Merge pull request #26 from rs-station/sfcif

Support sfCIF files as alternatives to MTZs

docs/quickstart.md

@@ -39,10 +39,10 @@ At this point, you should be good to go! Please [file an issue on github](https:
 ## Input files
 
 To run `matchmaps`, you will need:
- - one `.pdb` file containing a refined structural model corresponding to your "off" data.
- - two `.mtz` files corresponding to your "on" and "off" data respectively.
+ - one `.pdb` (or `.cif`) file containing a refined structural model corresponding to your "off" data.
+ - two `.mtz` (or `.cif`) files corresponding to your "on" and "off" data respectively.
 
-You will also need to know the names of the columns in these `mtz`s containing your observed structure factor amplitudes and uncertainties. Depending on what software you used to produce these files, this may be something like `FP`/`SIGFP`, `Fobs`/`SIGFobs`, or similar. If you don't know these off-hand, you can figure it out using [`reciprocalspaceship`](https://rs-station.github.io/reciprocalspaceship/)'s `rs.mtzdump` utility, which is installed along with `matchmaps`. You can do this right in the command-line as:
+You will also need to know the names of the columns in these `mtz`/`cif`s containing your observed structure factor amplitudes and uncertainties. Depending on what software you used to produce these files, this may be something like `FP`/`SIGFP`, `Fobs`/`SIGFobs`, or similar. If you don't know these off-hand and your input is an `.mtz` file, you can figure it out using [`reciprocalspaceship`](https://rs-station.github.io/reciprocalspaceship/)'s `rs.mtzdump` utility, which is installed along with `matchmaps`. You can do this right in the command-line as:
 ```bash
 rs.mtzdump mymtz.mtz
 ```

pyproject.toml

@@ -29,7 +29,7 @@ dynamic = ["version"]
 dependencies = [
     "numpy",
     "tqdm",
-    "reciprocalspaceship",
+    "reciprocalspaceship>=1.0.1",
     "rs-booster>=0.0.1",
     "gemmi"
 ]

src/matchmaps/_compute_mr_diff.py

@@ -23,6 +23,7 @@
     phaser_wrapper,
     _clean_up_files,
     _cif_or_pdb_to_pdb,
+    _cif_or_mtz_to_mtz,
 )
 
 
@@ -94,17 +95,16 @@ def compute_mr_difference_map(
 
     output_dir_contents = list(output_dir.glob("*"))
 
-    off_name = mtzoff.name.removesuffix(".mtz")
-    on_name = mtzon.name.removesuffix(".mtz")
-
     pdboff = _cif_or_pdb_to_pdb(pdboff, output_dir)
 
+    mtzoff, off_name = _cif_or_mtz_to_mtz(mtzoff, output_dir)
+    mtzon, on_name = _cif_or_mtz_to_mtz(mtzon, output_dir)
+
     # take in the list of rbr selections and parse them into phenix and gemmi selection formats
     # if rbr_groups = None, just returns (None, None)
     rbr_phenix, rbr_gemmi = _rbr_selection_parser(rbr_selections)
 
     # this is where scaling takes place in the usual pipeline, but that doesn't make sense with different-spacegroup inputs
-    # side note: I need to test the importance of scaling even in the normal case!! Might be more artifact than good, who knows
 
     pdboff = _handle_special_positions(pdboff, output_dir)
 
@@ -133,8 +133,6 @@ def compute_mr_difference_map(
         output_dir=output_dir,
     )
 
-    # the refinement process *should* be identical. Waters are gone already
-    # I just need to make sure that the phaser outputs go together
     print(f"{time.strftime('%H:%M:%S')}: Running phenix.refine for the 'on' data...")
 
     nickname_on = rigid_body_refinement_wrapper(
@@ -166,7 +164,6 @@ def compute_mr_difference_map(
     )
 
     # from here down I just copied over the stuff from the normal version
-    # this should be proofread for compatibility but should all work
 
     # read back in the files created by phenix
     # these have knowable names
@@ -255,7 +252,7 @@ def parse_arguments():
         metavar=("mtzfileoff", "Foff", "SigFoff"),
         required=True,
         help=(
-            "MTZ containing off/apo/ground/dark state data. "
+            "MTZ or sfCIF containing off/apo/ground/dark state data. "
             "Specified as [filename F SigF]"
         ),
     )
@@ -267,7 +264,7 @@ def parse_arguments():
         metavar=("mtzfileon", "Fon", "SigFon"),
         required=True,
         help=(
-            "MTZ containing on/bound/excited/bright state data. "
+            "MTZ or SFCIF containing on/bound/excited/bright state data. "
             "Specified as [filename F SigF]"
             "This file may be in a different spacegroup / crystal packing than mtzoff"
         ),
@@ -278,7 +275,7 @@ def parse_arguments():
         "-p",
         required=True,
         help=(
-            "Reference pdb/cif corresponding to the off/apo/ground/dark state. "
+            "Reference PDB or mmCIF corresponding to the off/apo/ground/dark state. "
             "Used as a molecular replacement solution for mtzon and for rigid-body refinement of both input MTZs to generate phases."
             "Should match mtzoff well enough that molecular replacement is not necessary."
         ),
@@ -298,7 +295,7 @@ def parse_arguments():
         "-i",
         required=False,
         default="./",
-        help="Path to input mtzs and pdb. Optional, defaults to './' (current directory)",
+        help="Path to input files. Optional, defaults to './' (current directory)",
     )
 
     parser.add_argument(

src/matchmaps/_compute_ncs_diff.py

@@ -24,6 +24,7 @@
     _validate_inputs,
     _clean_up_files,
     _cif_or_pdb_to_pdb,
+    _cif_or_mtz_to_mtz,
 )
 
 
@@ -51,6 +52,8 @@ def compute_ncs_difference_map(
 
     pdb = _cif_or_pdb_to_pdb(pdb, output_dir)
 
+    mtz, _ = _cif_or_mtz_to_mtz(mtz, output_dir)
+
     rbr_phenix, rbr_gemmi = _rbr_selection_parser(ncs_chains)
 
     if Phi is None:  # do rigid-body refinement to get phases
@@ -109,7 +112,7 @@ def compute_ncs_difference_map(
     )
 
     print(f"{time.strftime('%H:%M:%S')}: Cleaning up files...")
-    print(keep_temp_files)
+
     _clean_up_files(output_dir, output_dir_contents, keep_temp_files)
 
     print(f"{time.strftime('%H:%M:%S')}: Done!")
@@ -136,7 +139,7 @@ def parse_arguments():
         # metavar=("mtzfile", "F", "SigF"),
         required=True,
         help=(
-            "MTZ file containing structure factor amplitudes. "
+            "MTZ or sfCIF file containing structure factor amplitudes. "
             "Specified as [filename F SigF] or [filename F]. "
             "SigF is not necessary if phases are also provided"
         ),
@@ -147,7 +150,7 @@ def parse_arguments():
         required=False,
         default=None,
         help=(
-            "Optional. Column in MTZ file containing phases. "
+            "Optional. Column in MTZ/sfCIF file containing phases. "
             "If phases are not provided, phases will be computed via rigid-body refinement of "
             "the provided model and structure factor amplitudes."
         ),
@@ -158,8 +161,8 @@ def parse_arguments():
         "-p",
         required=True,
         help=(
-            "Reference pdb/cif. "
-            "If phases are not provided, used for rigid-body refinement of input MTZ to generate phases."
+            "Reference PDB or mmCIF. "
+            "If phases are not provided, used for rigid-body refinement of input MTZ/sfCIF to generate phases."
         ),
     )
 
@@ -197,7 +200,7 @@ def parse_arguments():
         "-i",
         required=False,
         default="./",
-        help="Path to input mtz and pdb. Optional, defaults to './' (current directory)",
+        help="Path to input files. Optional, defaults to './' (current directory)",
     )
 
     parser.add_argument(

src/matchmaps/_compute_realspace_diff.py

@@ -22,7 +22,7 @@
     _clean_up_files,
     _validate_environment,
     _validate_inputs,
-    #_cif_or_mtz_to_mtz,
+    _cif_or_mtz_to_mtz,
     _cif_or_pdb_to_pdb,
 )
 
@@ -94,13 +94,10 @@ def compute_realspace_difference_map(
 
     output_dir_contents = list(output_dir.glob("*"))
 
-    off_name = mtzoff.name.removesuffix(".mtz")
-    on_name = mtzon.name.removesuffix(".mtz")
-
     pdboff = _cif_or_pdb_to_pdb(pdboff, output_dir)
 
-    # off_name = _cif_or_mtz_to_mtz(mtzoff)
-    # on_name = _cif_or_mtz_to_mtz(mtzon)
+    mtzoff, off_name = _cif_or_mtz_to_mtz(mtzoff, output_dir)
+    mtzon, on_name = _cif_or_mtz_to_mtz(mtzon, output_dir)
 
     # take in the list of rbr selections and parse them into phenix and gemmi selection formats
     # if rbr_groups = None, just returns (None, None)
@@ -262,7 +259,7 @@ def parse_arguments():
         metavar=("mtzfileoff", "Foff", "SigFoff"),
         required=True,
         help=(
-            "MTZ containing off/apo/ground/dark state data. "
+            "MTZ or sfCIF containing off/apo/ground/dark state data. "
             "Specified as [filename F SigF]"
         ),
     )
@@ -274,7 +271,7 @@ def parse_arguments():
         metavar=("mtzfileon", "Fon", "SigFon"),
         required=True,
         help=(
-            "MTZ containing on/bound/excited/bright state data. "
+            "MTZ or sfCIF containing on/bound/excited/bright state data. "
             "Specified as [filename F SigF]"
         ),
     )
@@ -284,7 +281,7 @@ def parse_arguments():
         "-p",
         required=True,
         help=(
-            "Reference pdb/cif corresponding to the off/apo/ground/dark state. "
+            "Reference PDB or mmCIF corresponding to the off/apo/ground/dark state. "
             "Used for rigid-body refinement of both input MTZs to generate phases."
         ),
     )
@@ -303,7 +300,7 @@ def parse_arguments():
         "-i",
         required=False,
         default="./",
-        help="Path to input mtzs and pdb. Optional, defaults to './' (current directory)",
+        help="Path to input files. Optional, defaults to './' (current directory)",
     )
 
     parser.add_argument(

src/matchmaps/_utils.py

@@ -885,17 +885,62 @@ def _validate_inputs(
     return input_dir, output_dir, ligands, *files
 
 
-# def _cif_or_mtz_to_mtz(input_file, output_dir):
+def _cif_or_mtz_to_mtz(input_file, output_dir):
+    """
+    Return a File() object and also the original file name as a string
+
+    Parameters
+    ----------
+    input_file : pathlib.Path
+        _description_
+    output_dir : pathlib.Path
+        _description_
+
+    Returns
+    -------
+    (output_file, filename)
+
+    """
 
-#     if path.suffix.lower() == '.mtz':
-#         reflections = rs.read_mtz(str(path))
+    if input_file.suffix.lower() == '.mtz':
+
+        output_file = output_dir / (input_file.name)
+
+        shutil.copy(input_file, output_file)
+
+    elif input_file.suffix.lower() == '.cif':
+
+        output_file = output_dir / (input_file.name.lower().removesuffix('.cif') + '.mtz')
+
+        reflections = rs.read_cif(str(input_file))
+
+        reflections.write_mtz(str(output_file))
 
-#     elif path.suffix.lower() == '.cif':
-#         reflections = rs.read_cif(str(path))
+    else:
+        raise ValueError(f"Invalid file type {input_file.suffix} for starting model, must be '.mtz' or '.cif'")
 
-#     return name
+    return (output_file, 
+            input_file.name.removesuffix(input_file.suffix))
 
 def _cif_or_pdb_to_pdb(input_file, output_dir):
+    """
+    _summary_
+
+    Parameters
+    ----------
+    input_file : pathlib.Path
+    output_dir : pathlib.Path
+
+    Returns
+    -------
+    pathlib.Path
+        path to output file
+
+    Raises
+    ------
+    ValueError
+        _description_
+    """
 
     if input_file.suffix.lower() == '.pdb':