diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.bumpversion.cfg b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.bumpversion.cfg
new file mode 100644
index 00000000..5d6e5cf5
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.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/pdbbind_refined_v2020/__init__.py]
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.dockerignore b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.dockerignore
new file mode 100644
index 00000000..7c603f81
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.dockerignore
@@ -0,0 +1,4 @@
+.venv
+out
+tests
+__pycache__
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.gitignore b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.gitignore
new file mode 100644
index 00000000..c04bc49f
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/.gitignore
@@ -0,0 +1 @@
+poetry.lock
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/CHANGELOG.md b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/CHANGELOG.md
new file mode 100644
index 00000000..b67793f7
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/CHANGELOG.md
@@ -0,0 +1,5 @@
+# CHANGELOG
+
+## 0.1.0
+
+Initial release.
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/Dockerfile b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/Dockerfile
new file mode 100644
index 00000000..dc2d33f1
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/Dockerfile
@@ -0,0 +1,32 @@
+# docker build -f Dockerfile -t polusai/pdbbind-refined-v2020-tool .
+
+FROM condaforge/mambaforge
+
+ENV EXEC_DIR="/opt/executables"
+ENV POLUS_LOG="INFO"
+RUN mkdir -p ${EXEC_DIR}
+
+RUN apt-get update && apt-get install -y wget
+# Download the bdbbind dataset
+# RUN wget --no-clobber http://www.pdbbind.org.cn/download/PDBbind_v2020_refined.tar.gz
+## update to the new download URL (around 10 times faster) from PDBbind website.
+RUN wget --no-clobber https://pdbbind.oss-cn-hangzhou.aliyuncs.com/download/PDBbind_v2020_refined.tar.gz
+RUN tar -xvf PDBbind_v2020_refined.tar.gz
+
+# Work directory defined in the base container
+
+COPY pyproject.toml ${EXEC_DIR}
+COPY VERSION ${EXEC_DIR}
+COPY README.md ${EXEC_DIR}
+COPY CHANGELOG.md ${EXEC_DIR}
+
+# Install needed packages here
+RUN mamba install -c conda-forge pandas
+
+COPY src ${EXEC_DIR}/src
+
+RUN pip3 install ${EXEC_DIR} --no-cache-dir
+
+CMD ["--help"]
+
+WORKDIR /outdir
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/README.md b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/README.md
new file mode 100644
index 00000000..51e39081
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/README.md
@@ -0,0 +1,16 @@
+# pdbbind_refined_v2020 (0.1.0)
+
+Extract pdbbind_refined_v2020 data
+
+## Options
+
+This plugin takes 7 input arguments
+
+| Name | Description | I/O | Type | Default |
+|---------------|-------------------------|--------|--------|---------|
+| index_file_name | | Input | string | string |
+| base_dir | | Input | string | string |
+| query | query str to search the dataset. Pandas query doesn't support slash(/) in column names please use Kd_Ki instead of Kd/Ki, Type: string, File type: input, Accepted formats: txt | Input | string | string |
+| min_row | The row min inex, Type: int | Input | int | int |
+| max_row | The row max inex, Type: int | Input | int | int |
+| convert_Kd_dG | If this is set to true, dG will be calculated | Input | boolean | boolean |
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/VERSION b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/VERSION
new file mode 100644
index 00000000..6e8bf73a
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/VERSION
@@ -0,0 +1 @@
+0.1.0
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/__init__.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/__init__.py
new file mode 100644
index 00000000..187d0cd9
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/__init__.py
@@ -0,0 +1 @@
+"""pdbbind_refined_v2020_plugin package."""
