pdb download (#12)

Co-authored-by: Brandon Duane Walker <[email protected]>

utils/extract-ligand-protein-plugin/Dockerfile → utils/pre-process/structure-change/extract-ligand-protein-tool/Dockerfile

@@ -1,3 +1,4 @@
+# docker build -f Dockerfile -t polusai/extract-ligand-protein-tool .
 FROM condaforge/mambaforge
 
 ENV EXEC_DIR="/opt/executables"
@@ -15,11 +16,6 @@ COPY CHANGELOG.md ${EXEC_DIR}
 
 # Install needed packages here
 
-RUN pip install filepattern
-RUN conda config --add channels conda-forge
-RUN conda install mdanalysis
-
-
 COPY src ${EXEC_DIR}/src
 
 RUN pip3 install ${EXEC_DIR} --no-cache-dir

utils/pre-process/structure-change/extract-ligand-protein-tool/build-docker.sh

@@ -0,0 +1,4 @@
+#!/bin/bash
+
+version=$(<VERSION)
+docker build . -t polusai/extract-ligand-protein-tool:${version}

utils/pre-process/structure-change/extract-ligand-protein-tool/extract_ligand_protein_0.1.0.cwl

@@ -0,0 +1,85 @@
+#!/usr/bin/env cwl-runner
+cwlVersion: v1.0
+
+class: CommandLineTool
+
+label: A tool that employs OpenMM to extract ligands and protein from a PDB file
+
+doc: |-
+  A tool that employs OpenMM to extract ligands and protein from a PDB file
+
+baseCommand: ["python", "-m", "polus.mm.utils.extract_ligand_protein"]
+
+hints:
+  DockerRequirement:
+    dockerPull: polusai/extract-ligand-protein-tool@sha256:38416f3d020c26869028c6ba66a243882d0b3c5885c79ff68355912ec5768fc1
+
+inputs:
+  input_pdb_path:
+    label: Input pdb file path
+    doc: |-
+      Input pdb file path
+      Type: string
+      File type: input
+      Accepted formats: pdb
+      Example file: https://github.com/bioexcel/biobb_structure_utils/raw/master/biobb_structure_utils/test/data/utils/cat_protein.pdb
+    type: File
+    format:
+    - edam:format_1476
+    inputBinding:
+      prefix: --input_pdb_path
+
+  output_pdb_path:
+    label: Output pdb file path
+    doc: |-
+      Output pdb file path
+      Type: string
+      File type: output
+      Accepted formats: pdb
+      Example file: https://github.com/bioexcel/biobb_structure_utils/raw/master/biobb_structure_utils/test/reference/utils/ref_cat_pdb.pdb
+    type: string
+    format:
+    - edam:format_1476
+    inputBinding:
+      prefix: --output_pdb_path
+    default: system.pdb
+
+  output_pdb_ligand_path:
+    label: Output pdb ligand file path
+    doc: |-
+      Output pdb ligand file path
+      Type: string
+      File type: output
+      Accepted formats: sdf
+    type: string
+    format:
+    - edam:format_1476
+    inputBinding:
+      prefix: --output_pdb_ligand_path
+    default: ligand_system.pdb
+
+outputs:
+  output_pdb_path:
+    label: Output pdb file path
+    doc: |-
+      Output pdb file path
+    type: File
+    outputBinding:
+      glob: $(inputs.output_pdb_path)
+    format: edam:format_1476
+
+  output_pdb_ligand_path:
+    label: Output ligand pdb file path
+    doc: |-
+      Output ligand pdb file path
+      Use optional File? since ligand may not exist in complex
+    type: File?
+    outputBinding:
+      glob: $(inputs.output_pdb_ligand_path)
+    format: edam:format_1476
+
+$namespaces:
+  edam: https://edamontology.org/
+
+$schemas:
+- https://raw.githubusercontent.com/edamontology/edamontology/master/EDAM_dev.owl

utils/extract-ligand-protein-plugin/ict.yml → utils/pre-process/structure-change/extract-ligand-protein-tool/ict.yml

@@ -1,12 +1,12 @@
 specVersion: "0.1.0"
 name: extract_ligand_protein
 version: 0.1.0
-container: extract-ligand-protein-plugin
+container: extract-ligand-protein-tool
 entrypoint:
 title: extract_ligand_protein
 description: A tool that employs OpenMM to extract ligands and protein from a PDB file
-author: Data Scientist
-contact: [email protected]
+author: Brandon Walker
+contact: [email protected]
 repository:
 documentation:
 citation:

utils/extract-ligand-protein-plugin/src/polus/mm/utils/extract_ligand_protein/extract_ligand_protein.py → utils/pre-process/structure-change/extract-ligand-protein-tool/src/polus/mm/utils/extract_ligand_protein/extract_ligand_protein.py

@@ -51,15 +51,15 @@ def extract_single_ligand_protein(  # noqa: PLR0912
     except MDAnalysis.exceptions.NoDataError:
         print("No bonds found in the PDB file.")  # noqa: T201
 
-    # Identify water molecules based on the
-    # connectivity pattern (Oxygen bonded to two Hydrogens)
+    # Identify water molecules based on the connectivity
+    # pattern (Oxygen bonded to two Hydrogens)
     if has_bonds:
         water_indices = set()
         for atom in dup_u.atoms:  # dont use selection resname == 'HOH',
             # pdb file may have different water residue names
-            h_bonds = 2
+            num_bonds = 2
             if (
-                atom.name == "O" and len(atom.bonds) == h_bonds
+                atom.name == "O" and len(atom.bonds) == num_bonds
             ):  # if hydrogens are added
                 bonded_atoms_names = {a.name for a in atom.bonded_atoms}
                 if bonded_atoms_names == {"H"}:  # Check if both bonds are Hydrogens
@@ -90,9 +90,14 @@ def extract_single_ligand_protein(  # noqa: PLR0912
     )  # needed for coordinates
     ligand_u.atoms = ligand_atoms
 
-    protein_u.atoms.write(str(output_pdb_path))
+    with output_pdb_path.open(mode="w", encoding="utf-8") as output_file:
+        protein_u.atoms.write(output_file)
     if len(ligand_u.atoms) > 0:  # will crash if no ligand atoms
-        ligand_u.atoms.write(str(output_pdb_ligand_path))
+        with output_pdb_ligand_path.open(
+            mode="w",
+            encoding="utf-8",
+        ) as output_ligand_file:
+            ligand_u.atoms.write(output_ligand_file)
 
 
 def extract_all_ligand_protein(input_pdb_path: list[Path], outdir: Path) -> None: