Update dev with changes in main

docs/src/api.rst

@@ -195,6 +195,20 @@ NCI60
     data.omics.nci60_datatypes
     data.omics.nci60_table
 
+.. _api-phosphoegf:
+
+Phospho-EGF meta-analysis
+~~~~~
+.. module::networkcommons.data.omics
+.. currentmodule:: networkcommons
+
+.. autosummary::
+    :toctree: api
+    :recursive:
+
+    data.omics.phospho_egf_datatypes
+    data.omics.phospho_egf_tables
+
 .. _api-eval:
 
 Evaluation and description

docs/src/datasets.rst

@@ -92,6 +92,25 @@ NCI60
 
 .. _details-pk:
 
+
+Phosphoproteomics in response to EGF
+-----
+
+**Alias:** PhosphoEGF
+
+**Description:** A meta-analysis of phosphoproteomics data in response to EGF stimulation
+
+**Publication Link:** `Garrido-Rodriguez et al. Evaluating signaling pathway inference from kinase-substrate interactions and phosphoproteomics data. bioRxiv (2024). <https://www.biorxiv.org/content/10.1101/2024.10.21.619348v1>`_
+
+**Data location:** `Supplementary Data files of the manuscript <https://www.biorxiv.org/content/10.1101/2024.10.21.619348v1.supplementary-material>`_
+
+**Detailed Description:** This dataset the results of a meta-analysis of phosphoproteomics data in response to EGF stimulation across different labs and stimulation times. The data is available at two different levels. First, the phosphosite differential abundance is provided for every combination of study and treatment time. In the table, 'This study' refers to the data generated in the manuscript. Second, we offer access to the kinase-level activities inerred using decoupleR and the different kinase-substrate networks described in the paper. Briefly, four different networks were employed: A first one based on literature (literature), one based on kinase-substrate interaction prediction via protein language models (phosformer), one based on positionl peptide array screening (kinlibrary) and a combination of all of them (combined).
+
+**Functions:** See API documentation for :ref:`Phospho-EGF meta-analysis<api-phosphoegf>`.
+
+.. _details-pk:
+
+
 ---------------
 Prior Knowledge
 ---------------

networkcommons/data/datasets.yaml

@@ -47,3 +47,9 @@ omics:
         It includes three files: TF activities from transcriptomics data,
         metabolite abundances and gene reads.
       path: NCI60/{cell_line}/{cell_line}__{data_type}.tsv
+    phosphoegf:
+      name: PhosphoEGF
+      description: Phosphoproteomics meta-analysis of the response to EGF stimulus
+      publication_link: https://www.biorxiv.org/content/10.1101/2024.10.21.619348v1
+      detailed_description: >-
+        This dataset contains phosphoproteomics data after EGF stimulus gathered and preprocessed from multiple studies.

networkcommons/data/omics/__init__.py

@@ -24,3 +24,4 @@
 from ._scperturb import *
 from ._nci60 import *
 from ._cptac import *
+from ._phosphoegf import *

networkcommons/data/omics/_phosphoegf.py

@@ -0,0 +1,58 @@
+#!/usr/bin/env python
+
+#
+# This file is part of the `networkcommons` Python module
+#
+# Copyright 2024
+# Heidelberg University Hospital
+#
+# File author(s): Saez Lab ([email protected])
+#
+# Distributed under the GPLv3 license
+# See the file `LICENSE` or read a copy at
+# https://www.gnu.org/licenses/gpl-3.0.txt
+#
+
+"""
+Meta-analysis of phosphoproteomics response to EGF stimulus.
+"""
+
+import pandas as pd
+import warnings
+
+def phospho_egf_datatypes() -> pd.DataFrame:
+    """
+    Table describing the available data types in the Phospho EGF dataset.
+
+    Returns:
+        DataFrame with all data types.
+    """
+
+    return pd.DataFrame({
+        'type': ['phosphosite', 'kinase'],
+        'description': ['Differential phosphoproteomics at the site level for all studies in the meta-analysis',
+                        'Kinase activities obtained using each of the kinase-substrate prior knowledge resources'],
+    })
+
+
+def phospho_egf_tables(type='diffabundance'):
+    """
+    A table with the corresponding data type for the phospho EGF dataset. 
+
+    Args:
+        type:
+            Either 'diffabundance' or 'kinase_scores'.
+
+    Returns:
+        A DataFrame with the corresponding data.
+    """
+
+    if type == 'phosphosite':
+        out_table = pd.read_csv('https://www.biorxiv.org/content/biorxiv/early/2024/10/22/2024.10.21.619348/DC3/embed/media-3.gz', compression='gzip', low_memory=False)
+    elif type == 'kinase':
+        out_table = pd.read_csv('https://www.biorxiv.org/content/biorxiv/early/2024/10/22/2024.10.21.619348/DC4/embed/media-4.gz', compression='gzip', low_memory=False)
+    else:
+        warnings.warn(f'Unknown data type "{type}"')
+        return None
+    return out_table
+