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/build-docker.sh b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/build-docker.sh
new file mode 100755
index 00000000..cd23f2cb
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/build-docker.sh
@@ -0,0 +1,4 @@
+#!/bin/bash
+
+version=$(<VERSION)
+docker build . -t polusai/pdbbind-refined-v2020-tool:${version}
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/extract_pdbbind_refined_0.1.0.cwl b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/extract_pdbbind_refined_0.1.0.cwl
new file mode 100644
index 00000000..34c94b86
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/extract_pdbbind_refined_0.1.0.cwl
@@ -0,0 +1,246 @@
+#!/usr/bin/env cwl-runner
+cwlVersion: v1.0
+
+class: CommandLineTool
+
+label: Download the PDBbind refined database
+
+doc: |-
+ Download the PDBbind refined database
+
+baseCommand: ["python", "-m", "polus.mm.utils.pdbbind_refined_v2020"]
+
+hints:
+ DockerRequirement:
+ dockerPull: polusai/pdbbind-refined-v2020-tool@sha256:dcadf55c53c15c9f8b179d43cf96682b96dc1114f57b4e593497156c0cde8016
+
+
+requirements:
+ InlineJavascriptRequirement: {}
+
+inputs:
+
+ index_file_name:
+ label: The index file name
+ type: string
+ format:
+ - edam:format_2330
+ inputBinding:
+ prefix: --index_file_name
+ position: 1
+ default: INDEX_refined_data.2020
+
+ query:
+ label: query str to search the dataset, Pandas query doesn't support slash(/) in column names please use Kd_Ki instead of Kd/Ki
+ doc: |-
+ query str to search the dataset. Pandas query doesn't support slash(/) in column names please use Kd_Ki instead of Kd/Ki
+ Type: string
+ File type: input
+ Accepted formats: txt
+ type: string
+ format:
+ - edam:format_2330
+ inputBinding:
+ prefix: --query
+ position: 2
+
+ 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
+ position: 3
+ default: system.log
+
+ min_row:
+ label: The row min index
+ doc: |-
+ The row min inex
+ Type: int
+ type: int?
+ format:
+ - edam:format_2330
+ inputBinding:
+ position: 4
+ prefix: --min_row
+
+ max_row:
+ label: The row max index
+ doc: |-
+ The row max inex
+ Type: int
+ type: int?
+ format:
+ - edam:format_2330
+ inputBinding:
+ position: 5
+ prefix: --max_row
+
+ convert_Kd_dG:
+ label: If this is set to true, dG will be calculated
+ doc: If this is set to true, dG will be calculated
+ type: boolean
+ format:
+ - edam:format_2330
+ inputBinding:
+ prefix: --convert_Kd_dG
+ position: 6
+ default: False
+
+ experimental_dGs:
+ label: Experimental Free Energies of Binding
+ doc: |-
+ Experimental Free Energies of Binding
+ type: string?
+ format:
+ - edam:format_2330
+
+ pdb_ids:
+ label: The PDBID of proteins
+ doc: |-
+ The PDBID of proteins
+ type: string?
+ format:
+ - edam:format_2330
+
+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_pdb_paths:
+ label: Path to the input file
+ doc: |-
+ Path to the input file
+ Type: string
+ File type: input
+ Accepted formats: pdb
+ type: File[]
+ outputBinding:
+ # NOTE: Do NOT just use glob: ./*.pdb !!! This will return an array sorted by filenames.
+ # We want the order of output_pdb_paths to match the order of experimental_dGs, etc
+ # Because we need to compare experimental ΔGs with predicted values.
+ glob: $(inputs.output_txt_path)
+ loadContents: true
+ outputEval: |
+ ${
+ var lines = self[0].contents.split("\n");
+ var pdbs = [];
+ for (var i = 0; i < lines.length; i++) {
+ var words = lines[i].split(" ");
+ var pdbid = words[0];
+ var pdbfile = {"class": "File", "path": pdbid + "_protein.pdb"};
+ pdbs.push(pdbfile);
+ }
+
+ return pdbs;
+ }
+ format: edam:format_1476
+
+ output_sdf_paths:
+ label: Path to the input file
+ doc: |-
+ Path to the input file
+ Type: string
+ File type: input
+ Accepted formats: sdf
+ type: File[]
+ outputBinding:
+ # NOTE: Do NOT just use glob: ./*.sdf !!! This will return an array sorted by filenames.
+ # We want the order of output_sdf_paths to match the order of experimental_dGs, etc
+ # Because we need to compare experimental ΔGs with predicted values.
