Ruff Ruff

Author: rodvrees

im2deep/__init__.py

@@ -40,7 +40,7 @@
 
 # Import main functionality for easier access
 from importlib.metadata import version
-from im2deep.utils import ccs2im, im2ccs
+
 from im2deep.core import predict, predict_and_calibrate
 from im2deep.utils import ccs2im, im2ccs
 

im2deep/__main__.py

@@ -41,21 +41,21 @@
 
 from __future__ import annotations
 
-import logging
 import cProfile
+import logging
+from pathlib import Path
 
 import click
 from rich.console import Console
 
 from im2deep import __version__, core
-from pathlib import Path
 from im2deep.utils import (
-    setup_logging,
-    parse_input,
+    DefaultCommandGroup,
     build_credits,
-    write_output,
     infer_output_name,
-    DefaultCommandGroup,
+    parse_input,
+    setup_logging,
+    write_output,
 )
 
 console = Console()

im2deep/_architecture.py

@@ -1,10 +1,11 @@
-import sys
+import logging
 from pathlib import Path
+
+import lightning as L
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-import lightning as L
-import logging
+from scipy.stats import pearsonr
 
 try:
     import wandb
@@ -19,10 +20,12 @@
 
 logger = logging.getLogger(__name__)
 
+MAE = nn.L1Loss()
+
 
 class LogLowestMAE(L.Callback):
     def __init__(self, config):
-        super(LogLowestMAE, self).__init__()
+        super().__init__()
         self.bestMAE = float("inf")
         self.config = config
 
@@ -40,7 +43,7 @@ def on_validation_end(self, trainer, pl_module):
 
 class LRelu_with_saturation(nn.Module):
     def __init__(self, negative_slope, saturation):
-        super(LRelu_with_saturation, self).__init__()
+        super().__init__()
         self.negative_slope = negative_slope
         self.saturation = saturation
         self.leaky_relu = nn.LeakyReLU(self.negative_slope)
@@ -61,7 +64,7 @@ def __init__(
         negative_slope,
         saturation,
     ):
-        super(Conv1dActivation, self).__init__()
+        super().__init__()
         self.conv = nn.Conv1d(in_channels, out_channels, kernel_size, padding=padding)
         self.initializer = initializer
         self.activation = LRelu_with_saturation(
@@ -76,7 +79,7 @@ def forward(self, x):
 
 class DenseActivation(nn.Module):
     def __init__(self, in_features, out_features, initializer, negative_slope, saturation):
-        super(DenseActivation, self).__init__()
+        super().__init__()
         self.linear = nn.Linear(in_features, out_features)
         self.initializer = initializer
         self.activation = LRelu_with_saturation(
@@ -91,7 +94,7 @@ def forward(self, x):
 
 class SelfAttention(nn.Module):
     def __init__(self, feature_dim, heads=1):
-        super(SelfAttention, self).__init__()
+        super().__init__()
         self.feature_dim = feature_dim
         self.heads = heads
         # self.padded_dim = self.feature_dim + (self.feature_dim % self.heads)
@@ -152,7 +155,7 @@ def forward(self, x):
 
 class Branch(nn.Module):
     def __init__(self, input_size, output_size, add_layer=1, dropout_rate=0.0):
-        super(Branch, self).__init__()
+        super().__init__()
         self.add_layer = add_layer
         if self.add_layer:
             self.fc1 = nn.Linear(input_size, output_size)
@@ -172,7 +175,7 @@ def forward(self, x):
 
 class IM2Deep(L.LightningModule):
     def __init__(self, config, criterion):
-        super(IM2Deep, self).__init__()
+        super().__init__()
         self.config = config
         self.criterion = criterion
         self.mae = nn.L1Loss()
@@ -628,7 +631,7 @@ def configure_init(self):
 
 class IM2DeepMulti(L.LightningModule):
     def __init__(self, config, criterion):
-        super(IM2DeepMulti, self).__init__()
+        super().__init__()
         self.config = config
         self.criterion = criterion
 
@@ -1106,7 +1109,7 @@ def configure_init(self):
 
 class IM2DeepMultiTransfer(L.LightningModule):
     def __init__(self, config, criterion):
-        super(IM2DeepMultiTransfer, self).__init__()
+        super().__init__()
         # TODO: config should be adapted in config file
         self.config = config
         self.criterion = criterion
@@ -1123,7 +1126,7 @@ def __init__(self, config, criterion):
         self.ConvGlobal = self.backbone.ConvGlobal
         self.OneHot = self.backbone.OneHot
 
