                f"# Input data file was {data_adaptor.get_location()}, "
+                f"which was last modified on {lastmodstr}\n"
+            )
+
+            fname = self._get_genesets_filename(data_adaptor)
+            self._backup(fname)
+            with open(fname, "w", newline="") as f:
+                f.write(header)
+                f.write(self.genesets_to_csv(genesets))
+
     def _get_userdata_idhash(self, data_adaptor):
         """
         Return a short hash that weakly identifies the user and dataset.
         Used to create safe annotations output file names.
         """
-        uid = current_app.auth.get_user_id()
+        uid = current_app.auth.get_user_id() or ""
         id = (uid + data_adaptor.get_location()).encode()
         idhash = base64.b32encode(blake2b(id, digest_size=5).digest()).decode("utf-8")
         return idhash
@@ -109,16 +163,27 @@ class AnnotationsLocalFile(Annotations):
         if self.output_dir:
             return self.output_dir
 
-        if self.output_file:
-            return os.path.dirname(self.path.abspath(self.output_dir))
+        output_file = self.label_output_file or self.genesets_output_file
+        if output_file:
+            return os.path.dirname(self.path.abspath(output_file))
 
         return os.getcwd()
 
-    def _get_filename(self, data_adaptor):
+    def _get_celllabels_filename(self, data_adaptor):
         """ return the current annotation file name """
-        if self.output_file:
-            return self.output_file
+        if self.label_output_file:
+            return self.label_output_file
 
+        return self._get_filename(data_adaptor, "celllabels")
+
+    def _get_genesets_filename(self, data_adaptor):
+        """ return the current genesets file name """
+        if self.genesets_output_file:
+            return self.genesets_output_file
+
+        return self._get_filename(data_adaptor, "genesets")
+
+    def _get_filename(self, data_adaptor, anno_name):
         # we need to generate a file name, which we can only do if we have a UID and collection name
         if session is None:
             raise AnnotationsError("unable to determine file name for annotations")
@@ -131,7 +196,7 @@ class AnnotationsLocalFile(Annotations):
             raise AnnotationsError("unable to determine file name for annotations")
 
         idhash = self._get_userdata_idhash(data_adaptor)
-        return os.path.join(self._get_output_dir(), f"{collection}-{idhash}.csv")
+        return os.path.join(self._get_output_dir(), f"{collection}-{anno_name}-{idhash}.csv")
 
     def _backup(self, fname, max_backups=9):
         """
@@ -170,7 +235,8 @@ class AnnotationsLocalFile(Annotations):
 
     def update_parameters(self, parameters, data_adaptor):
         params = {}
-        params["annotations"] = True
+        params["annotations"] = self.user_annotations_enabled()
+        params["annotations_genesets_readonly"] = not self.genesets_save_enabled()
         params["user_annotation_collection_name_enabled"] = True
 
         if self.ontology_data:
@@ -179,18 +245,108 @@ class AnnotationsLocalFile(Annotations):
         else:
             params["annotations_cell_ontology_enabled"] = False
 
-        if self.output_file is not None:
-            # user has hard-wired the name of the annotation data collection
-            fname = os.path.basename(self.output_file)
+        if self.label_output_file is not None:
+            # user has hard-wired the name of the annotation cell label data collection
+            fname = os.path.basename(self.label_output_file)
             collection_fname = os.path.splitext(fname)[0]
             params["annotations-data-collection-is-read-only"] = True
             params["annotations-data-collection-name"] = collection_fname
 
         elif session is not None:
             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-name"] = collection
+            params["annotations-data-collection-is-read-only"] = False
+            params["annotations-data-collection-name"] = collection
+
+        if current_app.auth.is_user_authenticated():
+            params["annotations-user-data-idhash"] = self._get_userdata_idhash(data_adaptor)
 