+ glob: $(inputs.output_txt_path)
+ loadContents: true
+ outputEval: |
+ ${
+ var lines = self[0].contents.split("\n");
+ var sdfs = [];
+ for (var i = 0; i < lines.length; i++) {
+ var words = lines[i].split(" ");
+ var pdbid = words[0];
+ var sdffile = {"class": "File", "path": pdbid + "_ligand.sdf"};
+ sdfs.push(sdffile);
+ }
+
+ return sdfs;
+ }
+ format: edam:format_3814
+
+ experimental_dGs:
+ label: Experimental Free Energies of Binding
+ doc: |-
+ Experimental Free Energies of Binding
+ type: ["null", {"type": "array", "items": "float"}]
+ outputBinding:
+ glob: $(inputs.output_txt_path)
+ loadContents: true
+ outputEval: |
+ ${
+ var lines = self[0].contents.split("\n");
+ var experimental_dGs = [];
+ for (var i = 0; i < lines.length; i++) {
+ var words = lines[i].split(" ");
+ if (words.length > 2) {
+ var experimental_dG = parseFloat(words[2]);
+ experimental_dGs.push(experimental_dG);
+ }
+ }
+
+ if (experimental_dGs.length == 0) {
+ return null;
+ } else {
+ return experimental_dGs;
+ }
+ }
+
+ pdb_ids:
+ label: The PDBID of proteins
+ doc: |-
+ The PDBID of proteins
+ type:
+ type: array
+ items: string
+ outputBinding:
+ glob: $(inputs.output_txt_path)
+ loadContents: true
+ outputEval: |
+ ${
+ var lines = self[0].contents.split("\n");
+ var pdbids = [];
+ for (var i = 0; i < lines.length; i++) {
+ var words = lines[i].split(" ");
+ pdbids.push(words[0]);
+ }
+
+ if (pdbids.length == 0) {
+ throw new Error("Error! pdbids are empty!");
+ } else {
+ return pdbids;
+ }
+ }
+
+ stdout:
+ type: File
+ outputBinding:
+ glob: stdout
+
+stdout: stdout
+
+$namespaces:
+ edam: https://edamontology.org/
+
+$schemas:
+- https://raw.githubusercontent.com/edamontology/edamontology/master/EDAM_dev.owl
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/ict.yml b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/ict.yml
new file mode 100644
index 00000000..feb05f30
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/ict.yml
@@ -0,0 +1,42 @@
+specVersion: 0.1.0
+name: labshare/pdbbind-refined-v2020
+version: 0.1.0
+container: polusai/pdbbind-refined-v2020-tool:0.1.0
+entrypoint: ""
+title: pdbbind_refined_v2020
+description: Extracts data from the PDBBind refined dataset
+author: Brandon Walker (brandon.walker@axleinfo.com), Nazanin Donyapour (nazanin.donyapour@nih.gov)
+repository: https://github.com/labshare/mmtools
+documentation: https://ncats.nih.gov/preclinical/core/informatics
+citation: ""
+inputs:
+- name: output_txt_path
+ required: false
+ description: Path to the text dataset file
+ type: string
+ default: /outdir/system.log
+- name: index_file_name
+ required: false
+ description: The index file name
+ type: string
+ default: INDEX_refined_data.2020
+- name: base_dir
+ required: true
+ description: The base_dir path
+ type: string
+- name: query
+ required: false
+ description: "query str to search the dataset, Pandas query doesn't support slash(/) in column names please use Kd_Ki instead of Kd/Ki"
+ type: string
+- name: min_row
+ required: false
+ description: The row min index
+ type: int
+- name: max_row
+ required: false
+ description: The row max index
+ type: int
+- name: convert_Kd_dG
+ required: false
+ description: If this is set to true, dG will be calculated
+ type: boolean
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/pyproject.toml b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/pyproject.toml
new file mode 100644
index 00000000..c8732d8b
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/pyproject.toml
@@ -0,0 +1,32 @@
+[tool.poetry]
+name = "polus-mm-utils-pdbbind-refined-v2020"
+version = "0.1.0"
+description = "An awesome function."
+authors = ["Data Scientist <data.scientist@labshare.org>"]
+readme = "README.md"
+packages = [{include = "polus", from = "src"}]
+
+[tool.poetry.dependencies]
+python = ">=3.9,<3.12"
+typer = "^0.7.0"
+pandas = "^1.3.3"
+cwl-utils = "0.33"
+cwltool = "3.1.20240404144621"
+
+[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 = [
+ "."
+]
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__init__.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__init__.py
new file mode 100644
index 00000000..d5992f46
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__init__.py
@@ -0,0 +1,7 @@
+"""pdbbind_refined_v2020."""
