oxymetag 1.0.0__py3-none-any.whl → 1.1.0__py3-none-any.whl

oxymetag/__init__.py

@@ -2,7 +2,7 @@
 OxyMetaG: Oxygen metabolism profiling from metagenomic data
 """
 
-__version__ = "1.0.0"
+__version__ = "1.1.0"
 __author__ = "Clifton P. Bueno de Mesquita"
 __email__ = "cliff.buenodemesquita@colorado.edu"
 

oxymetag/cli.py

@@ -11,7 +11,6 @@ from . import __version__
 from .core import extract_reads, profile_samples, predict_aerobes
 from .utils import check_dependencies, run_kraken2_setup, OxyMetaGError
 
-# Set up logging
 logging.basicConfig(
     level=logging.INFO,
     format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
@@ -29,10 +28,8 @@ def main():
     
     subparsers = parser.add_subparsers(dest='command', help='Available commands')
     
-    # Setup command
     setup_parser = subparsers.add_parser('setup', help='Setup Kraken2 database')
     
-    # Extract command
     extract_parser = subparsers.add_parser('extract', help='Extract bacterial reads')
     extract_parser.add_argument('-i', '--input', nargs='+', required=True,
                                help='Input fastq.gz files')
@@ -43,27 +40,31 @@ def main():
     extract_parser.add_argument('--kraken-db', default='kraken2_db',
                                help='Kraken2 database path (default: kraken2_db)')
     
-    # Profile command  
-    profile_parser = subparsers.add_parser('profile', help='Profile samples with DIAMOND')
+    profile_parser = subparsers.add_parser('profile', help='Profile samples with DIAMOND or MMseqs2')
     profile_parser.add_argument('-i', '--input', default='BactReads',
                                help='Input directory (default: BactReads)')
-    profile_parser.add_argument('-o', '--output', default='diamond_output',
-                               help='Output directory (default: diamond_output)')
+    profile_parser.add_argument('-o', '--output', default=None,
+                               help='Output directory (default: diamond_output or mmseqs_output)')
     profile_parser.add_argument('-t', '--threads', type=int, default=4,
                                help='Number of threads (default: 4)')
+    profile_parser.add_argument('-m', '--method', choices=['diamond', 'mmseqs2'],
+                               default='diamond',
+                               help='Profiling method (default: diamond)')
     profile_parser.add_argument('--diamond-db',
                                help='DIAMOND database path (default: package database)')
+    profile_parser.add_argument('--mmseqs-db',
+                               help='MMseqs2 database path (default: package database)')
     
-    # Predict command
     predict_parser = subparsers.add_parser('predict', help='Predict aerobe levels')
-    predict_parser.add_argument('-i', '--input', default='diamond_output',
-                               help='Input directory (default: diamond_output)')
+    predict_parser.add_argument('-i', '--input', default=None,
+                               help='Input directory (default: diamond_output for modern, mmseqs_output for ancient)')
     predict_parser.add_argument('-o', '--output', default='per_aerobe_predictions.tsv',
                                help='Output file (default: per_aerobe_predictions.tsv)')
     predict_parser.add_argument('-t', '--threads', type=int, default=4,
                                help='Number of threads (default: 4)')
     predict_parser.add_argument('-m', '--mode', choices=['modern', 'ancient', 'custom'],
-                               default='modern', help='Filtering mode (default: modern)')
+                               default='modern', 
+                               help='Filtering mode: modern=DIAMOND, ancient=MMseqs2 (default: modern)')
     predict_parser.add_argument('--idcut', type=float,
                                help='Custom identity cutoff (for custom mode)')
     predict_parser.add_argument('--bitcut', type=float,
@@ -87,11 +88,12 @@ def main():
             extract_reads(args.input, args.output, args.threads, args.kraken_db)
             
         elif args.command == 'profile':
-            profile_samples(args.input, args.output, args.threads, args.diamond_db)
+            profile_samples(args.input, args.output, args.threads, args.method,
+                          args.diamond_db, args.mmseqs_db)
             
         elif args.command == 'predict':
             predict_aerobes(args.input, args.output, args.mode,
-                           args.idcut, args.bitcut, args.ecut, args.threads)
+                          args.idcut, args.bitcut, args.ecut, args.threads)
         
         logger.info("Command completed successfully")
         

oxymetag/core.py

@@ -108,20 +108,18 @@ def extract_reads(input_files: List[str], output_dir: str = "BactReads",
             continue
 
