Webapp data and scripts

8.shiny-app/scripts/0.single-dependency-umaps.py

@@ -0,0 +1,188 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# In[1]:
+
+
+import pandas as pd
+import pathlib
+import sys
+import joblib
+import numpy as np
+
+script_directory = pathlib.Path("../utils/").resolve()
+sys.path.insert(0, str(script_directory))
+from data_loader import load_model_data, load_train_test_data
+
+
+# In[2]:
+
+
+# Load dependency data
+data_directory = pathlib.Path("../0.data-download/data").resolve()
+dependency_file = pathlib.Path(f"{data_directory}/CRISPRGeneEffect.parquet").resolve()
+gene_dict_file = pathlib.Path(f"{data_directory}/CRISPR_gene_dictionary.parquet").resolve()
+dependency_df, gene_dict_df= load_model_data(dependency_file, gene_dict_file)
+dependency_df = dependency_df.set_index("ModelID")
+
+
+# In[3]:
+
+
+cancer_type_input_file = pathlib.Path("../0.data-download/data/Model.parquet")
+cancer_type_df = pd.read_parquet(cancer_type_input_file)
+
+
+# In[4]:
+
+
+combined_df = dependency_df.merge(
+    cancer_type_df[["ModelID", "OncotreePrimaryDisease"]], 
+    on="ModelID", 
+    how="left"
+)
+
+
+# In[5]:
+
+
+combined_df.head()
+
+
+# In[6]:
+
+
+gene_cols = combined_df.columns.drop("ModelID")
+gene_cols = gene_cols.drop("OncotreePrimaryDisease")
+
+
+# In[7]:
+
+
+features = combined_df[gene_cols]
+
+
+# In[8]:
+
+
+features
+
+
+# In[9]:
+
+
+import random
+import colorsys
+
+def generate_random_palette(num_colors, seed=12):
+    # Generate random colors with varied lightness and saturation
+    random.seed(seed)
+
+    colors = []
+
+    for _ in range(num_colors):
+        h = random.random()  # Random hue between 0 and 1
+        l = random.uniform(0.2, 0.8)  # Lightness between 0.3 and 0.9 for contrast
+        s = random.uniform(0.5, 1.0)  # Saturation between 0.6 and 1.0 for vivid colors
+        color = colorsys.hls_to_rgb(h, l, s)  # Convert HLS to RGB color
+        colors.append(color)
+
+    return colors
+
+
+# In[10]:
+
+
+import plotly.graph_objects as go
+import plotly.express as px
+from sklearn.decomposition import PCA
+from sklearn.preprocessing import StandardScaler
+
+# Prepare PCA input (excluding non-numeric columns)
+pca_input = combined_df[gene_cols].apply(pd.to_numeric, errors="coerce")
+pca = PCA(n_components=2, random_state=0)
+pca_embedding = pca.fit_transform(pca_input)
+
+# Add PCA components to the dataframe
+combined_df["PCA1"] = pca_embedding[:, 0]
+combined_df["PCA2"] = pca_embedding[:, 1]
+
+# Prepare color map for each cancer type
+cancer_types = combined_df["OncotreePrimaryDisease"].unique()
+color_map = px.colors.qualitative.Plotly + px.colors.qualitative.Light24 + px.colors.qualitative.Dark24
+highlight_color_map = {cancer: color_map[i % len(color_map)] for i, cancer in enumerate(cancer_types)}
+
+# Create one trace per cancer type
+traces = []
+for cancer in cancer_types:
+    df_subset = combined_df[combined_df["OncotreePrimaryDisease"] == cancer]
+    trace = go.Scatter(
+        x=df_subset["PCA1"],
+        y=df_subset["PCA2"],
+        mode='markers',
+        name=cancer,
+        marker=dict(size=7),
+        text=[f"{cancer} | {model_id}" for model_id in df_subset["ModelID"]],
+        hoverinfo='text',
+    )
+    traces.append(trace)
+
+# Create dropdown buttons
+dropdown_buttons = []
+
+for i, cancer in enumerate(cancer_types):
+    # Make all grey except the selected one
+    visibility = [True] * len(cancer_types)
+    colors = ['lightgrey'] * len(cancer_types)
+    colors[i] = highlight_color_map[cancer]
+
+    button = dict(
+        method="update",
+        label=cancer,
+        args=[
+            {"visible": visibility,
+             "marker": [{'color': colors[j]} for j in range(len(cancer_types))]},
+            {"title": f"PCA Highlighted: {cancer}"}
+        ]
+    )
+    dropdown_buttons.append(button)
+
+# Add a button to show all with normal colors
+default_colors = [highlight_color_map[cancer] for cancer in cancer_types]
+dropdown_buttons.insert(0, dict(
+    method="update",
+    label="Show All",
+    args=[
+        {"visible": [True]*len(cancer_types),
+         "marker": [{'color': default_colors[j]} for j in range(len(cancer_types))]},
+        {"title": "PCA of DepMap Single Gene Data"}
+    ]
+))
+
+# Create the figure
+fig = go.Figure(data=traces)
+
+# Set default colors
+for i, trace in enumerate(fig.data):
+    trace.marker.color = default_colors[i]
+
+fig.update_layout(
+    title="PCA of DepMap Single Gene Data",
+    xaxis_title="PCA1",
+    yaxis_title="PCA2",
+    updatemenus=[{
+        "buttons": dropdown_buttons,
+        "direction": "down",
+        "showactive": True,
+        "x": 1.15,
+        "xanchor": "left",
+        "y": 1.15,
+        "yanchor": "top"
+    }],
+    width=1500,
+    height=800
+)
+
+fig.show()
+fig.write_html("achilles_pca.html")
+