+
+__version__ = "0.1.0"
+
+from polus.mm.utils.pdbbind_refined_v2020.pdbbind_refined_v2020 import ( # noqa # pylint: disable=unused-import
+ pdbbind_refined_v2020,
+)
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__main__.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__main__.py
new file mode 100644
index 00000000..5ff8d576
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/__main__.py
@@ -0,0 +1,78 @@
+"""Package entrypoint for the pdbbind_refined_v2020 package."""
+
+# Base packages
+import logging
+from os import environ
+from pathlib import Path
+
+import typer
+from polus.mm.utils.pdbbind_refined_v2020.pdbbind_refined_v2020 import (
+ pdbbind_refined_v2020,
+)
+
+logging.basicConfig(
+ format="%(asctime)s - %(name)-8s - %(levelname)-8s - %(message)s",
+ datefmt="%d-%b-%y %H:%M:%S",
+)
+POLUS_LOG = getattr(logging, environ.get("POLUS_LOG", "INFO"))
+logger = logging.getLogger("polus.mm.utils.pdbbind_refined_v2020.")
+logger.setLevel(POLUS_LOG)
+
+app = typer.Typer(help="pdbbind_refined_v2020.")
+
+
+@app.command()
+def main( # noqa: PLR0913
+ output_txt_path: str = typer.Option(
+ "system.log",
+ "--output_txt_path",
+ help="Path to the text dataset file",
+ ),
+ index_file_name: str = typer.Option(
+ "INDEX_refined_data.2020",
+ "--index_file_name",
+ help="",
+ ),
+ query: str = typer.Option(
+ ...,
+ "--query",
+ help="query str to search the dataset.",
+ ),
+ min_row: int = typer.Option(
+ 1,
+ "--min_row",
+ help="The row min inex, Type: int",
+ ),
+ max_row: int = typer.Option(
+ ...,
+ "--max_row",
+ help="The row max inex, Type: int",
+ ),
+ convert_kd_dg: bool = typer.Option(
+ False,
+ "--convert_Kd_dG",
+ help="If this is set to true, dG will be calculated",
+ ),
+) -> None:
+ """pdbbind_refined_v2020."""
+ logger.info(f"output_txt_path: {output_txt_path}")
+ logger.info(f"index_file_name: {index_file_name}")
+ logger.info(f"query: {query}")
+ logger.info(f"min_row: {min_row}")
+ logger.info(f"max_row: {max_row}")
+ logger.info(f"convert_Kd_dG: {convert_kd_dg}")
+ base_dir = "/refined-set"
+ Path(base_dir)
+ pdbbind_refined_v2020(
+ output_txt_path,
+ index_file_name,
+ base_dir,
+ query,
+ min_row,
+ max_row,
+ convert_kd_dg,
+ )
+
+
+if __name__ == "__main__":
+ app()
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/pdbbind_refined_v2020.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/pdbbind_refined_v2020.py
new file mode 100644
index 00000000..cbcc59e4
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/src/polus/mm/utils/pdbbind_refined_v2020/pdbbind_refined_v2020.py
@@ -0,0 +1,174 @@
+"""extract pdbbind_refined_v2020."""
+import math
+import re
+import subprocess
+from collections import defaultdict
+from pathlib import Path
+from typing import Any
+
+import pandas as pd
+
+
+def pdbbind_refined_v2020( # noqa: PLR0913
+ output_txt_path: str,
+ index_file_name: str,
+ base_dir: str,
+ query: str,
+ min_row: int,
+ max_row: int,
+ convert_kd_dg: bool,
+) -> None:
+ """pdbbind_refined_v2020.
+
+ Args:
+ output_txt_path: Path to the text dataset file
+ index_file_name: The PDBbind index file name
+ base_dir: The base directory of the dataset
+ query: query str to search the dataset.
+ min_row: The row min index
+ max_row: The row max index
+ convert_kd_dg: If this is set to true, dG will be calculated
+ Returns:
+ None
+ """
+ load_data(
+ index_file_name,
+ base_dir,
+ query,
+ output_txt_path,
+ min_row,
+ max_row,
+ convert_kd_dg,
+ )
+
+
+def calculate_dg(kd: float) -> float:
+ """Calculates binding free energy from Kd.
