Mercurial > repos > recetox > target_screen

comparison target_screen.py @ 1:6d51be3d7bb5 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/RECETOX/galaxytools/tree/master/tools/misc commit d6102c60e41d91adf1c7a876f84ef420a69262e2
author recetox
date Mon, 12 May 2025 14:05:37 +0000
parents d4c2d5bc0524
children
comparison
equal deleted inserted replaced
0:d4c2d5bc0524 1:6d51be3d7bb5
1 import argparse 1 import argparse
2 from typing import Tuple
2 3
3 import numpy as np 4 import numpy as np
4 import pandas as pd 5 import pandas as pd
5 6
6 7
7 def mz_match(marker, peak, ppm): 8 class LoadDataAction(argparse.Action):
9 """
10 Custom argparse action to load data from a file into a pandas DataFrame.
11 Supports CSV, TSV, and Parquet file formats.
12 """
13 def __call__(self, parser: argparse.ArgumentParser, namespace: argparse.Namespace, values: Tuple[str, str], option_string: str = None) -> None:
14 file_path, file_extension = values
15 file_extension = file_extension.lower()
16 if file_extension == "csv":
17 df = pd.read_csv(file_path)
18 elif file_extension in ["tsv", "tabular"]:
19 df = pd.read_csv(file_path, sep="\t")
20 elif file_extension == "parquet":
21 df = pd.read_parquet(file_path)
22 else:
23 raise ValueError(f"Unsupported file format: {file_extension}")
24 setattr(namespace, self.dest, df)
25
26
27 def mz_match(marker: np.ndarray, peak: np.ndarray, ppm: int) -> np.ndarray:
28 """
29 Check if the mass-to-charge ratio (m/z) of markers and peaks match within a given PPM tolerance.
30
31 Args:
32 marker (np.ndarray): Array of marker m/z values.
33 peak (np.ndarray): Array of peak m/z values.
34 ppm (int): PPM tolerance for matching.
35
36 Returns:
37 np.ndarray: Boolean array indicating matches.
38 """
8 return np.abs(marker - peak) <= ((peak + marker) / 2) * ppm * 1e-06 39 return np.abs(marker - peak) <= ((peak + marker) / 2) * ppm * 1e-06
9 40
10 41
11 def rt_match(marker, peak, tol): 42 def rt_match(marker: np.ndarray, peak: np.ndarray, tol: int) -> np.ndarray:
43 """
44 Check if the retention time (rt) of markers and peaks match within a given tolerance.
45
46 Args:
47 marker (np.ndarray): Array of marker retention times.
48 peak (np.ndarray): Array of peak retention times.
49 tol (int): Retention time tolerance for matching.
50
51 Returns:
52 np.ndarray: Boolean array indicating matches.
53 """
12 return np.abs(marker - peak) <= tol 54 return np.abs(marker - peak) <= tol
13 55
14 56
15 def find_matches(peaks, markers, ppm, rt_tol): 57 def find_matches(peaks: pd.DataFrame, markers: pd.DataFrame, ppm: int, rt_tol: int) -> pd.DataFrame:
58 """
59 Find matches between peaks and markers based on m/z and retention time tolerances.
60
61 Args:
62 peaks (pd.DataFrame): DataFrame containing peak data with 'mz' and 'rt' columns.
63 markers (pd.DataFrame): DataFrame containing marker data with 'mz' and 'rt' columns.
64 ppm (int): PPM tolerance for m/z matching.
65 rt_tol (int): Retention time tolerance for rt matching.
66
67 Returns:
68 pd.DataFrame: DataFrame containing matched rows with all columns from peaks and markers.
69 """
16 # Create a meshgrid of all combinations of mz and rt values 70 # Create a meshgrid of all combinations of mz and rt values
17 marker_mz = markers['mz'].values[:, np.newaxis] 71 marker_mz = markers['mz'].values[:, np.newaxis]
18 peak_mz = peaks['mz'].values 72 peak_mz = peaks['mz'].values
19 marker_rt = markers['rt'].values[:, np.newaxis] 73 marker_rt = markers['rt'].values[:, np.newaxis]
20 peak_rt = peaks['rt'].values 74 peak_rt = peaks['rt'].values
27 match_indices = np.where(mz_matches & rt_matches) 81 match_indices = np.where(mz_matches & rt_matches)
28 82
29 # Create a DataFrame of hits 83 # Create a DataFrame of hits
30 matched_markers = markers.iloc[match_indices[0]].reset_index(drop=True) 84 matched_markers = markers.iloc[match_indices[0]].reset_index(drop=True)
31 matched_peaks = peaks.iloc[match_indices[1]].reset_index(drop=True) 85 matched_peaks = peaks.iloc[match_indices[1]].reset_index(drop=True)
32 hits = pd.concat([matched_markers[['formula']].reset_index(drop=True), matched_peaks], axis=1)
33 86
34 # Calculate mz and rt differences 87 # Calculate mz and rt differences
35 hits['mz_diff'] = np.abs(matched_markers['mz'].values - matched_peaks['mz'].values) 88 matched_markers['mz_diff'] = np.abs(matched_markers['mz'].values - matched_peaks['mz'].values)
36 hits['rt_diff'] = np.abs(matched_markers['rt'].values - matched_peaks['rt'].values) 89 matched_markers['rt_diff'] = np.abs(matched_markers['rt'].values - matched_peaks['rt'].values)
37 90
91 # Drop mz and rt columns from the marker table
92 matched_markers = matched_markers.drop(columns=['mz', 'rt'])
93
94 # Combine all columns from peaks and markers
95 hits = pd.concat([matched_markers.reset_index(drop=True), matched_peaks.reset_index(drop=True)], axis=1)
38 return hits 96 return hits
39 97
40 98
41 def main(): 99 def main() -> None:
100 """
101 Main function to parse arguments, find matches between peaks and markers, and save the results.
102 """
42 parser = argparse.ArgumentParser(description='Find matches between peaks and markers.') 103 parser = argparse.ArgumentParser(description='Find matches between peaks and markers.')
43 parser.add_argument('--peaks', required=True, help='Path to the peaks parquet file.') 104 parser.add_argument('--peaks', required=True, nargs=2, action=LoadDataAction, help='Path to the peaks file and its format (e.g., "file.parquet parquet").')
44 parser.add_argument('--markers', required=True, help='Path to the markers CSV file.') 105 parser.add_argument('--markers', required=True, nargs=2, action=LoadDataAction, help='Path to the markers file and its format (e.g., "file.tsv tsv").')
45 parser.add_argument('--output', required=True, help='Path to the output TSV file.') 106 parser.add_argument('--output', required=True, help='Path to the output TSV file.')
46 parser.add_argument('--ppm', type=int, default=5, help='PPM tolerance for mz matching.') 107 parser.add_argument('--ppm', type=int, default=5, help='PPM tolerance for mz matching.')
47 parser.add_argument('--rt_tol', type=int, default=10, help='RT tolerance for rt matching.') 108 parser.add_argument('--rt_tol', type=int, default=10, help='RT tolerance for rt matching.')
48 args = parser.parse_args() 109 args = parser.parse_args()
49 110
50 peaks = pd.read_parquet(args.peaks) 111 hits = find_matches(args.peaks, args.markers, args.ppm, args.rt_tol)
51 markers = pd.read_csv(args.markers, sep='\t')
52
53 hits = find_matches(peaks, markers, args.ppm, args.rt_tol)
54 112
55 hits.to_csv(args.output, sep='\t', index=False) 113 hits.to_csv(args.output, sep='\t', index=False)
56 114
57 115
58 if __name__ == "__main__": 116 if __name__ == "__main__":