 
-def profile_samples(input_dir: str = "BactReads", output_dir: str = "diamond_output", 
-                   threads: int = 4, diamond_db: str = None):
+def profile_samples(input_dir: str = "BactReads", output_dir: str = None, 
+                   threads: int = 4, method: str = "diamond",
+                   diamond_db: str = None, mmseqs_db: str = None):
     """
-    Profile samples using DIAMOND blastx against Pfam database
+    Profile samples using DIAMOND or MMseqs2 against Pfam database
     """
     from .utils import get_package_data_path
     
-    logger.info(f"Starting sample profiling with {threads} threads")
+    logger.info(f"Starting sample profiling with {method} using {threads} threads")
     
-    if diamond_db is None:
-        diamond_db = get_package_data_path("oxymetag_pfams.dmnd")
-    
-    if not Path(diamond_db).exists():
-        raise OxyMetaGError(f"DIAMOND database not found: {diamond_db}")
+    if output_dir is None:
+        output_dir = 'diamond_output' if method == 'diamond' else 'mmseqs_output'
     
     output_path = Path(output_dir)
     output_path.mkdir(exist_ok=True)
@@ -132,6 +130,8 @@ def profile_samples(input_dir: str = "BactReads", output_dir: str = "diamond_out
     patterns = [
         '*_R1_bacterial.fastq.gz',
         '*_1_bacterial.fastq.gz',
+        '*_bacterial_R1.fastq.gz',
+        '*_bacterial_1.fastq.gz',
         '*_bacterial.fastq.gz'
     ]
     
@@ -147,9 +147,32 @@ def profile_samples(input_dir: str = "BactReads", output_dir: str = "diamond_out
         logger.error(f"FASTQ files in {input_dir}: {[f.name for f in all_files[:5]]}")
         raise OxyMetaGError(f"No bacterial read files found in {input_dir}")
     
+    if method == 'diamond':
+        _profile_with_diamond(input_files, output_path, threads, diamond_db)
+    elif method == 'mmseqs2':
+        _profile_with_mmseqs(input_files, output_path, threads, mmseqs_db)
+    else:
+        raise OxyMetaGError(f"Unknown method: {method}")
+
+
+def _profile_with_diamond(input_files: List[Path], output_path: Path, 
+                         threads: int, diamond_db: str = None):
+    """Profile samples using DIAMOND blastx"""
+    from .utils import get_package_data_path
+    
+    if diamond_db is None:
+        diamond_db = get_package_data_path("oxymetag_pfams.dmnd")
+    
+    if not Path(diamond_db).exists():
+        raise OxyMetaGError(f"DIAMOND database not found: {diamond_db}")
+    
     for input_file in input_files:
         base_name = input_file.stem.replace('.fastq', '').replace('.gz', '')
-        base_name = base_name.replace('_R1_bacterial', '').replace('_1_bacterial', '').replace('_bacterial', '')
+        base_name = (base_name.replace('_R1_bacterial', '')
+                             .replace('_1_bacterial', '')
+                             .replace('_bacterial_R1', '')
+                             .replace('_bacterial_1', '')
+                             .replace('_bacterial', ''))
         
         logger.info(f"Processing {input_file}")
         
@@ -172,20 +195,95 @@ def profile_samples(input_dir: str = "BactReads", output_dir: str = "diamond_out
             continue
 
