feat: isolate visualization dependencies behind optional extra

README.md

@@ -21,6 +21,7 @@ Find paths between node A and node B in the Translator KG <br>
 Find a subnetwork given a list of nodes in the Translator KG <br>
 Developer-friendly wrappers for resolving labels/CURIEs, caching Translator resources, and returning parsed finder results <br>
 Connecting user's API with Translator API <br>
+*Note: Visualization capabilities (pyvis, matplotlib, seaborn) can be installed separately via the `vision` extra.*
 
 
 ## How to use TCT
@@ -34,8 +35,13 @@ pip install TCT
 # TCT is in development, to get the most recent update, user can install it through the github repo
 ```
 
-**This the recommended approach for installation.**
+**This is the recommended approach for a minimal installation.**
 
+Visualization support is optional. Install it with the `vision` extra when you need the plotting and graph-rendering utilities:
+
+```bash
+pip install "TCT[vision]"
+```
 
 #### Development Installation
 
@@ -57,6 +63,12 @@ cd Translator_component_toolkit
 uv sync
 ```
 
+To include visualization support in the UV environment:
+
+```bash
+uv sync --extra vision
+```
+
 #### Building and Deployment
 **Using pip:**
 - Build: `python -m build`

TCT/TCT.py

@@ -4,17 +4,12 @@
 import requests
 import json
 import pandas as pd
-import  seaborn as sns
-import matplotlib.pyplot as plt
-import networkx as nx
 import numpy as np
 #import openai
 from . import name_resolver, node_normalizer, translator_query
 
 # plt.switch_backend('module://ipykernel.pylab.backend_inline')
 
-from IPython.display import display
-
 __all__ = [
     'TCT_help',
     'list_functions',
@@ -851,6 +846,9 @@ def plot_heatmap(predicates_by_nodes_df,num_of_nodes = 20,
                                  fontsize = 6,
                                  title_fontsize = 10,
                                  output_png="NE_heatmap.png"):
+    import matplotlib.pyplot as plt
+    import seaborn as sns
+
     #matplotlib.use('Agg')
 
     #title = "Ranking of one-hop nodes by primary infores"
@@ -891,7 +889,8 @@ def plot_heatmap_ui(predicates_by_nodes_df,num_of_nodes = 20,
                                  fontsize = 6,
                                  title_fontsize = 10,
                                  output_png="NE_heatmap.png"):
-
+    import matplotlib.pyplot as plt
+    import seaborn as sns
 
     title = "Ranking of one-hop nodes by primary infores"
     ylab = "infores"
@@ -1658,7 +1657,8 @@ def merge_by_ranking_index(result_ranked_by_primary_infores,
                            title_fontsize = 12,
                            fontsize = 12,
                            ):
-
+    import matplotlib.pyplot as plt
+    import seaborn as sns
 
     dic_rank1 = {}
     for i in range(0, result_ranked_by_primary_infores.shape[0]):
@@ -1774,6 +1774,9 @@ def plot_path_bar(x,
                     fontsize = 8,
                     title_fontsize = 10,
                     output_png="NE_heatmap.png"):
+    import matplotlib.pyplot as plt
+    import seaborn as sns
+
     #matplotlib.use('Agg')
 
     # title = "Bridging nodes"  # Unused variable
@@ -1965,6 +1968,9 @@ def select_result_to_analysis(sele_genes,Temp_result_df1, Temp_result_df2 ):
 
 
 def plot_graph_by_predicates(for_plot):
+    import networkx as nx
+    from IPython.display import display
+
     graph = nx.from_pandas_edgelist(for_plot,
                                 source='Subject',
                                 target='Object',
@@ -2013,6 +2019,8 @@ def plot_graph_by_predicates(for_plot):
 
 
 def plot_graph_by_infores(for_plot):
+    import networkx as nx
+    from IPython.display import display
 
     graph = nx.from_pandas_edgelist(for_plot,
                                     source='Subject',
@@ -2062,6 +2070,8 @@ def plot_graph_by_infores(for_plot):
 
 
 def plot_graph_by_API(for_plot):
+    import networkx as nx
+    from IPython.display import display
 
     graph = nx.from_pandas_edgelist(for_plot,
                                     source='Subject',
@@ -2273,6 +2283,9 @@ def load_translator_resources():
 
 
 def visulize_path(input_node1_id, intermediate_node, input_node3_id, result, result2):
+    import networkx as nx
+    from IPython.display import display
+
     forplot_subject = []
     forplot_object = []
     forplot_predicate = []

TCT/TCT_Visualization.py

@@ -1,9 +1,6 @@
 
 from .node_normalizer import ID_convert_to_preferred_name_nodeNormalizer
 
-import networkx as nx
-from pyvis.network import Network
-
 def visualize_neighborhood_graph(result, show_label=True, height="1000px", width="100%", output_filename_prefix=None):
     '''Visualize the neighborhood graph using pyvis
     Args:
@@ -18,6 +15,9 @@ def visualize_neighborhood_graph(result, show_label=True, height="1000px", width
         dic_graph = visualize_neighborhood_graph(result, show_label=True, height="500", width="100%")
     '''
 
+    import networkx as nx
+    from pyvis.network import Network
+
     # Your JSON (as Python dict)
     data = result
     IDs = []

docs/source/intro.md

@@ -12,6 +12,7 @@ Allowing testing whether a user defined API follows a [TRAPI](https://github.com
 Faciliting to explore knowledge graphs from both Translator ecosystem and user defined APIs.<br>
 Developer-friendly `pathfinder` and `neighborhood_finder` wrappers for resolving labels/CURIEs, caching Translator resources, and returning parsed finder results (`from TCT import pathfinder, neighborhood_finder`). <br>
 Connecting large language models to convert user's questions into TRAPI queries. <br>
+*Note: Visualization capabilities (pyvis, matplotlib, seaborn) can be installed via the `vision` extra using `pip install TCT[vision]`*
 
 ### Contributing
 TCT is a tool that helps to explore knowledge graphs developed in the Biomedical Data Translator Consortium. Consortium members and external contributors are encouraged to submit issues and pull requests. 
@@ -22,7 +23,17 @@ Guangrong Qin, guangrong.qin@isbscience.org
 ## How to use TCT
 ### Install Requirements
 
-To install TCT as a python library, you can install the library using `pip install TCT` from the command line. 
+To install the minimal TCT package, run:
+
+```bash
+pip install TCT
+```
+
+Visualization support is optional. Install it with the `vision` extra when you need the plotting and graph-rendering utilities:
+
+```bash
+pip install "TCT[vision]"
+```
 
 The TCT is continuously updated, if you would like to use the latest functions, you can also  clone this most recent github repository (https://github.com/NCATSTranslator/Translator_component_toolkit/tree/main), 
 `git clone https://github.com/NCATSTranslator/Translator_component_toolkit.git`

pyproject.toml

@@ -24,20 +24,21 @@ dependencies = [
     "requests",
     "jsons",
     "pandas",
-    "seaborn",
-    "matplotlib",
-    "ipycytoscape",
-    "networkx",
     "numpy",
     "openai",
     "ipykernel",
-    'networkx',
-    'igraph',
-    'pyvis',
-    'zstandard',
+    "igraph",
+    "zstandard",
 ]
 
 [project.optional-dependencies]
+vision = [
+    "ipycytoscape",
+    "matplotlib",
+    "networkx",
+    "pyvis",
+    "seaborn",
+]
 mcp = [
     "fastmcp>=2.12.2",
     "click>=8.2.1",