         parameters.update(params)
+
+
+def read_geneset_tidycsv(f, context=None):
+    """
+    Read & parse the Tidy CSV format, applying validation checks for mandatory
+    values, and de-duping rules.
+
+    Format is a four-column CSV, with a mandatory header row, and optional "#" prefixed
+    comments.  Format:
+
+        geneset_name, geneset_description, gene_symbol, gene_description
+
+    geneset_name and gene_symbol must be non-null; others are optional.
+
+    Returns: a dictionary of the shape (values in angle-brackets vary):
+
+        {
+            <string, a gene set name>: {
+                "geneset_name": <string, a gene set name>,
+                "geneset_description": <a string or None>,
+                "genes": [
+                    {
+                        "gene_symbol": <string, a gene symbol or name>,
+                        "gene_description": <a string or None>
+                    },
+                    ...
+                ]
+            },
+            ...
+        }
+    """
+
+    class myDialect(csv.excel):
+        skipinitialspace = True
+
+    def just(n, seq):
+        it = iter(seq)
+        for _ in range(n - 1):
+            yield next(it, "")
+        yield tuple(it)
+
+    messagefn = context["messagefn"] if context else (lambda x: None)
+
+    reader = csv.reader(f, dialect=myDialect())
+    genesets = {}
+    haveReadHeader = False
+    lineno = 0
+    for row in reader:
+        lineno += 1
+        # ignore empty rows
+        if len(row) == 0:
+            continue
+        # if row starts with '#' it is a comment
+        if row[0].startswith("#"):
+            continue
+        # if this is the first non-comment row, assume it is a header
+        if not haveReadHeader:
+            if row != Annotations.Genesets_Header:
+                raise AnnotationsError("Geneset CSV file missing the required column header.")
+            haveReadHeader = True
+            continue
+
+        geneset_name, geneset_description, gene_symbol, gene_description, _ = just(5, row)
+        if not geneset_name:
+            raise AnnotationsError(f"Geneset CSV missing required geneset or gene name on line {lineno}")
+        if (not gene_symbol) and gene_description:
+            messagefn(f"Warning: Missing gene name in geneset name {geneset_name} on line {lineno}.")
+
+        if geneset_name in genesets:
+            gs = genesets[geneset_name]
+        else:
+            gs = genesets[geneset_name] = {
+                "geneset_name": geneset_name,
+                "geneset_description": geneset_description,
+                "genes": [],
+            }
+        # Use first geneset_description with a value
+        if not gs["geneset_description"] and geneset_description:
+            gs["geneset_description"] = geneset_description
+        # add the gene if the gene_symbol is defined
+        if gene_symbol:
+            gs["genes"].append(
+                {
+                    "gene_symbol": gene_symbol,
+                    "gene_description": gene_description,
+                }
+            )
+
+    return genesets

local_server/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": dataset_config.user_annotations__genesets__readonly,
+        "annotations_genesets_summary_methods": ["mean"],
         "annotations_cell_ontology_enabled": False,
         "annotations_cell_ontology_obopath": None,
         "annotations_cell_ontology_terms": None,

local_server/common/config/dataset_config.py

@@ -3,7 +3,7 @@ from os.path import splitext, isdir
 
 from local_server.common.annotations.local_file_csv import AnnotationsLocalFile
 from local_server.common.config.base_config import BaseConfig
-from local_server.common.errors import ConfigurationError, OntologyLoadFailure
+from local_server.common.errors import ConfigurationError, OntologyLoadFailure, AnnotationsError
 from local_server.compute.scanpy import get_scanpy_module
 from local_server.data_common.matrix_loader import MatrixDataLoader
 
@@ -32,6 +32,10 @@ class DatasetConfig(BaseConfig):
             self.user_annotations__ontology__obo_location = default_config["user_annotations"]["ontology"][
                 "obo_location"
             ]
+            self.user_annotations__genesets__readonly = default_config["user_annotations"]["genesets"]["readonly"]
+            self.user_annotations__local_file_csv__genesets_file = default_config["user_annotations"]["local_file_csv"][
+                "genesets_file"
+            ]
 
             self.embeddings__names = default_config["embeddings"]["names"]
             self.embeddings__enable_reembedding = default_config["embeddings"]["enable_reembedding"]
@@ -56,9 +60,7 @@ class DatasetConfig(BaseConfig):
     def get_data_adaptor(self):
         server_config = self.app_config.server_config
         if not server_config.data_adaptor:
-            matrix_data_loader = MatrixDataLoader(
-                server_config.single_dataset__datapath, app_config=self.app_config
-            )
+            matrix_data_loader = MatrixDataLoader(server_config.single_dataset__datapath, app_config=self.app_config)
             server_config.data_adaptor = matrix_data_loader.open(self.app_config)
 
         return server_config.data_adaptor
@@ -96,11 +98,16 @@ class DatasetConfig(BaseConfig):
         self.validate_correct_type_of_configuration_attribute(
             "user_annotations__local_file_csv__file", (type(None), str)
         )
+        self.validate_correct_type_of_configuration_attribute(
+            "user_annotations__local_file_csv__genesets_file", (type(None), str)
+        )
         self.validate_correct_type_of_configuration_attribute("user_annotations__ontology__enable", bool)
         self.validate_correct_type_of_configuration_attribute(
             "user_annotations__ontology__obo_location", (type(None), str)
         )
-        if self.user_annotations__enable:
+        self.validate_correct_type_of_configuration_attribute("user_annotations__genesets__readonly", bool)
+
+        if self.user_annotations__enable or not self.user_annotations__genesets__readonly:
             server_config = self.app_config.server_config
             if not self.app__authentication_enable:
                 raise ConfigurationError("user annotations requires authentication to be enabled")
@@ -108,59 +115,84 @@ class DatasetConfig(BaseConfig):
                 auth_type = server_config.authentication__type
                 raise ConfigurationError(f"authentication method {auth_type} is not compatible with user annotations")
 
