Add a script to filter mutations in a MAF file

Dockerfile

@@ -1,45 +1,57 @@
+# Stage 1: Build the Java application JAR
 FROM maven:3-eclipse-temurin-21 as jar_builder
 
 # Set the working directory in the Maven image
 WORKDIR /app
 
-# Copy the java source files and the pom.xml file into the image
+# Copy the Java source files and the pom.xml file into the image
 COPY src ./src
 COPY pom.xml .
 
 # Build the application
 RUN mvn clean package -DskipTests
 
+# Stage 2: Prepare the final image
 FROM maven:3-eclipse-temurin-21
 
-# download system dependencies first to take advantage of docker caching
-RUN apt-get update; apt-get install -y --no-install-recommends \
+# Download system dependencies first to take advantage of Docker caching
+RUN apt-get update && apt-get install -y --no-install-recommends \
         build-essential \
         default-mysql-client \
         default-libmysqlclient-dev \
         python3 \
-        python3-setuptools \
+        python3-venv \
         python3-dev \
-        python3-pip \
         unzip \
         perl \
-    && rm -rf /var/lib/apt/lists/* \
-    && pip3 install wheel
+    && rm -rf /var/lib/apt/lists/*
 
-# Install any needed packages specified in requirements.txt
+# Set up a Python virtual environment
+RUN python3 -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Install Python packages in the virtual environment
 COPY requirements.txt ./
 RUN pip install --no-cache-dir -r requirements.txt
 
-RUN ln -s $(which python3) /usr/local/bin/python || true
+# Link python3 to python for compatibility
+RUN ln -s /opt/venv/bin/python /usr/local/bin/python || true
 
+# Copy the built JAR from the first stage
 COPY --from=jar_builder /app/core-*.jar /
+
+# Copy and set permissions for scripts
 COPY scripts/ scripts/
 RUN chmod -R a+x /scripts/
 
 # Set the working directory in the container
 WORKDIR /scripts/
 
+# Environment variable
 ENV PORTAL_HOME=/
 
-# This file is empty. It has to be overriden by bind mounting the actual application.properties
+# Create an empty application.properties file
 RUN touch /application.properties
+
+# Entry command
+CMD ["python", "your_script.py"]

src/main/java/org/mskcc/cbio/portal/dao/DaoGeneOptimized.java

@@ -343,30 +343,6 @@ public List<CanonicalGene> guessGene(String geneId, String chr) {
         return Collections.emptyList();
     }
 
-
-    private static Map<String,String> validChrValues = null;
-    public static String normalizeChr(String strChr) {
-        if (strChr==null) {
-            return null;
-        }
-
-        if (validChrValues==null) {
-            validChrValues = new HashMap<String,String>();
-            for (int lc = 1; lc<=24; lc++) {
-                    validChrValues.put(Integer.toString(lc),Integer.toString(lc));
-                    validChrValues.put("CHR" + Integer.toString(lc),Integer.toString(lc));
-            }
-            validChrValues.put("X","23");
-            validChrValues.put("CHRX","23");
-            validChrValues.put("Y","24");
-            validChrValues.put("CHRY","24");
-            validChrValues.put("NA","NA");
-            validChrValues.put("MT","MT"); // mitochondria
-        }
-
-        return validChrValues.get(strChr);
-    }
-
     private static String getChrFromCytoband(String cytoband) {
         if (cytoband==null) {
             return null;

