Optimized compare_ref function from loop logic to dataframe logic

bamsurgeon.def

@@ -0,0 +1,44 @@
+Bootstrap: docker
+From: ubuntu:20.04
+
+# This definition file will allow to create a Singularity image, similar to Dockerfile
+
+%environment
+    export BAMSURGEON_PICARD_JAR=/picard.jar
+    export PATH=/app/bin:$PATH  # Force /app/bin to stay within the container
+
+%post
+    # To avoids interactive prompts during build)
+    export DEBIAN_FRONTEND=noninteractive
+
+    # Preconfigure tzdata
+    echo "tzdata tzdata/Areas select US" | debconf-set-selections
+    echo "tzdata tzdata/Zones/US select Eastern" | debconf-set-selections
+
+    # Install dependencies
+    apt-get update && apt-get upgrade -y
+    apt-get install -y python3 python3-pip git wget build-essential libz-dev \
+        libglib2.0-dev libbz2-dev liblzma-dev default-jre autoconf samtools bwa
+
+    mkdir -p /app/bin  # Create a custom directory for binaries
+    wget https://github.com/dzerbino/velvet/archive/refs/tags/v1.2.10.tar.gz
+    tar -xvzf v1.2.10.tar.gz
+    cd velvet-1.2.10
+    make
+    cp velvetg /app/bin
+    cp velveth /app/bin
+
+    git clone https://github.com/adamewing/exonerate.git
+    cd exonerate
+    autoreconf -fi
+    ./configure
+    make
+    make install
+
+    wget https://github.com/broadinstitute/picard/releases/download/2.27.3/picard.jar -O /picard.jar
+
+    # Install python dependencies : pysam, pandas
+    pip install pysam pandas
+
+    # Bamsurgeon
+    git clone https://github.com/adamewing/bamsurgeon.git

bin/bamsurgeon/replace_reads.py

@@ -7,7 +7,7 @@
 from collections import defaultdict
 
 from bamsurgeon.common import rc
-
+import pandas as pd
 import logging
 FORMAT = '%(levelname)s %(asctime)s %(message)s'
 logging.basicConfig(format=FORMAT)
@@ -68,16 +68,35 @@ def get_excluded_reads(file):
             ex.add(line.strip())
     return ex
 
+# Changed from loop logic to dataframe logic
+# def compare_ref(targetbam, donorbam):
+#     ''' if targetbam and donorbam are aligned to different references 
+#         and the references are in a different order it's a problem
+#     '''
+#     for ref in targetbam.references:
+#         if ref not in donorbam.references or donorbam.gettid(ref) != targetbam.gettid(ref):
+#             logger.error("contig mismatch: %s\n" % ref)
+#             return False
+#     return True
+
+## Modified on 05/22/2024: changed loop logic to dataframe logic
 def compare_ref(targetbam, donorbam):
     ''' if targetbam and donorbam are aligned to different references 
         and the references are in a different order it's a problem
     '''
-    for ref in targetbam.references:
-        if ref not in donorbam.references or donorbam.gettid(ref) != targetbam.gettid(ref):
-            logger.error("contig mismatch: %s\n" % ref)
-            return False
-    return True
+    donor_df = pd.DataFrame({"donorbam": donorbam.references})
+    target_df = pd.DataFrame({"targetbam": targetbam.references})
+
+    mismatched_refs = donor_df[~target_df.isin(donor_df)].dropna()
 
+    if len(mismatched_refs):
+        for i in range(len(mismatched_refs)):
+            logger.error("contig mismatch: %s \n"%mismatched_refs.loc[i, "donorbam"])
+            return False
+    else:
+        return True
+
+
 
 def replace_reads(origbamfile, mutbamfile, outbamfile, fasta_ref, nameprefix=None, excludefile=None, allreads=False, keepqual=False, progress=False, keepsecondary=False, keepsupplementary=False, seed=None, quiet=False):
     ''' outputbam must be writeable and use targetbam as template