436 integrate ptms from evidence into alphafold predictions

backend/.gitignore

@@ -13,6 +13,10 @@
 !user_data/workflows/.test-run-empty.yaml
 !user_data/example_dataset
 
+# PTM cif files - necessary to explicitly include parent
+!user_data/external_data/
+user_data/external_data/alphafold
+
 # Uploads (e.g. Uniprot databases)
 /uploads/*
 

backend/protzilla/all_steps.py

@@ -15,6 +15,7 @@
     importing.EvidenceImport,
     importing.ExampleDatasetImport,
     importing.FastaImport,
+    importing.CifImport,
     importing.AlphaFoldPredictionLoad,
     importing.CrosslinkingImport,
     importing.AlphaFoldQueryJsonGeneration,
@@ -94,11 +95,13 @@
     data_integration.PlotGOEnrichmentDotPlot,
     data_integration.PlotGSEADotPlot,
     data_integration.PlotGSEAEnrichmentPlot,
+    data_integration.AddPTMsFromEvidenceToPrediction,
 ]
 
 # Steps excluded from UI, i.e. users cannot instantiate them
 _hidden_steps: list[Step] = [
     importing.ArbitraryCSVImport,
+    importing.CifImport,
     data_integration.DatabaseIntegrationByUniprot,
     data_integration.PlotGSEAEnrichmentPlot,
     data_analysis.PTMsProteinAndPerSample,