+
+ Args:
+ kd (float): The binding affinity of the protein-ligand complex
+
+ Returns:
+ float: The binding free energy
+ """
+ # Calculate the binding free energy from kd so we can make the correlation plots.
+ # See https://en.wikipedia.org/wiki/Binding_constant
+ ideal_gas_constant = 8.31446261815324 # J/(Mol*K)
+ kcal_per_joule = 4184
+ # NOTE: Unfortunately, the temperature at which
+ # experimental kd binding data was taken
+ # is often not recorded. Thus, we are forced to guess. The two standard guesses are
+ # physiological body temperature (310K) or room temperature (298K).
+ temperature = 298
+ rt = (ideal_gas_constant / kcal_per_joule) * temperature
+ # NOTE: For performance, simulations are often done in a very small unit cell, and
+ # thus at a very high concentration. The size of the unit cell bounds the volume.
+ # For shorter simulations where the ligand has not explored the entire box, it may
+ # be less. See the Yank paper for a method of calculating the correct volumes.
+ standard_concentration = 1 # Units of mol / L, but see comment above.
+ return rt * math.log(kd / standard_concentration)
+
+
+def read_index_file(index_file_path: str) -> pd.DataFrame:
+ """Reads the PDBbind index file and extracts binding data.
+
+ Args:
+ index_file_path (str): The path to the index file
+
+ Returns:
+ pd.DataFrame: The kd data
+ """
+ data: dict[str, Any] = defaultdict(list)
+ # The file format
+ # PDB code, resolution, release year, -logkd/Ki, kd/Ki, reference, ligand name
+ unit_conv = {"uM": 1, "mM": 1000.0, "nM": 0.001, "pM": 0.000001}
+
+ with Path(index_file_path).open(encoding="utf-8") as rfile:
+ lines = [line for line in rfile.readlines() if line[0] != "#" and "Kd=" in line]
+ for line in lines:
+ words = line.split()
+ data["PDB_code"].append(words[0])
+ data["resolution"].append(words[1])
+ data["release_year"].append(words[2])
+
+ # Kd conversion to micro molar
+ unit = re.split(r"=[-+]?(?:\d*\.\d+|\d+)", words[4])[1]
+ standard_type = re.split(r"=[-+]?(?:\d*\.\d+|\d+)", words[4])[0]
+ kd = float(re.findall(r"[-+]?(?:\d*\.\d+|\d+)", words[4])[0])
+ data["Kd_Ki"].append(standard_type)
+ data["value"].append(kd * unit_conv[unit])
+ data["ligand_name"].append(re.findall(r"\((.*?)\)", words[7])[0])
+
+ return pd.DataFrame.from_dict(data)
+
+
+# pylint: disable=too-many-arguments,too-many-locals
+
+
+def load_data( # noqa: PLR0913
+ index_file_name: str,
+ base_dir: str,
+ query: str,
+ output_txt_path: str,
+ min_row: int = 1,
+ max_row: int = -1,
+ convert_kd_dg: bool = False,
+) -> None:
+ """Filters Kd data beased on a query.
+
+ Args:
+ index_file_name (str): The PDBbind index file name
+ base_dir (str): The base directory of the dataset
+ query (str): The Query to perform
+ output_txt_path (str): The output text file
+ min_row (int, optional): min index of rows. Defaults to 1.
+ max_row (int, optional): max index of rows. Defaults to -1.
+ convert_kd_dg (bool, optional): If this set to True,
+ The dG will be calculated. Defaults to False.
+ """
+ index_file_path = Path(base_dir).joinpath("index", index_file_name)
+ df = read_index_file(str(index_file_path))
+ # perform query
+ df = df.query(query)
+
+ # Perform row slicing (if any)
+ if int(min_row) != 1 or int(max_row) != -1:
+ # We want to convert to zero-based indices and we also want
+ # the upper index to be inclusive (i.e. <=) so -1 lower index.