-        if self.config.get("add_X_mol", False) == True:
+        if self.config.get("add_X_mol", False):
             self.MolDesc = self.backbone.MolDesc
 
         self.concat = list(self.backbone.Concat.children())[:-1]
@@ -1294,7 +1297,7 @@ def configure_optimizers(self):
 
 class IM2DeepTransfer(L.LightningModule):
     def __init__(self, config, criterion):
-        super(IM2DeepTransfer, self).__init__()
+        super().__init__()
 
         self.config = config
         self.criterion = criterion
@@ -1312,7 +1315,7 @@ def __init__(self, config, criterion):
         self.ConvGlobal = self.backbone.ConvGlobal
         self.OneHot = self.backbone.OneHot
 
-        if self.config.get("add_X_mol", False) == True:
+        if self.config.get("add_X_mol", False):
             self.MolDesc = self.backbone.MolDesc
 
         self.concat = self.backbone.Concat
@@ -1439,7 +1442,7 @@ def configure_optimizers(self):
 
 class FlexibleLossSorted(nn.Module):
     def __init__(self, diversity_weight=0.1):
-        super(FlexibleLossSorted, self).__init__()
+        super().__init__()
         self.diversity_weight = diversity_weight
 
     def forward(self, y1, y2, y_hat1, y_hat2):
@@ -1479,7 +1482,7 @@ def forward(self, y1, y2, y_hat1, y_hat2):
 
 class FlexibleLoss(nn.Module):
     def __init__(self, diversity_weight=0.1):
-        super(FlexibleLoss, self).__init__()
+        super().__init__()
         self.diversity_weight = diversity_weight
 
     def forward(self, y1, y2, y_hat1, y_hat2):

im2deep/_model_ops.py

@@ -2,16 +2,16 @@
 """Training, predicting, and evaluating using IM2Deep (PyTorch)."""
 
 from __future__ import annotations
-import copy
+
 import logging
 import warnings
 from os import PathLike
 from pathlib import Path
 
+import lightning as L
 import torch
 from rich.progress import track
 from torch.utils.data import DataLoader, Dataset
-import lightning as L
 
 # Suppress PyTorch padding warning for conv1d with even kernels and odd dilation
 warnings.filterwarnings(

im2deep/calibration.py

@@ -8,15 +8,13 @@
 
 import logging
 from abc import ABC, abstractmethod
-from typing import cast
 
-import pandas as pd
 import numpy as np
-from psm_utils import PSMList, Peptidoform
+import pandas as pd
+from psm_utils import Peptidoform, PSMList
 
 from im2deep._exceptions import CalibrationError
-from im2deep.utils import parse_input
-from im2deep.constants import DEFAULT_REFERENCE_DATASET_PATH, DEFAULT_MULTI_REFERENCE_DATASET_PATH
+from im2deep.constants import DEFAULT_MULTI_REFERENCE_DATASET_PATH, DEFAULT_REFERENCE_DATASET_PATH
 
 LOGGER = logging.getLogger(__name__)
 
@@ -110,7 +108,7 @@ def fit(
                 LOGGER.warning(
                     f"Could not calculate charge-specific shift factors: {e}. Using 0.0 as fallback."
                 )
-                self.charge_shifts = {charge: 0.0 for charge in range(1, 7)}
+                self.charge_shifts = dict.fromkeys(range(1, 7), 0.0)
 
             # Set general shift as the mean of calculated charge shifts or charge 2 if available
             if 2 in self.charge_shifts and self.charge_shifts[2] != 0.0:
@@ -153,7 +151,7 @@ def fit(
                     f"Could not calculate general shift factor: {e}. Using 0.0 as fallback."
                 )
                 self.general_shift = 0.0
-            self.charge_shifts = {charge: self.general_shift for charge in range(1, 7)}
+            self.charge_shifts = dict.fromkeys(range(1, 7), self.general_shift)
 
         self.used_charges = set(self.charge_shifts.keys())
         self.fitted = True
@@ -172,7 +170,7 @@ def transform(
         if "peptidoform" not in psm_df.columns:
             raise CalibrationError("Input DataFrame must contain 'peptidoform' column.")
 
-        if not "predicted_CCS_uncalibrated" in psm_df.columns and "metadata" in psm_df.columns:
+        if "predicted_CCS_uncalibrated" not in psm_df.columns and "metadata" in psm_df.columns:
             psm_df["predicted_CCS_uncalibrated"] = psm_df["metadata"].apply(
                 lambda x: (
                     x["predicted_CCS_uncalibrated"]

im2deep/core.py

@@ -4,17 +4,16 @@
 
 import logging
 from os import PathLike
-from pathlib import Path
 
 import numpy as np
-from psm_utils.psm_list import PSMList
 import torch
 from deeplc.data import DeepLCDataset
+from psm_utils.psm_list import PSMList
 
