[ENH] pdb to String loader using SEQREQ records

pyaptamer/datasets/__init__.py

@@ -2,7 +2,7 @@
 
 from pyaptamer.datasets._loaders._one_gnh import load_1gnh_structure
 from pyaptamer.datasets._loaders._online_databank import load_from_rcsb
-from pyaptamer.datasets._loaders._pfoa_loader import load_pfoa_structure
+from pyaptamer.datasets._loaders._pfoa import load_pfoa_structure
 
 __all__ = [
     "load_pfoa_structure",

pyaptamer/datasets/_loaders/__init__.py

@@ -1,6 +1,6 @@
 """Loaders for different data structures."""
 
 from pyaptamer.datasets._loaders._one_gnh import load_1gnh_structure
-from pyaptamer.datasets._loaders._pfoa_loader import load_pfoa_structure
+from pyaptamer.datasets._loaders._pfoa import load_pfoa_structure
 
 __all__ = ["load_pfoa_structure", "load_1gnh_structure"]

pyaptamer/utils/pdb_to_aaseq.py

@@ -0,0 +1,43 @@
+__author__ = "satvshr"
+__all__ = ["pdb_to_aaseq"]
+
+import os
+
+import pandas as pd
+from Bio import SeqIO
+
+
+def pdb_to_aaseq(pdb_file_path, return_df=False):
+    """
+    Extract amino-acid sequences (SEQRES) from a PDB file.
+
+    Parameters
+    ----------
+    pdb_file_path : str or os.PathLike
+        Path to a PDB file.
+    return_df : bool, optional, default=False
+        If True, return a pandas.DataFrame with columns:
+          - 'chain' (if available) and
+          - 'sequence' (one-letter amino-acid string per chain).
+        If False, return a list of strings (one per chain).
+
+    Returns
+    -------
+    list of str or pandas.DataFrame
+        List of amino-acid sequences (one-letter codes), or a DataFrame containing the
+        sequences.extracted from the SEQRES records in the PDB file. Each element is a
+        string representing the full sequence for a chain (e.g. "MKWVTFISLL..."). The
+        order of sequences matches the order in which SEQRES records are encountered in
+        the file. Returns an empty list if no SEQRES records are found.
+    """
+    pdb_path = os.fspath(pdb_file_path)
+    sequences = []
+
+    with open(pdb_path) as handle:
+        for record in SeqIO.parse(handle, "pdb-seqres"):
+            sequences.append(str(record.seq))
+
+    if return_df:
+        return pd.DataFrame({"sequence": sequences})
+
+    return sequences

pyaptamer/utils/tests/test_pdb_to_aaseq.py

@@ -0,0 +1,32 @@
+__author__ = "satvshr"
+
+import os
+
+from pyaptamer.utils.pdb_to_aaseq import pdb_to_aaseq
+
+
+def test_pdb_to_aaseq():
+    """
+    Test that `pdb_to_aaseq` converts a PDB file path into a non-empty string
+    containing alphabetic characters.
+    """
+    pdb_file_path = os.path.join(
+        os.path.dirname(__file__), "..", "..", "datasets", "data", "1gnh.pdb"
+    )
+    sequences = pdb_to_aaseq(pdb_file_path)
+
+    assert isinstance(sequences, list), "pdb_to_aaseq should return a list"
+    assert len(sequences) > 0, "Returned list should not be empty"
+
+    for seq in sequences:
+        assert isinstance(seq, str), "Each entry should be a string"
+        assert len(seq) > 0, "Each sequence string should not be empty"
+
+    sequences = pdb_to_aaseq(pdb_file_path, return_df=True)
+
+    assert not sequences.empty, "Returned DataFrame should not be empty"
+    assert "sequence" in sequences.columns, "DataFrame should have a 'sequence' column"
+
+    for seq in sequences["sequence"]:
+        assert isinstance(seq, str), "Each entry should be a string"
+        assert len(seq) > 0, "Each sequence string should not be empty"