Release 1.38.0

.github/workflows/minify_ontologies.yml

@@ -3,9 +3,9 @@ name: Minify ontologies
 on:
   pull_request:
     types: [opened]  # Only trigger on PR "opened" event
-#  push: # Uncomment, update branches to develop / debug
-#    branches:
-#      jlc_show_gene_name 
+# push: # Uncomment, update branches to develop / debug
+#   branches:
+#     jlc_show_de_pairwise
 
 jobs:
   build:

ingest/anndata_.py

@@ -6,6 +6,7 @@
 import scipy
 from scipy.io.mmio import MMFile
 
+
 # scipy.io.mmwrite uses scientific notation by default
 # https://stackoverflow.com/questions/64748513
 class MMFileFixedFormat(MMFile):
@@ -72,7 +73,7 @@ def create_cell_data_arrays(self):
             linear_data_id=self.study_file_id,
             cluster_name=raw_filename,
             study_file_id=self.study_file_id,
-            study_id=self.study_id
+            study_id=self.study_id,
         ):
             data_arrays.append(data_array)
 

ingest/cell_metadata.py

@@ -7,6 +7,7 @@
 PREREQUISITES
 Must have python 3.6 or higher.
 """
+
 import collections
 import ntpath
 from collections import defaultdict, OrderedDict
@@ -115,6 +116,7 @@ def validate_header_for_coordinate_values(self):
                 "error", msg, "format:cap:metadata-no-coordinates"
             )
             return False
+
     @staticmethod
     def make_multiindex_name(modality):
         """From modality, generate column name in multi-index format"""
@@ -127,7 +129,9 @@ def create_boolean_modality_metadatum(df, modality):
         """Translate presence of single modality to boolean for BigQuery"""
         # check for empty cells (aka. nan) or empty strings
         modality_multiindex = CellMetadata.make_multiindex_name(modality)
-        no_modality_info = df[modality_multiindex].isna() | df[modality_multiindex].str.len().eq(0)
+        no_modality_info = df[modality_multiindex].isna() | df[
+            modality_multiindex
+        ].str.len().eq(0)
         bool_name = modality + "_bool"
         bool_multiindex = CellMetadata.make_multiindex_name(bool_name)
         # store inverse of no_modality_info (ie. True = has modality info)
@@ -145,12 +149,12 @@ def hide_modality_metadatum(self):
             m_to_rename[bool_name] = has_m
         self.modality_urls = self.file.filter(m_to_hide, axis=1)
         self.file.drop(m_to_hide, axis=1, inplace=True)
-        self.file.rename(columns= m_to_rename, inplace=True)
+        self.file.rename(columns=m_to_rename, inplace=True)
         return
 
     def booleanize_modality_metadata(self):
         """Translate presence of modality data to boolean for BigQuery
-           If no modality data, self.files is unchanged
+        If no modality data, self.files is unchanged
         """
         if self.modalities is not None:
             df = copy.deepcopy(self.file)

ingest/cli_parser.py

@@ -274,6 +274,13 @@ def create_parser():
         help="Indicates that differential expression analysis should be invoked",
     )
 
+    parser_differential_expression.add_argument(
+        "--de-type",
+        default="rest",
+        choices=['rest', 'pairwise'],
+        help="Accepted values: 'pairwise' or 'rest' (default)",
+    )
+
     parser_differential_expression.add_argument(
         "--study-accession",
         required=True,
@@ -336,6 +343,17 @@ def create_parser():
         "--gene-file", help="Path to .genes.tsv file"
     )
 
+    # For pairwise analyses
+    parser_differential_expression.add_argument(
+        "--group1",
+        help="1st annotation label to use for pairwise DE analysis",
+    )
+
+    parser_differential_expression.add_argument(
+        "--group2",
+        help="2nd annotation label to use for pairwise DE analysis",
+    )
+
     parser_ingest_differential_expression = subparsers.add_parser(
         "ingest_differential_expression",
         help="Indicates author differential expression analysis processing",
@@ -377,40 +395,38 @@ def create_parser():
     parser_ingest_differential_expression.add_argument(
         "--differential-expression-file",
         required=True,
-        help="Path to DE file uploaded by author."
+        help="Path to DE file uploaded by author.",
     )
 
     parser_ingest_differential_expression.add_argument(
         "--gene-header",
         required=True,
-        help="Header used for gene names / symbols in DE file"
+        help="Header used for gene names / symbols in DE file",
     )
 
     parser_ingest_differential_expression.add_argument(
-        "--group-header",
-        required=True,
-        help="Header used for group in DE file"
+        "--group-header", required=True, help="Header used for group in DE file"
     )
 
     parser_ingest_differential_expression.add_argument(
         "--comparison-group-header",
         required=False,
         help=(
-            "Header used for comparison group in DE file.  " +
-            "For pairwise comparisons.  Can omit if DE file is in one-vs-rest-only format."
-        )
+            "Header used for comparison group in DE file.  "
+            + "For pairwise comparisons.  Can omit if DE file is in one-vs-rest-only format."
+        ),
     )
 
     parser_ingest_differential_expression.add_argument(
         "--size-metric",
         required=True,
-        help='Header used as size metric in DE file, e.g. "logfoldchanges", "avg_log2FC", etc.'
+        help='Header used as size metric in DE file, e.g. "logfoldchanges", "avg_log2FC", etc.',
     )
 
     parser_ingest_differential_expression.add_argument(
         "--significance-metric",
         required=True,
-        help='Header used as significance metric in DE file, e.g. "pvals_adj", "p_val_adj", etc.'
+        help='Header used as significance metric in DE file, e.g. "pvals_adj", "p_val_adj", etc.',
     )
 
     # AnnData subparsers
@@ -493,7 +509,7 @@ def create_parser():
     parser_rank_genes.add_argument(
         '--publication',
         help="URL of the study's publicly-accessible research article, or GS URL or local path to publication text file",
-        required=True
+        required=True,
     )
 
     return parser

ingest/clusters.py

@@ -136,9 +136,11 @@ def transform(self):
                 {
                     "name": annot_name,
                     "type": annot_type,
-                    "values": self.file[annot_headers].unique().tolist()
-                    if annot_type == "group"
-                    else [],
+                    "values": (
+                        self.file[annot_headers].unique().tolist()
+                        if annot_type == "group"
+                        else []
+                    ),
                 }
             )
         Annotations.dev_logger.info(f"Creating model for {self.study_id}")
@@ -148,9 +150,11 @@ def transform(self):
             cell_annotations=cell_annotations,
             study_file_id=self.study_file_id,
             study_id=self.study_id,
-            domain_ranges=DomainRanges(**self.domain_ranges)
-            if self.domain_ranges is not None
-            else None,
+            domain_ranges=(
+                DomainRanges(**self.domain_ranges)
+                if self.domain_ranges is not None
+                else None
+            ),
         )
 
     def get_data_array_annot(self, linear_data_id):