-from im2deep.utils import validate_psm_list
 from im2deep import _model_ops
-from im2deep.calibration import LinearCCSCalibration, Calibration
+from im2deep.calibration import Calibration, LinearCCSCalibration
 from im2deep.constants import DEFAULT_MODEL, DEFAULT_MULTI_MODEL
+from im2deep.utils import validate_psm_list
 
 LOGGER = logging.getLogger(__name__)
 

im2deep/utils.py

@@ -15,28 +15,25 @@
 
 from __future__ import annotations
 
-import sys
-from pathlib import Path
-from typing import Any
 import logging
-from rich.text import Text
-import gzip
+from pathlib import Path
 
 import click
 import numpy as np
-import psm_utils.io
 import pandas as pd
+import psm_utils.io
+from psm_utils.psm import PSM
+from psm_utils.psm_list import PSMList
 from rich.console import Console
 from rich.logging import RichHandler
-from psm_utils.psm_list import PSMList
-from psm_utils.psm import PSM
+from rich.text import Text
 
 from im2deep._exceptions import IM2DeepError
 from im2deep.constants import (
-    SUMMARY_CONSTANT,
     MASS_GAS_N2,
-    TEMP,
+    SUMMARY_CONSTANT,
     T_DIFF,
+    TEMP,
 )
 
 console = Console()
@@ -286,7 +283,7 @@ def parse_input(
     is_legacy_format = False
     try:
         # Read first line to check column names
-        with open(input_file, "r") as f:
+        with open(input_file) as f:
             first_line = f.readline().strip()
 
         # Check if it has legacy format columns
@@ -307,7 +304,7 @@ def parse_input(
         # Try to parse with psm_utils
         try:
             psm_list = psm_utils.io.read_file(input_file, filetype=filetype or "infer")
-            LOGGER.debug(f"Successfully read file using psm_utils.")
+            LOGGER.debug("Successfully read file using psm_utils.")
         except Exception as e:
             # If psm_utils fails, try legacy format as fallback
             LOGGER.warning(f"Failed to read PSM file using psm_utils: {e}")
@@ -345,7 +342,7 @@ def _parse_legacy_format(input_file: str | Path) -> PSMList:
         df = pd.read_csv(input_file, sep=None, engine="python")
         df = df.fillna("")  # Replace NaN with empty strings
     except Exception as e:
-        raise IM2DeepError(f"Failed to read file as delimited text: {e}")
+        raise IM2DeepError(f"Failed to read file as delimited text: {e}") from e
 
     required_cols_legacy = ["seq", "modifications", "charge"]
     missing_cols = set(required_cols_legacy) - set(df.columns)

tests/conftest.py

@@ -1,10 +1,10 @@
 """Pytest configuration and fixtures for IM2Deep tests."""
 
-import pytest
+
 import numpy as np
 import pandas as pd
-from pathlib import Path
-from psm_utils import PSM, PSMList, Peptidoform
+import pytest
+from psm_utils import PSM, Peptidoform, PSMList
 
 
 @pytest.fixture
@@ -51,7 +51,7 @@ def sample_psm_list_with_ccs():
             is_decoy=False,
             score=0.95,
             retention_time=100.5,
-            metadata={"CCS": float(450.5)},
+            metadata={"CCS": 450.5},
         ),
         PSM(
             peptidoform=Peptidoform("SEQUENCE/3"),
@@ -61,7 +61,7 @@ def sample_psm_list_with_ccs():
             is_decoy=False,
             score=0.92,
             retention_time=120.3,
-            metadata={"CCS": float(520.8)},
+            metadata={"CCS": 520.8},
         ),
         PSM(
             peptidoform=Peptidoform("TESTPEPTIDE/2"),
@@ -71,7 +71,7 @@ def sample_psm_list_with_ccs():
             is_decoy=False,
             score=0.88,
             retention_time=135.7,
-            metadata={"CCS": float(480.2)},
+            metadata={"CCS": 480.2},
         ),
         PSM(
             peptidoform=Peptidoform("ANOTHER/3"),
@@ -81,7 +81,7 @@ def sample_psm_list_with_ccs():
             is_decoy=False,
             score=0.90,
             retention_time=142.1,
-            metadata={"CCS": float(510.5)},
+            metadata={"CCS": 510.5},
         ),
     ]
     return PSMList(psm_list=psms)

tests/test_calibration.py

@@ -1,12 +1,12 @@
 """Tests for calibration module."""
 
-import pytest
 import numpy as np
 import pandas as pd
-from psm_utils import Peptidoform, PSM, PSMList
+import pytest
+from psm_utils import Peptidoform
 
-from im2deep.calibration import LinearCCSCalibration, get_default_reference
 from im2deep._exceptions import CalibrationError
+from im2deep.calibration import LinearCCSCalibration, get_default_reference
 
 
 class TestLinearCCSCalibration: