dayhoff-tools 1.1.7__tar.gz → 1.1.9__tar.gz

{dayhoff_tools-1.1.7 → dayhoff_tools-1.1.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: dayhoff-tools
-Version: 1.1.7
+Version: 1.1.9
 Summary: Common tools for all the repos at Dayhoff Labs
 Author: Daniel Martin-Alarcon
 Author-email: dma@dayhofflabs.com

{dayhoff_tools-1.1.7 → dayhoff_tools-1.1.9}/dayhoff_tools/deployment/processors.py

@@ -160,7 +160,7 @@ class MMSeqsProfileProcessor(Processor):
         self.num_threads = str(num_threads)  # MMseqs2 expects string for threads
 
         default_mmseqs_args = {
-            "memory_limit_gb": "30",
+            "memory_limit_gb": "25",
             "evalue": "10",
             "sensitivity": "7.5",
             "max_seqs_search": "300",
@@ -221,12 +221,16 @@ class MMSeqsProfileProcessor(Processor):
 
         The input_file is the target FASTA. The query FASTA is provided
         during initialization.
+        The method creates an output directory (e.g., {target_stem})
+        which contains the result files, now named meaningfully using the target stem
+        (e.g., {target_stem}_results.m8 and {target_stem}_hits.fasta).
 
         Args:
             input_file: Path to the input target FASTA file.
 
         Returns:
-            Path to the output directory containing results.m8 and results.fasta.
+            Path to the output directory (e.g., {target_stem}) containing
+            the meaningfully named result files.
 
         Raises:
             subprocess.CalledProcessError: If any MMseqs2 command fails.
@@ -237,10 +241,11 @@ class MMSeqsProfileProcessor(Processor):
 
         input_file_path = Path(input_file).resolve()  # Ensure absolute path
         target_fasta_filename = input_file_path.name
+        target_fasta_stem = input_file_path.stem  # Get stem for naming
 
         # Create a unique base directory for this run's outputs and temp files
         # This directory will be returned and subsequently uploaded by the Operator
-        run_base_dir_name = f"mmseqs_run_{Path(target_fasta_filename).stem}"
+        run_base_dir_name = f"{target_fasta_stem}"  # Use stem as the dir name
         run_base_dir = Path(run_base_dir_name).resolve()
         run_base_dir.mkdir(parents=True, exist_ok=True)
         logger.info(f"Created run base directory: {run_base_dir}")
@@ -259,9 +264,13 @@ class MMSeqsProfileProcessor(Processor):
         mmseqs_temp_dir.mkdir(parents=True, exist_ok=True)
         logger.info(f"Created MMseqs2 temporary directory: {mmseqs_temp_dir}")
 
-        # Define output file paths directly within run_base_dir
-        local_results_m8_file = run_base_dir / "results.m8"
-        local_hits_fasta_file = run_base_dir / "results.fasta"
+        # Define INTERMEDIATE output file paths within mmseqs_temp_dir
+        intermediate_results_m8_file = mmseqs_temp_dir / "results.m8"
+        intermediate_hits_fasta_file = mmseqs_temp_dir / "results.fasta"
+
+        # Define FINAL output file paths within run_base_dir, using target stem
+        final_results_m8_file = run_base_dir / f"{target_fasta_stem}_results.m8"
+        final_hits_fasta_file = run_base_dir / f"{target_fasta_stem}_hits.fasta"
 
         # --- MMseqs2 Workflow Paths (intermediate files in mmseqs_temp_dir) ---
         query_db = mmseqs_temp_dir / "queryDB"
@@ -381,7 +390,7 @@ class MMSeqsProfileProcessor(Processor):
                 run_base_dir,
             )
 
-            # 8. Convert results to tabular format (M8)
+            # 8. Convert results to tabular format (M8) -> to intermediate file
             self._run_mmseqs_command(
                 [
                     "mmseqs",
@@ -389,7 +398,7 @@ class MMSeqsProfileProcessor(Processor):
                     str(profile_db),  # Query DB used for search (profileDB)
                     str(target_db),
                     str(result_db),
-                    str(local_results_m8_file),  # Output M8 file
+                    str(intermediate_results_m8_file),  # Output M8 file to temp dir
                     "--threads",
                     self.num_threads,
                 ],
@@ -404,26 +413,55 @@ class MMSeqsProfileProcessor(Processor):
                 run_base_dir,
             )
 
-            # 10. Convert hit sequences to FASTA
+            # 10. Convert hit sequences to FASTA -> to intermediate file
             self._run_mmseqs_command(
-                ["mmseqs", "convert2fasta", str(hits_db), str(local_hits_fasta_file)],
+                [
+                    "mmseqs",
+                    "convert2fasta",
+                    str(hits_db),
+                    str(intermediate_hits_fasta_file),
+                ],
                 "Convert hits to FASTA",
                 run_base_dir,
             )
 
             logger.info(
-                f"MMseqs2 workflow completed successfully. Outputs in {run_base_dir}"
+                f"MMseqs2 workflow completed successfully. Intermediate outputs in {mmseqs_temp_dir}"
             )
 
+            # Move and rename final output files from mmseqs_temp_dir to run_base_dir
+            if intermediate_results_m8_file.exists():
+                shutil.move(
+                    str(intermediate_results_m8_file), str(final_results_m8_file)
+                )
+                logger.info(f"Moved and renamed M8 results to {final_results_m8_file}")
+            else:
+                logger.warning(
+                    f"Intermediate M8 file {intermediate_results_m8_file} not found. Creating empty target file."
+                )
+                final_results_m8_file.touch()  # Create empty file in run_base_dir if not found
+
+            if intermediate_hits_fasta_file.exists():
+                shutil.move(
+                    str(intermediate_hits_fasta_file), str(final_hits_fasta_file)
+                )
+                logger.info(f"Moved and renamed hits FASTA to {final_hits_fasta_file}")
+            else:
+                logger.warning(
+                    f"Intermediate hits FASTA {intermediate_hits_fasta_file} not found. Creating empty target file."
+                )
+                final_hits_fasta_file.touch()  # Create empty file in run_base_dir if not found
+
         finally:
-            # Clean up the MMseqs2 temporary directory
+            # Clean up the MMseqs2 temporary directory (mmseqs_tmp) which contains intermediate DBs etc.
             if mmseqs_temp_dir.exists():
                 shutil.rmtree(mmseqs_temp_dir)
                 logger.info(
                     f"Cleaned up MMseqs2 temporary directory: {mmseqs_temp_dir}"
                 )
-            # The input_file (original target) is managed by the Operator
-            # The local_target_file (copy inside run_base_dir) will be cleaned up
-            # by the Operator when run_base_dir is deleted after upload.
+            # The local_target_file (copy of input) inside run_base_dir will be cleaned up
+            # by the Operator when run_base_dir itself is deleted after upload.
 
-        return str(run_base_dir)  # Return the path to the directory containing results
+        return str(
+            run_base_dir
+        )  # Return the path to the directory containing meaningfully named results

{dayhoff_tools-1.1.7 → dayhoff_tools-1.1.9}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "poetry.core.masonry.api"
 
 [project]
 name = "dayhoff-tools"
-version = "1.1.7"
+version = "1.1.9"
 description = "Common tools for all the repos at Dayhoff Labs"
 authors = [
     {name = "Daniel Martin-Alarcon", email = "dma@dayhofflabs.com"}