 
-def predict_aerobes(input_dir: str = "diamond_output", output_file: str = "per_aerobe_predictions.tsv",
+def _profile_with_mmseqs(input_files: List[Path], output_path: Path,
+                        threads: int, mmseqs_db: str = None):
+    """Profile samples using MMseqs2 easy-search"""
+    from .utils import get_package_data_path
+    
+    if mmseqs_db is None:
+        mmseqs_db = get_package_data_path("oxymetag_pfams_n117_db")
+    
+    if not Path(mmseqs_db).exists():
+        raise OxyMetaGError(f"MMseqs2 database not found: {mmseqs_db}")
+    
+    data_dir = Path(get_package_data_path(""))
+    vtml_matrix = data_dir / "VTML20.out"
+    nucl_matrix = data_dir / "nucleotide.out"
+    
+    if not vtml_matrix.exists():
+        raise OxyMetaGError(f"VTML20.out matrix not found: {vtml_matrix}")
+    if not nucl_matrix.exists():
+        raise OxyMetaGError(f"nucleotide.out matrix not found: {nucl_matrix}")
+    
+    for input_file in input_files:
+        base_name = input_file.stem.replace('.fastq', '').replace('.gz', '')
+        base_name = (base_name.replace('_R1_bacterial', '')
+                             .replace('_1_bacterial', '')
+                             .replace('_bacterial_R1', '')
+                             .replace('_bacterial_1', '')
+                             .replace('_bacterial', ''))
+        
+        logger.info(f"Processing {input_file} with MMseqs2")
+        
+        output_file = output_path / f"{base_name}_mmseqs.tsv"
+        tmp_dir = output_path / f"{base_name}_tmp"
+        tmp_dir.mkdir(exist_ok=True)
+        
+        cmd = [
+            'mmseqs', 'easy-search',
+            str(input_file),
+            str(mmseqs_db),
+            str(output_file),
+            str(tmp_dir),
+            '--min-length', '12',
+            '-e', '10.0',
+            '--min-seq-id', '0.86',
+            '-c', '0.65',
+            '--cov-mode', '2',
+            '--format-mode', '0',
+            '--format-output', 'query,target,fident,alnlen,mismatch,gapopen,qstart,qend,tstart,tend,evalue,bits,qlen,tlen,cigar,qaln,taln',
+            '--comp-bias-corr', '0',
+            '--mask', '0',
+            '--exact-kmer-matching', '1',
+            '--sub-mat', f'aa:{vtml_matrix},nucl:{nucl_matrix}',
+            '--seed-sub-mat', f'aa:{vtml_matrix},nucl:{nucl_matrix}',
+            '-s', '2',
+            '-k', '6',
+            '--spaced-kmer-pattern', '11011101',
+            '--max-seqs', '10000',
+            '--max-rejected', '10',
+            '--threads', str(threads),
+            '--remove-tmp-files', '0',
+            '--use-all-table-starts', '1'
+        ]
+        
+        try:
+            subprocess.run(cmd, check=True)
+            logger.info(f"MMseqs2 profiling completed for {input_file}")
+            
+        except subprocess.CalledProcessError as e:
+            logger.error(f"Failed to process {input_file}: {e}")
+            continue
+
+
+def predict_aerobes(input_dir: str = None, output_file: str = "per_aerobe_predictions.tsv",
                    mode: str = "modern", id_cut: float = None, bit_cut: float = None, 
                    e_cut: float = None, threads: int = 4):
     """
-    Predict aerobe levels from DIAMOND results
+    Predict aerobe levels from DIAMOND or MMseqs2 results
+    Mode determines method: modern=DIAMOND, ancient=MMseqs2
     """
     from .utils import get_package_data_path
     
     logger.info(f"Starting aerobe level prediction in {mode} mode")
     
+    if input_dir is None:
+        input_dir = 'diamond_output' if mode == 'modern' else 'mmseqs_output'
+    
     if mode == "modern":
         identity_cutoff, bitscore_cutoff, evalue_cutoff = 60.0, 50.0, 0.001
     elif mode == "ancient":
-        identity_cutoff, bitscore_cutoff, evalue_cutoff = 45.0, 25.0, 0.1
+        identity_cutoff, bitscore_cutoff, evalue_cutoff = 86.0, 50.0, 0.001
     elif mode == "custom":
         if any(x is None for x in [id_cut, bit_cut, e_cut]):
             raise OxyMetaGError("Custom mode requires id_cut, bit_cut, and e_cut parameters")
@@ -194,7 +292,8 @@ def predict_aerobes(input_dir: str = "diamond_output", output_file: str = "per_a
         raise OxyMetaGError("Mode must be 'modern', 'ancient', or 'custom'")
     
     package_data_dir = str(Path(get_package_data_path("")).parent / "data")
-    r_script_path = get_package_data_path("../scripts/predict_oxygen.R")
+    package_base = Path(__file__).parent
+    r_script_path = str(package_base / "scripts" / "predict_oxygen.R")
     
     if not Path(input_dir).exists():
         raise OxyMetaGError(f"Input directory not found: {input_dir}")
@@ -223,4 +322,4 @@ def predict_aerobes(input_dir: str = "diamond_output", output_file: str = "per_a
         logger.info(f"Results saved to {output_file}")
         return results_df
     else:
-        raise OxyMetaGError(f"Output file was not created: {output_file}")
+        raise OxyMetaGError(f"Output file was not created: {output_file}")

oxymetag/data/VTML20.out

@@ -0,0 +1,33 @@
+# VTML_20
+#
+# This matrix was produced from: vtml_20qij.mat using vtml_P.mat background frequencies
+#
+# VTML_20 substitution matrix, Units = bits/2.0
+# Expected score = -2.916179 bits; Entropy = 2.912514 bits
+# Target fraction identity = 0.8307
+# Lowest Score = -16, Highest Score= 12
+#
+# Background (precomputed optional): 0.0721 0.0135 0.0522 0.0728 0.038 0.0804 0.0256 0.0699 0.0703 0.107 0.0232 0.0503 0.0393 0.0339 0.0523 0.0698 0.05 0.0683 0.0143 0.0384 0.00001
+# Lambda     (precomputed optional): 0.34657
+	A	C	D	E	F	G	H	I	K	L	M	N	P	Q	R	S	T	V	W	Y	X
+A	7	-3	-6	-5	-8	-4	-7	-7	-6	-7	-5	-6	-4	-5	-7	-2	-3	-3	-9	-8	0
+C	-3	12	-14	-14	-13	-7	-6	-5	-13	-12	-4	-8	-9	-13	-7	-3	-5	-3	-15	-4	0
+D	-6	-14	8	-1	-16	-6	-4	-12	-5	-15	-9	-1	-6	-4	-12	-5	-6	-9	-10	-14	0
+E	-5	-14	-1	7	-14	-6	-6	-10	-2	-8	-8	-5	-6	-1	-10	-5	-6	-7	-16	-7	0
+F	-8	-13	-16	-14	9	-11	-5	-5	-14	-3	-3	-10	-9	-8	-10	-7	-8	-6	-3	0	0
+G	-4	-7	-6	-6	-11	7	-7	-15	-7	-11	-10	-5	-8	-8	-7	-4	-8	-10	-9	-10	0
+H	-7	-6	-4	-6	-5	-7	10	-9	-5	-7	-12	-3	-6	-2	-3	-5	-5	-8	-6	-1	0
+I	-7	-5	-12	-10	-5	-15	-9	7	-9	-2	-2	-9	-10	-9	-8	-9	-5	1	-6	-8	0
+K	-6	-13	-5	-2	-14	-7	-5	-9	7	-8	-5	-3	-6	-2	0	-5	-4	-8	-9	-8	0
+L	-7	-12	-15	-8	-3	-11	-7	-2	-8	6	0	-9	-7	-6	-8	-8	-7	-3	-6	-6	0
+M	-5	-4	-9	-8	-3	-10	-12	-2	-5	0	10	-7	-10	-4	-6	-8	-4	-3	-13	-11	0
+N	-6	-8	-1	-5	-10	-5	-3	-9	-3	-9	-7	8	-8	-4	-5	-2	-4	-9	-10	-6	0
+P	-4	-9	-6	-6	-9	-8	-6	-10	-6	-7	-10	-8	9	-5	-7	-4	-6	-7	-9	-15	0
+Q	-5	-13	-4	-1	-8	-8	-2	-9	-2	-6	-4	-4	-5	9	-2	-4	-5	-7	-15	-12	0
+R	-7	-7	-12	-10	-10	-7	-3	-8	0	-8	-6	-5	-7	-2	8	-6	-6	-9	-8	-7	0
+S	-2	-3	-5	-5	-7	-4	-5	-9	-5	-8	-8	-2	-4	-4	-6	7	-1	-8	-8	-6	0
+T	-3	-5	-6	-6	-8	-8	-5	-5	-4	-7	-4	-4	-6	-5	-6	-1	8	-4	-15	-8	0
+V	-3	-3	-9	-7	-6	-10	-8	1	-8	-3	-3	-9	-7	-7	-9	-8	-4	7	-13	-8	0
+W	-9	-15	-10	-16	-3	-9	-6	-6	-9	-6	-13	-10	-9	-15	-8	-8	-15	-13	12	-2	0
+Y	-8	-4	-14	-7	0	-10	-1	-8	-8	-6	-11	-6	-15	-12	-7	-6	-8	-8	-2	9	0
+X	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	1

oxymetag/data/nucleotide.out

@@ -0,0 +1,9 @@
+# NUCL in 1/2 Bit
+# Background (precomputed optional): 0.2499975 0.2499975 0.2499975 0.2499975 0.00001
+# Lambda     (precomputed optional): 0.6337314
+   A       C       T       G       X
+A  2.0000 -3.0000 -3.0000 -3.0000 -3.0000
+C -3.0000  2.0000 -3.0000 -3.0000 -3.0000
+T -3.0000 -3.0000  2.0000 -3.0000 -3.0000
+G -3.0000 -3.0000 -3.0000  2.0000 -3.0000
+X -3.0000 -3.0000 -3.0000 -3.0000 -3.0000