score_pdb_structures

utils/score-pdb-structures-plugin/.bumpversion.cfg

@@ -0,0 +1,29 @@
+[bumpversion]
+current_version = 0.1.0
+commit = False
+tag = False
+parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\-(?P<release>[a-z]+)(?P<dev>\d+))?
+serialize = 
+	{major}.{minor}.{patch}-{release}{dev}
+	{major}.{minor}.{patch}
+
+[bumpversion:part:release]
+optional_value = _
+first_value = dev
+values = 
+	dev
+	_
+
+[bumpversion:part:dev]
+
+[bumpversion:file:pyproject.toml]
+search = version = "{current_version}"
+replace = version = "{new_version}"
+
+[bumpversion:file:VERSION]
+
+[bumpversion:file:README.md]
+
+[bumpversion:file:plugin.json]
+
+[bumpversion:file:src/polus/mm/utils/score_pdb_structures/__init__.py]

utils/score-pdb-structures-plugin/.dockerignore

@@ -0,0 +1,4 @@
+.venv
+out
+tests
+__pycache__

utils/score-pdb-structures-plugin/.gitignore

@@ -0,0 +1 @@
+poetry.lock

utils/score-pdb-structures-plugin/CHANGELOG.md

@@ -0,0 +1,5 @@
+# CHANGELOG
+
+## 0.1.0
+
+Initial release.

utils/score-pdb-structures-plugin/Dockerfile

@@ -0,0 +1,22 @@
+FROM condaforge/mambaforge
+
+ENV EXEC_DIR="/opt/executables"
+ENV POLUS_LOG="INFO"
+RUN mkdir -p ${EXEC_DIR}
+
+
+# Work directory defined in the base container
+# WORKDIR ${EXEC_DIR}
+
+COPY pyproject.toml ${EXEC_DIR}
+COPY VERSION ${EXEC_DIR}
+COPY README.md ${EXEC_DIR}
+COPY CHANGELOG.md ${EXEC_DIR}
+
+# Install needed packages here
+
+COPY src ${EXEC_DIR}/src
+
+RUN pip3 install ${EXEC_DIR} --no-cache-dir
+
+CMD ["--help"]

utils/score-pdb-structures-plugin/README.md

@@ -0,0 +1,18 @@
+# score_pdb_structures (0.1.0)
+
+Fetches the PDB information from RCSB and scores PDB structures.
+
+## Options
+
+This plugin takes     6     input arguments and 2 output argument:
+
+| Name          | Description             | I/O    | Type   | Default |
+|---------------|-------------------------|--------|--------|---------|
+| input_pdbids | List of input PDBIDs to score, Type: string[], File type: input, Accepted formats: list[string] | Input | ['null', {'type': 'array', 'items': 'string'}] | ['null', {'type': 'array', 'items': 'string'}] |
+| min_row | The row min inex, Type: int | Input | int | int |
+| max_row | The row max inex, Type: int | Input | int | int |
+| output_txt_path | Path to the text dataset file, Type: string, File type: output, Accepted formats: txt | Input | string | string |
+| timeout_duration | The maximum time in seconds to wait for a response from the API before timing out, Type: int | Input | int | int |
+| max_retries | The maximum number of times to retry the request in case of failure, Type: int | Input | int | int |
+| output_txt_path | Path to the txt file | Output | File | File |
+| output_pdbids | The selected PDB IDs | Output | {'type': 'array', 'items': 'string'} | {'type': 'array', 'items': 'string'} |