+ df = df[
+ (int(min_row) - 1) : int(max_row)
+ ] # pylint: disable=unsubscriptable-object
+
+ # Calculate dG
+ df = df[["PDB_code", "value", "Kd_Ki"]]
+ binding_data: list[str] = []
+ micromolar = 0.000001 # uM
+ for _, row in enumerate(df.values):
+ (pdbcode, binding_datum, kd_ki) = row
+ binding_datum = binding_datum * micromolar
+
+ if convert_kd_dg:
+ dg = calculate_dg(binding_datum)
+ binding_data.append(f"{pdbcode} {binding_datum} {dg} {kd_ki}")
+ else:
+ binding_data.append(f"{pdbcode} {binding_datum} {kd_ki}")
+
+ with Path(output_txt_path).open(mode="w", encoding="utf-8") as f:
+ f.write("\n".join(binding_data))
+
+ # copy pdb and sdf files
+ for _, row in df.iterrows():
+ pdbcode = row["PDB_code"]
+ source_pdb_path = Path(base_dir).joinpath(pdbcode, f"{pdbcode}_protein.pdb")
+ dist_pdb_path = f"{pdbcode}_protein.pdb"
+ subprocess.run(
+ ["cp", f"{source_pdb_path}", f"{dist_pdb_path}"], # noqa: S603, S607
+ check=True,
+ )
+ source_sdf_path = Path(base_dir).joinpath(pdbcode, f"{pdbcode}_ligand.sdf")
+
+ dist_sdf_path = f"{pdbcode}_ligand.sdf"
+ subprocess.run(
+ ["cp", f"{source_sdf_path}", f"{dist_sdf_path}"], # noqa: S603, S607
+ check=True,
+ )
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/__init__.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/__init__.py
new file mode 100644
index 00000000..e0368608
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/__init__.py
@@ -0,0 +1 @@
+"""Tests for pdbbind_refined_v2020."""
diff --git a/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/test_pdbbind_refined_v2020.py b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/test_pdbbind_refined_v2020.py
new file mode 100644
index 00000000..65dd02dc
--- /dev/null
+++ b/utils/pre-process/data-download/pdbbind_refined_v2020_tool/tests/test_pdbbind_refined_v2020.py
@@ -0,0 +1,56 @@
+"""Tests for pdbbind_refined_v2020."""
+import sys
+from pathlib import Path
+
+sys.path.append("src")
+from polus.mm.utils.pdbbind_refined_v2020.pdbbind_refined_v2020 import ( # noqa: E402
+ pdbbind_refined_v2020,
+)
+
+current_dir = Path(__file__).resolve().parent
+target_dir = current_dir.parent.parent.parent.parent.parent / "cwl_utils"
+sys.path.append(str(target_dir))
+
+from cwl_utilities import call_cwltool # noqa: E402
+from cwl_utilities import create_input_yaml # noqa: E402
+from cwl_utilities import parse_cwl_arguments # noqa: E402
+
+
+def test_pdbbind_refined_v2020() -> None:
+ """Test pdbbind_refined_v2020."""
+ output_txt_path = "output.txt"
+ index_file_name = "INDEX_refined_data.2020"
+ base_dir = Path.cwd() / "refined-set"
+ query = '(Kd_Ki == "Kd") and (value < 0.001)'
+ min_row = 1
+ max_row = 1
+ convert_kd_dg = True
+ pdbbind_refined_v2020(
+ output_txt_path,
+ index_file_name,
+ base_dir,
+ query,
+ min_row,
+ max_row,
+ convert_kd_dg,
+ )
+ current_directory = Path.cwd() # Get the current directory
+ pdb_files = list(current_directory.glob("*.pdb")) # List all *.pdb files
+ sdf_files = list(current_directory.glob("*.sdf")) # List all *.sdf files
+
+ assert pdb_files, "No .pdb files found in the directory"
+ assert sdf_files, "No .sdf files found in the directory"
+
+
+def test_extract_pdbbind_refined() -> None:
+ """Test pdb."""
+ cwl_file = Path("extract_pdbbind_refined_0.1.0.cwl")
+ input_to_props = parse_cwl_arguments(cwl_file)
+ input_to_props["query"] = '(Kd_Ki == "Kd") and (value < 0.001)'
+ input_to_props["convert_Kd_dG"] = True
+ input_to_props["max_row"] = 1
+ input_yaml_path = Path("extract_pdbbind_refined_0.1.0.yml")
+ create_input_yaml(input_to_props, input_yaml_path)
+ call_cwltool(cwl_file, input_yaml_path)
+
+ assert Path("1e3g_protein.pdb").is_file()