-            if self.user_annotations__type == "local_file_csv":
-                self.handle_local_file_csv_annotations()
-            else:
-                raise ConfigurationError('The only annotation type support is "local_file_csv"')
+        # Must always have an annotations instance to support genesets. User annotation (cell labels) are optional
+        # as are writable gene sets
+        if self.user_annotations__type == "local_file_csv":
+            self.handle_local_file_csv_annotations(context)
+        else:
+            raise ConfigurationError('The only annotation type support is "local_file_csv"')
+
+        if self.user_annotations__enable:
             if self.user_annotations__ontology__enable or self.user_annotations__ontology__obo_location:
                 try:
                     self.user_annotations.load_ontology(self.user_annotations__ontology__obo_location)
                 except OntologyLoadFailure as e:
                     raise ConfigurationError("Unable to load ontology terms\n" + str(e))
-        else:
-            self.check_annotation_config_vars_not_set(context)
 
-    def handle_local_file_csv_annotations(self):
+        self.check_annotation_config_vars_not_set(context)
+
+    def handle_local_file_csv_annotations(self, context):
         dirname = self.user_annotations__local_file_csv__directory
         filename = self.user_annotations__local_file_csv__file
-        if filename is not None and dirname is not None:
-            raise ConfigurationError("'annotations-file' and 'annotations-dir' may not be used together.")
+        genesets_filename = self.user_annotations__local_file_csv__genesets_file
+
+        if dirname is not None and (filename is not None or genesets_filename is not None):
+            raise ConfigurationError(
+                "'user-generated-data-dir' may not be used with annotations-file' or 'genesets-file'."
+            )
 
         if filename is not None:
             lf_name, lf_ext = splitext(filename)
             if lf_ext and lf_ext != ".csv":
                 raise ConfigurationError(f"annotation file type must be .csv: {filename}")
 
+        if genesets_filename is not None:
+            lf_name, lf_ext = splitext(genesets_filename)
+            if lf_ext and lf_ext != ".csv":
+                raise ConfigurationError(f"genesets file type must be .csv: {genesets_filename}")
+
         if dirname is not None and not isdir(dirname):
             try:
                 os.mkdir(dirname)
             except OSError:
-                raise ConfigurationError("Unable to create directory specified by --annotations-dir")
+                raise ConfigurationError("Unable to create directory specified by --user-generated-data-dir")
 
-        self.user_annotations = AnnotationsLocalFile(dirname, filename)
+        anno_config = {
+            "user-annotations": self.user_annotations__enable,
+            "genesets-save": not self.user_annotations__genesets__readonly,
+        }
+        self.user_annotations = AnnotationsLocalFile(anno_config, dirname, filename, genesets_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.
         server_config = self.app_config.server_config
-        if server_config.single_dataset__datapath and self.user_annotations__local_file_csv__file:
+        if server_config.single_dataset__datapath:
             data_adaptor = self.get_data_adaptor()
-            data_adaptor.check_new_labels(self.user_annotations.read_labels(data_adaptor))
+            if self.user_annotations__local_file_csv__file:
+                data_adaptor.check_new_labels(self.user_annotations.read_labels(data_adaptor))
+            if self.user_annotations__local_file_csv__genesets_file:
+                try:
+                    data_adaptor.check_new_genesets(self.user_annotations.read_genesets(data_adaptor, context), context)
+                except (ValueError, AnnotationsError, KeyError) as e:
+                    raise ConfigurationError(f"Unable to read genesets CSV file: {str(e)}") from e
 
     def check_annotation_config_vars_not_set(self, context):
         if self.user_annotations__type is not None:
             dirname = self.user_annotations__local_file_csv__directory
             filename = self.user_annotations__local_file_csv__file
-            if filename is not None:
-                context["messagefn"]("Warning: --annotations-file ignored as annotations are disabled.")
-            if dirname is not None:
-                context["messagefn"]("Warning: --annotations-dir ignored as annotations are disabled.")
-
-        if self.user_annotations__ontology__enable:
-            context["messagefn"]("Warning: --experimental-annotations-ontology ignored as annotations are disabled.")
-        if self.user_annotations__ontology__obo_location is not None:
-            context["messagefn"](
-                "Warning: --experimental-annotations-ontology-obo ignored as annotations are disabled."
-            )
+            if not self.user_annotations__enable:
+                if filename is not None:
+                    context["messagefn"]("Warning: --annotations-file ignored as annotations are disabled.")
+                if self.user_annotations__ontology__enable:
+                    context["messagefn"](
+                        "Warning: --experimental-annotations-ontology ignored as annotations are disabled."
+                    )
+                if self.user_annotations__ontology__obo_location is not None:
+                    context["messagefn"](
+                        "Warning: --experimental-annotations-ontology-obo ignored as annotations are disabled."
+                    )
+                if dirname is not None:
+                    context["messagefn"]("Warning: --user-generated-data-dir ignored as annotations are disabled.")
 