src/main/java/org/mskcc/cbio/portal/scripts/FilterMutationData.java

@@ -0,0 +1,142 @@
+/*
+ * Copyright (c) 2015 - 2022 Memorial Sloan Kettering Cancer Center.
+ *
+ * This library is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY, WITHOUT EVEN THE IMPLIED WARRANTY OF MERCHANTABILITY OR FITNESS
+ * FOR A PARTICULAR PURPOSE. The software and documentation provided hereunder
+ * is on an "as is" basis, and Memorial Sloan Kettering Cancer Center has no
+ * obligations to provide maintenance, support, updates, enhancements or
+ * modifications. In no event shall Memorial Sloan Kettering Cancer Center be
+ * liable to any party for direct, indirect, special, incidental or
+ * consequential damages, including lost profits, arising out of the use of this
+ * software and its documentation, even if Memorial Sloan Kettering Cancer
+ * Center has been advised of the possibility of such damage.
+ */
+
+/*
+ * This file is part of cBioPortal.
+ *
+ * cBioPortal is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+package org.mskcc.cbio.portal.scripts;
+
+import joptsimple.OptionException;
+import joptsimple.OptionParser;
+import joptsimple.OptionSet;
+import joptsimple.OptionSpec;
+import org.mskcc.cbio.maf.MafRecord;
+import org.mskcc.cbio.maf.MafUtil;
+import org.mskcc.cbio.portal.util.*;
+
+import java.io.*;
+import java.util.*;
+
+/**
+ * Read MAF records, filter records of interest and writes back to the file. The script backs up original file under {filename with extension}_backup.
+ *
+ * @author Ruslan Forostianov
+ */
+public class FilterMutationData extends ConsoleRunnable {
+
+    /**
+     * Instantiates a ConsoleRunnable to run with the given command line args.
+     *
+     * @param args the command line arguments to be used
+     * @see {@link #run()}
+     */
+    public FilterMutationData(String[] args) {
+        super(args);
+    }
+
+    public void run() {
+        String description = "Filter MAF file for records of interest and rewrites it with selected mutations.";
+        OptionParser parser = new OptionParser();
+        OptionSpec<String> data = parser.accepts( "data",
+                "MAF data file" ).withRequiredArg().describedAs( "data_mutations.txt" ).ofType( String.class );
+        OptionSpec<String> meta = parser.accepts( "meta",
+                "meta (description) file" ).withOptionalArg().describedAs( "meta_mutations.txt" ).ofType( String.class );
+
+        OptionSet options = null;
+        File originalMutationFile;
+        Set<String> namespaces = null;
+        Set<String> filteredMutations = null;
+
+        try {
+            options = parser.parse( args );
+            originalMutationFile = new File((String) options.valueOf("data"));
+            if (options.has("meta")) {
+                File descriptorFile = new File((String) options.valueOf( "meta" ) );
+                filteredMutations = GeneticProfileReader.getVariantClassificationFilter(descriptorFile);
+                namespaces = GeneticProfileReader.getNamespaces(descriptorFile);
+            }
+        } catch (OptionException e) {
+            throw new UsageException(
+                    this.getClass().getName(), description, parser,
+                    e.getMessage());
+        } catch (Exception e) {
+           throw new RuntimeException(e);
+        }
+        ProgressMonitor.setCurrentMessage("Start filtering mutation records in the MAF file ...");
+        File resultMutationFile = new File(originalMutationFile.getAbsolutePath() + "_filtered");
+        final MutationFilter mutationFilter = new MutationFilter();
+        try (
+                BufferedReader originalFileBufferedReader = new BufferedReader(new FileReader(originalMutationFile));
+                BufferedWriter resultFileBufferedWriter = new BufferedWriter(new FileWriter(resultMutationFile))
+        ) {
+            String line;
+            MafUtil mafUtil = null;
+            while ((line = originalFileBufferedReader.readLine()) != null) {
+                ProgressMonitor.incrementCurValue();
+                ConsoleUtil.showProgress();
+
+                if (TsvUtil.isDataLine(line)) {
+                    if (mafUtil == null) {
+                        mafUtil = new MafUtil(line, namespaces);
+                    } else {
+                        MafRecord record = mafUtil.parseRecord(line);
+                        if (!mutationFilter.acceptMutation(record, filteredMutations)) {
+                            continue;
+                        }
+                    }
+                }
+                resultFileBufferedWriter.write(line);
+                resultFileBufferedWriter.write(System.lineSeparator());
+            }
+        } catch (IOException e) {
+            e.printStackTrace();
+        }
+        File backupMutationFile = new File(originalMutationFile.getAbsolutePath() + "_backup");
+        if (originalMutationFile.renameTo(backupMutationFile)) {
+            ProgressMonitor.setCurrentMessage("The original file is backed up to:"
+                    + backupMutationFile.getAbsolutePath());
+            if (resultMutationFile.renameTo(originalMutationFile)) {
+                ProgressMonitor.setCurrentMessage("The MAF file has been overwritten with filtered records.");
+            } else {
+                throw new RuntimeException("Failed to rename the filtered MAF file ("
+                        + resultMutationFile.getAbsolutePath() + ") to the input MAF file ("
+                        + originalMutationFile.getAbsolutePath() + ").");
+            }
+        } else {
+            throw new RuntimeException("Failed to rename MAF file ("
+                    + originalMutationFile.getAbsolutePath() + ") for backup.");
+        }
+        ProgressMonitor.setCurrentMessage(mutationFilter.getStatistics());
+    }
+
+    public static void main(String[] args) {
+        ConsoleRunnable runner = new FilterMutationData(args);
+        runner.runInConsole();
+    }
+}