changes to build liverpdo dataset

Author: alexandriai168

build/build_dataset.py

@@ -48,7 +48,8 @@ def process_docker(dataset,validate):
         'genes': ['genes'],
         'upload': ['upload'],
         'crcpdo': ['crcpdo'], 
-        'bladderpdo': ['bladderpdo']
+        'bladderpdo': ['bladderpdo'],
+        'liverpdo': ['liverpdo']
     }
 
     # Collect container names to build based on the dataset provided. Always build 'genes'.
@@ -131,7 +132,8 @@ def process_omics(executor, dataset, should_continue):
         'sarcpdo': ['mutations', 'transcriptomics'],
         'pancpdo': ['transcriptomics'],
         'bladderpdo': ['copy_number', 'mutations', 'transcriptomics'],
-        'crcpdo':['copy_number', 'mutations', 'transcriptomics']
+        'crcpdo':['copy_number', 'mutations', 'transcriptomics'],
+        'liverpdo':['copy_number', 'mutations', 'transcriptomics']
     }
 
     expected_omics = dataset_omics_files.get(dataset, [])

build/docker/Dockerfile.liverpdo

@@ -0,0 +1,61 @@
+FROM r-base:4.4.1
+
+ENV DEBIAN_FRONTEND=noninteractive
+
+# Update package list and install required packages
+RUN apt-get update && \
+    apt-get install -y build-essential wget curl libcurl4-openssl-dev libxml2-dev \
+    zlib1g-dev libssl-dev libbz2-dev libreadline-dev libsqlite3-dev libffi-dev
+
+# Download and compile Python 3.10 with shared library support
+RUN wget https://www.python.org/ftp/python/3.10.12/Python-3.10.12.tgz && \
+    tar -xf Python-3.10.12.tgz && \
+    cd Python-3.10.12 && \
+    ./configure --enable-optimizations --enable-shared && \
+    make -j$(nproc) && \
+    make altinstall && \
+    cd .. && \
+    rm -rf Python-3.10.12.tgz Python-3.10.12
+
+# Set Python 3.10 as default
+RUN ln -s /usr/local/bin/python3.10 /usr/bin/python3 && \
+    ln -s /usr/local/bin/pip3.10 /usr/bin/pip3
+
+# Update library paths for Python shared library
+RUN echo "/usr/local/lib" >> /etc/ld.so.conf.d/python3.10.conf && ldconfig
+
+# Create a Python virtual environment
+#RUN python3 -m venv /opt/venv
+#RUN /opt/venv/bin/pip install --upgrade pip
+
+# Set environment variables for reticulate
+#ENV RETICULATE_PYTHON="/opt/venv/bin/python3"
+ENV PYTHONPATH=/app#"${PYTHONPATH}:/app"
+WORKDIR /app
+
+# Set MPLCONFIGDIR to a writable directory and create it.
+ENV MPLCONFIGDIR=/app/tmp/matplotlib
+RUN mkdir -p /app/tmp/matplotlib
+
+
+ADD build/liverpdo/requirements.R .
+# installing r libraries
+RUN Rscript requirements.R
+
+
+
+# installing python libraries
+ADD build/liverpdo/requirements.txt .
+#RUN /opt/venv/bin/pip3 install -r requirements.txt
+RUN pip3 install -r requirements.txt
+
+RUN python3 --version
+RUN which Rscript
+
+#ENV PATH="/opt/venv/bin:$PATH"
+
+ADD build/liverpdo/CNV-segfile-annotation.R ./
+ADD build/liverpdo/*py ./
+ADD build/liverpdo/*sh ./
+
+ADD build/utils/* ./

build/docker/docker-compose.yml

@@ -112,4 +112,13 @@ services:
       args:
         HTTPS_PROXY: ${HTTPS_PROXY}
     platform: linux/amd64
-    image: crcpdo:latest
+    image: crcpdo:latest
+
+  liverpdo:
+    build:
+      context: ../../
+      dockerfile: build/docker/Dockerfile.liverpdo
+      args:
+        HTTPS_PROXY: ${HTTPS_PROXY}
+    platform: linux/amd64
+    image: liverpdo:latest

build/liverpdo/requirements.txt

@@ -0,0 +1,19 @@
+pandas
+numpy==1.23
+#os
+#gzip
+requests
+argparse
+synapseclient 
+matplotlib
+#math
+#time
+#threading
+#signal
+rdkit
+mordred
+#multiprocessing
+tqdm
+#itertools
+scikit-learn
+openpyxl