     def handle_embeddings(self):
         self.validate_correct_type_of_configuration_attribute("embeddings__names", list)
@@ -186,6 +218,5 @@ class DatasetConfig(BaseConfig):
         data_adaptor = self.get_data_adaptor()
         if self.diffexp__enable and data_adaptor.parameters.get("diffexp_may_be_slow", False):
             context["messagefn"](
-                "CAUTION: due to the size of your dataset, "
-                "running differential expression may take longer or fail."
+                "CAUTION: due to the size of your dataset, " "running differential expression may take longer or fail."
             )

local_server/common/errors.py

@@ -55,3 +55,4 @@ define_exception("OntologyLoadFailure", "Raised when reading the ontology file f
 define_exception("ConfigurationError", "Raised when checking configuration errors")
 define_exception("PrepareError", "Raised when data is misprepared")
 define_exception("SecretKeyRetrievalError", "Raised when get_secret_key from AWS fails")
+define_exception("ObsoleteRequest", "Raised when the request is no longer valid.")

local_server/common/rest.py

@@ -17,6 +17,8 @@ from local_server.common.errors import (
     ExceedsLimitError,
     DatasetAccessError,
     ColorFormatException,
+    AnnotationsError,
+    ObsoleteRequest,
 )
 
 import json
@@ -44,7 +46,7 @@ def _query_parameter_to_filter(args):
 
     Query param filters look like:  <axis>:name=value, where value
     may be one of:
-        - a range, min,max, where either may be an open range by using an asterisc, eg, 10,*
+        - a range, min,max, where either may be an open range by using an asterisk, eg, 10,*
         - a value
     Eg,
         ...?tissue=lung&obs:tissue=heart&obs:num_reads=1000,*
@@ -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)
@@ -196,9 +198,6 @@ def annotations_var_get(request, data_adaptor):
 
     try:
         labels = None
-        annotations = data_adaptor.dataset_config.user_annotations
-        if annotations is not None:
-            labels = annotations.read_labels(data_adaptor)
         return make_response(
             data_adaptor.annotation_to_fbs_matrix(Axis.VAR, fields, labels),
             HTTPStatus.OK,
@@ -328,3 +327,56 @@ 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) = data_adaptor.check_new_genesets(annotations.read_genesets(data_adaptor))
+
+        if preferred_mimetype == "text/csv":
+            return make_response(
+                annotations.genesets_to_csv(genesets),
+                HTTPStatus.OK,
+                {
+                    "Content-Type": "text/csv",
+                    "Content-Disposition": "attachment; filename=genesets.csv",
+                },
+            )
+        else:
+            return make_response(
+                jsonify({"genesets": annotations.genesets_to_response(genesets), "tid": tid}), HTTPStatus.OK
+            )
+    except (ValueError, KeyError, AnnotationsError) as e:
+        return abort_and_log(HTTPStatus.BAD_REQUEST, str(e))
+
+
+def genesets_put(request, data_adaptor):
+    annotations = data_adaptor.dataset_config.user_annotations
+    if not annotations.genesets_save_enabled():
+        return abort(HTTPStatus.NOT_IMPLEMENTED)
+
+    anno_collection = request.args.get("annotation-collection-name", default=None)
+    if anno_collection is not None:
+        if not annotations.is_safe_collection_name(anno_collection):
+            return abort(HTTPStatus.BAD_REQUEST, "Bad annotation collection name")
+        annotations.set_collection(anno_collection)
+
+    args = request.get_json()
+    try:
+        genesets = args.get("genesets", None)
+        tid = args.get("tid", None)
+        if genesets is None:
+            abort(HTTPStatus.BAD_REQUEST)
+
+        (gs, _) = data_adaptor.check_new_genesets((genesets, tid))
+        annotations.write_genesets(gs, tid, data_adaptor)
+        return make_response(jsonify({"status": "OK"}), HTTPStatus.OK)
+    except (ValueError, DisabledFeatureError, KeyError) as e:
+        return abort_and_log(HTTPStatus.BAD_REQUEST, str(e), include_exc_info=True)
+    except (ObsoleteRequest, TypeError) as e:
+        return abort(HTTPStatus.NOT_FOUND, description=str(e))