utils/score-pdb-structures-plugin/VERSION

@@ -0,0 +1 @@
+0.1.0

utils/score-pdb-structures-plugin/build-docker.sh

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

utils/score-pdb-structures-plugin/ict.yml

@@ -0,0 +1,89 @@
+specVersion: "0.1.0"
+name: score_pdb_structures
+version: 0.1.0
+container: score-pdb-structures-plugin
+entrypoint:
+title: score_pdb_structures
+description: Fetches the PDB information from RCSB and scores PDB structures.
+author: Brandon Walker, Nazanin Donyapour
+contact: [email protected], [email protected]
+repository:
+documentation:
+citation:
+
+inputs:
+  - name: input_pdbids
+    required: true
+    description: List of input PDBIDs to score, Type string[], File type input, Accepted formats list[string]
+    type: ['null', {'type': 'array', 'items': 'string'}]
+    format:
+      uri: edam:format_2330
+  - name: min_row
+    required: true
+    description: The row min inex, Type int
+    type: int
+    format:
+      uri: edam:format_2330
+  - name: max_row
+    required: true
+    description: The row max inex, Type int
+    type: int
+    format:
+      uri: edam:format_2330
+  - name: output_txt_path
+    required: true
+    description: Path to the text dataset file, Type string, File type output, Accepted formats txt
+    type: string
+    defaultValue: system.log
+    format:
+      uri: edam:format_2330
+  - name: timeout_duration
+    required: true
+    description: The maximum time in seconds to wait for a response from the API before timing out, Type int
+    type: int
+    defaultValue: 10
+    format:
+      uri: edam:format_2330
+  - name: max_retries
+    required: true
+    description: The maximum number of times to retry the request in case of failure, Type int
+    type: int
+    defaultValue: 5
+    format:
+      uri: edam:format_2330
+outputs:
+  - name: output_txt_path
+    required: true
+    description: Path to the txt file
+    type: File
+    format:
+      uri: edam:format_2330
+  - name: output_pdbids
+    required: true
+    description: The selected PDB IDs
+    type: {'type': 'array', 'items': 'string'}
+ui:
+  - key: inputs.input_pdbids
+    title: "input_pdbids: "
+    description: "List of input PDBIDs to score, Type string[], File type input, Accepted formats list[string]"
+    type: ['null', {'type': 'array', 'items': 'string'}]
+  - key: inputs.min_row
+    title: "min_row: "
+    description: "The row min inex, Type int"
+    type: int
+  - key: inputs.max_row
+    title: "max_row: "
+    description: "The row max inex, Type int"
+    type: int
+  - key: inputs.output_txt_path
+    title: "output_txt_path: "
+    description: "Path to the text dataset file, Type string, File type output, Accepted formats txt"
+    type: string
+  - key: inputs.timeout_duration
+    title: "timeout_duration: "
+    description: "The maximum time in seconds to wait for a response from the API before timing out, Type int"
+    type: int
+  - key: inputs.max_retries
+    title: "max_retries: "
+    description: "The maximum number of times to retry the request in case of failure, Type int"
+    type: int

utils/score-pdb-structures-plugin/pyproject.toml

@@ -0,0 +1,31 @@
+[tool.poetry]
+name = "polus-mm-utils-score-pdb-structures"
+version = "0.1.0"
+description = "Fetches the PDB information from RCSB and scores PDB structures."
+authors = ["Nazanin Donyapour <[email protected]>", "Brandon Walker <[email protected]>"]
+readme = "README.md"
+packages = [{include = "polus", from = "src"}]
+
+[tool.poetry.dependencies]
+python = ">=3.9,<3.13"
+typer = "^0.7.0"
+sophios = "0.1.4"
+pandas = "2.2.2"
+
+[tool.poetry.group.dev.dependencies]
+bump2version = "^1.0.1"
+pytest = "^7.4"
+pytest-sugar = "^0.9.6"
+pre-commit = "^3.2.1"
+black = "^23.3.0"
+mypy = "^1.1.1"
+ruff = "^0.0.270"
+
+[build-system]
+requires = ["poetry-core"]
+build-backend = "poetry.core.masonry.api"
+
+[tool.pytest.ini_options]
+pythonpath = [
+  "."
+]

utils/score-pdb-structures-plugin/score_pdb_structures_0@1@0.cwl

@@ -0,0 +1,135 @@
+#!/usr/bin/env cwl-runner
+cwlVersion: v1.0
+
+class: CommandLineTool
+
+label: Fetches the PDB information from RCSB and scores PDB structures.
+
+doc: |-
+  Fetches the PDB information from RCSB and scores PDB structures.
+
+baseCommand: ["python", "-m", "polus.mm.utils.score_pdb_structures"]
+
+hints:
+  DockerRequirement:
+    dockerPull: polusai/score-pdb-structures-tool@sha256:eb16431eefabef6daacdd9554e6cdb77e2c9396d6be5ac310ef3f7918c15a9a2
+
+requirements:
+  InlineJavascriptRequirement: {}
+
+
+inputs:
+  input_pdbids:
+    label: List of input PDBIDs to score
+    doc: |-
+      List of input PDBIDs to score
+      Type: string[]
+      File type: input
+      Accepted formats: list[string]
+    type: ["null", {"type": "array", "items": "string"}]
+    format: edam:format_2330
+    inputBinding:
+      prefix: --input_pdbids
+    default: []
+
+  min_row:
+    label: The row min index
+    doc: |-
+      The row min inex
+      Type: int
+    type: int?
+    format:
+    - edam:format_2330
+    inputBinding:
+      prefix: --min_row
+    default: 1
+
+  max_row:
+    label: The row max index
+    doc: |-
+      The row max inex
+      Type: int
+    type: int?
+    format:
+    - edam:format_2330
+    inputBinding:
+      prefix: --max_row
+    default: -1
+
+  output_txt_path:
+    label: Path to the text dataset file
+    doc: |-
+      Path to the text dataset file
+      Type: string
+      File type: output
+      Accepted formats: txt
+    type: string
+    format:
+    - edam:format_2330
+    inputBinding:
+      prefix: --output_txt_path
+    default: system.log
+
+  timeout_duration:
+    label: The maximum time in seconds to wait for a response from the API before timing out
+    doc: |-
+      The maximum time in seconds to wait for a response from the API before timing out
+      Type: int
+    type: int?
+    format:
+    - edam:format_2330
+    inputBinding:
+      prefix: --timeout_duration
+    default: 10
+
+  max_retries:
+    label: The maximum number of times to retry the request in case of failure
+    doc: |-
+      The maximum number of times to retry the request in case of failure
+      Type: int
+    type: int?
+    format:
+    - edam:format_2330
+    inputBinding:
+      prefix: --max_retries
+    default: 5
+
+outputs:
+  output_txt_path:
+    label: Path to the txt file
+    doc: |-
+      Path to the txt file
+    type: File
+    outputBinding:
+      glob: $(inputs.output_txt_path)
+    format: edam:format_2330
+
+  output_pdbids:
+    label: The selected PDB IDs
+    doc: |-
+      The selected PDB IDs
+    type:
+      type: array
+      items: string
+    outputBinding:
+      glob: $(inputs.output_txt_path)
+      loadContents: true
+      outputEval: |
+        ${
+          // check if self[0] exists
+          if (!self[0]) {
+            return null;
+          }
+          var lines = self[0].contents.split("\n");
+          // remove black lines
+          lines = lines.filter(function(line) {return line.trim() !== '';});
+          // The outpus file has one line
+          // The format of the line is as follows: 6x7z,4yo7,5fyr,6ktl,4rxm,4irx,3v16,6b5z,4mio,7d5n
+          return lines[0].split(",").map(function(item) {return item.trim();});
+        }
+
+$namespaces:
+  edam: https://edamontology.org/
+
+$schemas:
+- https://raw.githubusercontent.com/edamontology/edamontology/master/EDAM_dev.owl

utils/score-pdb-structures-plugin/src/polus/mm/utils/score_pdb_structures/__init__.py

@@ -0,0 +1,7 @@
+"""score_pdb_structures."""
+
+__version__ = "0.1.0"
+
+from polus.mm.utils.score_pdb_structures.score_pdb_structures import (  # noqa # pylint: disable=unused-import
+    score_pdb_structures,
+)