backend/protzilla/constants/cif_constants.py

@@ -0,0 +1,187 @@
+import pandas as pd
+
+from enum import Enum, StrEnum
+from pathlib import Path
+
+from typing_extensions import override
+
+from backend.protzilla.constants.paths import PTM_PATH
+from backend.protzilla.form import Option
+
+
+ATOM_SITE_PREFIX = "_atom_site."
+
+
+class ATOM_SITE_COLUMNS(StrEnum):
+    """
+    Enum containing all column names that should be present in
+    the _atom_site. table for mmCIF files from PDB or AFDB
+    """
+
+    GROUP_PDB = f"{ATOM_SITE_PREFIX}group_PDB"
+    ID = f"{ATOM_SITE_PREFIX}id"
+    TYPE_SYMBOL = f"{ATOM_SITE_PREFIX}type_symbol"
+    LABEL_ATOM_ID = f"{ATOM_SITE_PREFIX}label_atom_id"
+    LABEL_ALT_ID = f"{ATOM_SITE_PREFIX}label_alt_id"
+    LABEL_COMP_ID = f"{ATOM_SITE_PREFIX}label_comp_id"
+    LABEL_ASYM_ID = f"{ATOM_SITE_PREFIX}label_asym_id"
+    LABEL_ENTITY_ID = f"{ATOM_SITE_PREFIX}label_entity_id"
+    LABEL_SEQ_ID = f"{ATOM_SITE_PREFIX}label_seq_id"
+    PDBX_PDB_INS_CODE = f"{ATOM_SITE_PREFIX}pdbx_PDB_ins_code"
+    CARTN_X = f"{ATOM_SITE_PREFIX}Cartn_x"
+    CARTN_Y = f"{ATOM_SITE_PREFIX}Cartn_y"
+    CARTN_Z = f"{ATOM_SITE_PREFIX}Cartn_z"
+    OCCUPANCY = f"{ATOM_SITE_PREFIX}occupancy"
+    B_ISO_OR_EQUIV = f"{ATOM_SITE_PREFIX}B_iso_or_equiv"
+    PDBX_FORMAL_CHARGE = f"{ATOM_SITE_PREFIX}pdbx_formal_charge"
+    AUTH_SEQ_ID = f"{ATOM_SITE_PREFIX}auth_seq_id"
+    AUTH_COMP_ID = f"{ATOM_SITE_PREFIX}auth_comp_id"
+    AUTH_ASYM_ID = f"{ATOM_SITE_PREFIX}auth_asym_id"
+    AUTH_ATOM_ID = f"{ATOM_SITE_PREFIX}auth_atom_id"
+    PDBX_PDB_MODEL_NUM = f"{ATOM_SITE_PREFIX}pdbx_PDB_model_num"
+
+
+ATOM_SITE_COLUMNS_NUMERIC = [
+    ATOM_SITE_COLUMNS.ID,
+    ATOM_SITE_COLUMNS.LABEL_ENTITY_ID,
+    ATOM_SITE_COLUMNS.LABEL_SEQ_ID,
+    ATOM_SITE_COLUMNS.CARTN_X,
+    ATOM_SITE_COLUMNS.CARTN_Y,
+    ATOM_SITE_COLUMNS.CARTN_Z,
+    ATOM_SITE_COLUMNS.OCCUPANCY,
+    ATOM_SITE_COLUMNS.B_ISO_OR_EQUIV,
+    ATOM_SITE_COLUMNS.AUTH_SEQ_ID,
+]
+
+CHEM_COMP_PREFIX = "_chem_comp."
+
+# convert flags to native booleans
+CIF_BOOL_MAP = {"y": True, "n": False, ".": pd.NA}
+
+
+class CHEM_COMP_COLUMNS(StrEnum):
+    """
+    Enum containing all column names that should be present in
+    the _chem_comp. table for mmCIF files from PDB or AFDB
+    """
+
+    ID = f"{CHEM_COMP_PREFIX}id"
+    TYPE = f"{CHEM_COMP_PREFIX}type"
+    MON_NSTD_FLAG = f"{CHEM_COMP_PREFIX}mon_nstd_flag"
+    NAME = f"{CHEM_COMP_PREFIX}name"
+    PDBX_SYNONYMS = f"{CHEM_COMP_PREFIX}pdbx_synonyms"
+    FORMULA = f"{CHEM_COMP_PREFIX}formula"
+    FORMULA_WEIGHT = f"{CHEM_COMP_PREFIX}formula_weight"
+
+
+BACKBONE_ATOMS = ["N", "CA", "C", "O"]
+
+CHEM_COMP_ATOM_PREFIX = "_chem_comp_atom."
+
+
+class CHEM_COMP_ATOM_COLUMNS(StrEnum):
+    """
+    Enum containing all column names expected in the _chem_comp_atom. table
+    in the cif files describing modified residues
+    """
+
+    COMP_ID = f"{CHEM_COMP_ATOM_PREFIX}comp_id"
+    ATOM_ID = f"{CHEM_COMP_ATOM_PREFIX}atom_id"
+    ALT_ATOM_ID = f"{CHEM_COMP_ATOM_PREFIX}alt_atom_id"
+    TYPE_SYMBOL = f"{CHEM_COMP_ATOM_PREFIX}type_symbol"
+    CHARGE = f"{CHEM_COMP_ATOM_PREFIX}charge"
+    MODEL_CARTN_X = f"{CHEM_COMP_ATOM_PREFIX}model_Cartn_x"
+    MODEL_CARTN_Y = f"{CHEM_COMP_ATOM_PREFIX}model_Cartn_y"
+    MODEL_CARTN_Z = f"{CHEM_COMP_ATOM_PREFIX}model_Cartn_z"
+    IDEAL_CARTN_X = f"{CHEM_COMP_ATOM_PREFIX}pdbx_model_Cartn_x_ideal"
+    IDEAL_CARTN_Y = f"{CHEM_COMP_ATOM_PREFIX}pdbx_model_Cartn_y_ideal"
+    IDEAL_CARTN_Z = f"{CHEM_COMP_ATOM_PREFIX}pdbx_model_Cartn_z_ideal"
+    LEAVING_FLAG = f"{CHEM_COMP_ATOM_PREFIX}pdbx_leaving_atom_flag"
+    BACKBONE_FLAG = f"{CHEM_COMP_ATOM_PREFIX}pdbx_backbone_atom_flag"
+
+
+# convert column name from chem_comp_atom columns to atom_site columns
+ATOM_SITE_TO_CHEM_COMP = {
+    ATOM_SITE_COLUMNS.LABEL_ATOM_ID: CHEM_COMP_ATOM_COLUMNS.ATOM_ID,
+    ATOM_SITE_COLUMNS.AUTH_ATOM_ID: CHEM_COMP_ATOM_COLUMNS.ATOM_ID,
+    ATOM_SITE_COLUMNS.TYPE_SYMBOL: CHEM_COMP_ATOM_COLUMNS.TYPE_SYMBOL,
+    ATOM_SITE_COLUMNS.PDBX_FORMAL_CHARGE: CHEM_COMP_ATOM_COLUMNS.CHARGE,
+    ATOM_SITE_COLUMNS.CARTN_X: CHEM_COMP_ATOM_COLUMNS.IDEAL_CARTN_X,
+    ATOM_SITE_COLUMNS.CARTN_Y: CHEM_COMP_ATOM_COLUMNS.IDEAL_CARTN_Y,
+    ATOM_SITE_COLUMNS.CARTN_Z: CHEM_COMP_ATOM_COLUMNS.IDEAL_CARTN_Z,
+}
+
+CHEM_COMP_BOOLEAN_COLUMNS = {
+    CHEM_COMP_ATOM_COLUMNS.LEAVING_FLAG,
+    CHEM_COMP_ATOM_COLUMNS.BACKBONE_FLAG,
+}
+
+
+class Modification(StrEnum):
+    Acetylation = "Acetyl"
+    Citrullination = "Deamidation"
+    Ubiquitination = "GlyGly"
+    Methylation = "Methyl"
+    Phosphorylation = "Phospho"
+
+
+class Residue(StrEnum):
+    Lysine = "K"
+    Arginine = "R"
+    Serine = "S"
+    Threonine = "T"
+    Tyrosine = "Y"
+
+
+class KnownPTM(Enum):
+    """
+    Enum holding all PTM pairings currently supported by the PTM insertion into cif files.
+    Allows for construction via
+        KnownPTM.from_strings("Acetylation", "Lysine")
+    and path retrieval through .get_cif_path() on a constructed member
+    """
+
+    ACETYLATION_LYSINE = (Modification.Acetylation, Residue.Lysine)
+    CITRULLINATION_ARGININE = (Modification.Citrullination, Residue.Arginine)
+    METHYLATION_ARGININE = (Modification.Methylation, Residue.Arginine)
+    METHYLATION_LYSINE = (Modification.Methylation, Residue.Lysine)
+    PHOSPHORYLATION_SERINE = (Modification.Phosphorylation, Residue.Serine)
+    PHOSPHORYLATION_THREONINE = (Modification.Phosphorylation, Residue.Threonine)
+    PHOSPHORYLATION_TYROSINE = (Modification.Phosphorylation, Residue.Tyrosine)
+    UBIQUITINATION_LYSINE = (Modification.Ubiquitination, Residue.Lysine)
+
+    @property
+    def modification(self) -> Modification:
+        return self.value[0]
+
+    @property
+    def residue(self) -> Residue:
+        return self.value[1]
+
+    @override
+    def __str__(self) -> str:
+        return f"{self.modification.name} on {self.residue.name}"
+
+    @classmethod
+    def to_options(cls) -> list[Option]:
+        return [Option(member.name, str(member)) for member in cls]
+
+    def get_cif_path(self) -> Path:
+        """
+        Retrieves the cif path of the PTM
+        """
+        mod, res = self.value
+        path = (PTM_PATH / mod.name / res.name).with_suffix(".cif")
+        return path
+
+    @classmethod
+    def from_strings(cls, modification: str, residue: str) -> "KnownPTM":
+        """
+        Factory method for known PTMs, by passing the modification and residue
+        """
+        try:
+            return cls((Modification(modification), Residue(residue)))
+        except ValueError:
+            raise ValueError(
+                f"There is currently no PTM data for {modification} on {residue}"
+            )

backend/protzilla/constants/data_types.py

@@ -29,6 +29,7 @@ class DataKey(StrEnum):
     CONFIDENCE_DF = "confidence_df"
     FULL_DATA_DF = "full_data_df"
     JOB_REQUEST_DF = "job_request_df"
+    MODIFICATION_DF = "modification_df"
 
 
 ProteinDf = NewType("ProteinDf", pd.DataFrame)

backend/protzilla/constants/paths.py

@@ -30,3 +30,4 @@
     EXAMPLE_DATASET_DIR / "txt_REL_FREE-REPASE/proteinGroups.txt"
 )
 EXAMPLE_DATASET_EVIDENCE_FILE = EXAMPLE_DATASET_DIR / "txt_REL_FREE-REPASE/evidence.txt"
+PTM_PATH = EXTERNAL_DATA_PATH / "ptm"

backend/protzilla/constants/peptide_columns.py

@@ -1,12 +1,66 @@
-MAX_QUANT_PEPTIDE_COLUMNS = [
-    "Leading razor protein",
-    "Sequence",
-    "Missed cleavages",
-    "PEP",
-]
-MAX_QUANT_EVIDENCE_COLUMNS = MAX_QUANT_PEPTIDE_COLUMNS + [
-    "Modifications",
-    "Modified sequence",
-    "Experiment",
-    "Raw file",
-]
+from enum import StrEnum
+
+
+class MAX_QUANT_PEPTIDE_COLUMNS(StrEnum):
+    """
+    Enum containing column names expected in raw MQ peptide.txt files
+    """
+
+    LEADING_RAZOR_PROTEIN = "Leading razor protein"
+    SEQUENCE = "Sequence"
+    MISSED_CLEAVAGES = "Missed cleavages"
+    PEP = "PEP"
+
+
+class PEPTIDE_DF_COLUMNS(StrEnum):
+    """
+    Enum containing column names expected in a peptide_df after import
+    """
+
+    SEQUENCE = "Sequence"
+    MISSED_CLEAVAGES = "Missed cleavages"
+    PEP = "PEP"
+    # During the peptide import, Leading razor protein is renamed to Protein ID
+    PROTEIN_ID = "Protein ID"
+
+
+class MAX_QUANT_EVIDENCE_COLUMNS(StrEnum):
+    """
+    Enum containing column names expected in raw MQ evidence.txt files
+    """
+
+    LEADING_RAZOR_PROTEIN = "Leading razor protein"
+    SEQUENCE = "Sequence"
+    MISSED_CLEAVAGES = "Missed cleavages"
+    PEP = "PEP"
+    MODIFICATIONS = "Modifications"
+    MODIFIED_SEQUENCE = "Modified sequence"
+    EXPERIMENT = "Experiment"
+    RAW_FILE = "Raw file"
+
+
+class PSM_DF_COLUMNS(StrEnum):
+    """
+    Enum containing column names expected in a psm_df after import
+    """
+
+    SEQUENCE = "Sequence"
+    MISSED_CLEAVAGES = "Missed cleavages"
+    PEP = "PEP"
+    MODIFICATIONS = "Modifications"
+    MODIFIED_SEQUENCE = "Modified sequence"
+    RAW_FILE = "Raw file"
+    # During the evidence import, Leading razor protein is renamed to Protein ID
+    PROTEIN_ID = "Protein ID"
+    SAMPLE = "Sample"
+
+
+class MODIFICATION_COLUMNS(StrEnum):
+    """
+    Enum containing all column names expected in a modification_df
+    """
+
+    PROTEIN_ID = "Protein ID"
+    MODIFICATION = "Modification"
+    PROTEIN_LOCATION = "Protein Location"
+    RESIDUE = "Residue"