interactive plot USPTO, patch batching

README.md

@@ -40,6 +40,21 @@ To test the HCASE method, you can run the `embed_uspto.py` script located in the
 
 This script will execute the embedding process and provide the results.
 
+## Interactive Heatmap Plotting and Features
+
+The HCASE repository includes an interactive heatmap visualization tool for exploring the distribution of embedded chemical structures. This tool is implemented in `hcase/plot.py` and provides the following features:
+
+- **Automatic Deployment**: The Dash app for the heatmap visualization is automatically launched when running the USPTO embedding workflow (e.g., `examples/EmbedUSPTO/embed_uspto.py`). No separate launch step is required.
+- **Frequency Heatmap**: Visualizes the frequency of structures in each region of the embedded space, using log-scaled bins for better dynamic range.
+- **Color Scheme**: The heatmap uses a discrete, inverted color scale—light colors represent low frequency counts, and dark colors represent high frequency counts.
+- **Plot Area Border**: The plot area (data region) is outlined with a clear border, making the data region visually distinct. Axis ticks and labels are positioned just outside this border.
+- **Interactive Search**: Users can search for specific reference or target structures by SMILES or InChIKey, and matching points are highlighted on the heatmap.
+- **Structure Display**: Clicking on a heatmap cell displays the reference structure and a grid of corresponding target structures, with options to highlight substructure overlaps.
+- **Frequency Range Slider**: A slider allows users to filter the heatmap by frequency range, using log-scale bins.
+- **Legend**: A color legend explains the mapping between frequency bins and colors.
+
+To use the heatmap tool, simply run the USPTO embedding workflow as described above. The Dash app will be deployed automatically, enabling detailed exploration of the chemical embedding space and helping users identify regions of interest, structure clusters, and outliers.
+
 ## For Contributors
 
 We welcome contributions to the HCASE method! Here's how you can contribute:

docker/Dockerfile.hcase

@@ -1,9 +1,47 @@
-FROM python:3.12-slim-bookworm
+# Use a base image with CUDA and Python (adjust CUDA version as needed)
+FROM nvidia/cuda:12.2.0-devel-ubuntu22.04
 
-RUN apt-get update && apt-get install -y gcc 
+# Set environment variables to prevent some interactive prompts
+ENV DEBIAN_FRONTEND=noninteractive
 
-COPY ./requirements.txt requirements.txt
-RUN pip install -r requirements.txt
+# Install dependencies
+RUN apt-get update && apt-get install -y \
+    wget \
+    git \
+    curl \
+    ca-certificates \
+    build-essential \
+    python3 \
+    python3-pip \
+    python3-venv \
+    && rm -rf /var/lib/apt/lists/*
 
-RUN mkdir /app
-WORKDIR /app
+# Symlink python3 and pip3
+RUN ln -s /usr/bin/python3 /usr/bin/python && ln -s /usr/bin/pip3 /usr/bin/pip
+
+# Upgrade pip and install conda (Miniconda)
+ENV CONDA_DIR=/opt/conda
+RUN wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh && \
+    bash miniconda.sh -b -p $CONDA_DIR && \
+    rm miniconda.sh
+ENV PATH=$CONDA_DIR/bin:$PATH
+
+# IMPORTANT: Build context must be the repo root, not the docker/ folder!
+# Example: docker build -f docker/Dockerfile.hcase .
+# This will copy the entire repo into /workspace in the container.
+COPY . /workspace
+
+# Create and activate environment
+RUN conda env create -f /workspace/environment.yml && conda clean -a
+
+# Activate the environment by default
+SHELL ["conda", "run", "-n", "base", "/bin/bash", "-c"]
+
+# Install cupy using pip or conda as needed (adjust CUDA version)
+RUN pip install cupy-cuda12x
+
+# Set working directory inside container
+WORKDIR /workspace
+
+# Default command
+CMD ["python"]

docker/Singularity.def

@@ -0,0 +1,61 @@
+Bootstrap: docker
+From: nvidia/cuda:12.2.0-devel-ubuntu22.04
+
+%environment
+    export PATH=/opt/conda/bin:$PATH
+    export DEBIAN_FRONTEND=noninteractive
+    export WORKDIR=/workspace
+
+%post
+    apt-get update && apt-get install -y \
+        wget \
+        git \
+        curl \
+        ca-certificates \
+        build-essential \
+        python3 \
+        python3-pip \
+        python3-venv \
+        && rm -rf /var/lib/apt/lists/*
+
+    # Symlink python3 and pip3
+    ln -s /usr/bin/python3 /usr/bin/python || true
+    ln -s /usr/bin/pip3 /usr/bin/pip || true
+
+    # Install Miniconda
+    wget --quiet https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
+    bash miniconda.sh -b -p /opt/conda
+    rm miniconda.sh
+
+    # Create workspace directory
+    mkdir -p /workspace
+
+    # Create conda environment
+    /opt/conda/bin/conda env create -f /workspace/environment.yml
+    /opt/conda/bin/conda clean -a
+
+    # Install cupy (adjust CUDA version if needed)
+    /opt/conda/bin/pip install cupy-cuda12x
+
+%files
+    aux_code /workspace/aux_code
+    data /workspace/data
+    examples /workspace/examples
+    hcase /workspace/hcase
+    log /workspace/log
+    plots /workspace/plots
+    tests /workspace/tests
+    workflows /workspace/workflows
+    environment.yml /workspace/environment.yml
+    requirements.txt /workspace/requirements.txt
+    README.md /workspace/README.md
+    LICENSE /workspace/LICENSE
+    Makefile /workspace/Makefile
+    pyproject.toml /workspace/pyproject.toml
+    setup.py /workspace/setup.py
+    DISCLAIMER /workspace/DISCLAIMER
+    NOTES /workspace/NOTES
+
+%runscript
+    cd /workspace
+    exec /opt/conda/bin/python "$@"

environment.yml

@@ -3,14 +3,17 @@ channels:
   - conda-forge
   - anaconda
 dependencies:
-  - python==3.13.2
-  - cupy==13.4.0
-  - rdkit==2024.09.6
+  - python==3.11
+  - rdkit==2025.03.3
+  - cupy==13.4.1
   - pandarallel==1.6.5
-  - ipywidgets==8.1.5
-  - pip==25.0.1
-  - numpy==2.2.3
-  - tqdm==4.65.0
-  - pytest
+  - ipywidgets==8.1.7
+  - pip==25.1.1
+  - numpy==2.3.0
+  - tqdm==4.67.1
+  - pytest==8.4.0
   - pip:
-        - -e .
+      - -e .
+      - molplotly==1.1.7
+      - dash==2.11.1
+      - jupyter-dash==0.4.2

environment_cpu.yml

@@ -3,13 +3,16 @@ channels:
   - conda-forge
   - anaconda
 dependencies:
-  - python==3.13.2
-  - rdkit==2024.09.6
+  - python==3.11
+  - rdkit==2025.03.3
   - pandarallel==1.6.5
-  - ipywidgets==8.1.5
-  - pip==25.0.1
-  - numpy==2.2.3
-  - tqdm==4.65.0
-  - pytest
+  - ipywidgets==8.1.7
+  - pip==25.1.1
+  - numpy==2.3.0
+  - tqdm==4.67.1
+  - pytest==8.4.0
   - pip:
-        - -e .
+      - -e .
+      - molplotly==1.1.7
+      - dash==2.11.1
+      - jupyter-dash==0.4.2