Merge pull request #587 from hechth/target_screen

added target screening tool

tools/misc/.shed.yml

@@ -13,3 +13,11 @@ repositories:
       - use_theoretical_mz_annotations.py
       - macros.xml
       - test-data
+  target_screen:
+    description: "Extract peaks from recetox-aplcms tables using a list of target ions"
+    long_description: |
+      "Extract peaks from recetox-aplcms tables using a list of target ions. MZ and RT tolerances are used to find matching features."
+    include:
+      - target_screen.xml
+      - target_screen.py
+      - test-data

tools/misc/target_screen.py

@@ -0,0 +1,59 @@
+import argparse
+
+import numpy as np
+import pandas as pd
+
+
+def mz_match(marker, peak, ppm):
+    return np.abs(marker - peak) <= ((peak + marker) / 2) * ppm * 1e-06
+
+
+def rt_match(marker, peak, tol):
+    return np.abs(marker - peak) <= tol
+
+
+def find_matches(peaks, markers, ppm, rt_tol):
+    # Create a meshgrid of all combinations of mz and rt values
+    marker_mz = markers['mz'].values[:, np.newaxis]
+    peak_mz = peaks['mz'].values
+    marker_rt = markers['rt'].values[:, np.newaxis]
+    peak_rt = peaks['rt'].values
+
+    # Calculate mz and rt matches
+    mz_matches = mz_match(marker_mz, peak_mz, ppm)
+    rt_matches = rt_match(marker_rt, peak_rt, rt_tol)
+
+    # Find the indices where both mz and rt match
+    match_indices = np.where(mz_matches & rt_matches)
+
+    # Create a DataFrame of hits
+    matched_markers = markers.iloc[match_indices[0]].reset_index(drop=True)
+    matched_peaks = peaks.iloc[match_indices[1]].reset_index(drop=True)
+    hits = pd.concat([matched_markers[['formula']].reset_index(drop=True), matched_peaks], axis=1)
+
+    # Calculate mz and rt differences
+    hits['mz_diff'] = np.abs(matched_markers['mz'].values - matched_peaks['mz'].values)
+    hits['rt_diff'] = np.abs(matched_markers['rt'].values - matched_peaks['rt'].values)
+
+    return hits
+
+
+def main():
+    parser = argparse.ArgumentParser(description='Find matches between peaks and markers.')
+    parser.add_argument('--peaks', required=True, help='Path to the peaks parquet file.')
+    parser.add_argument('--markers', required=True, help='Path to the markers CSV file.')
+    parser.add_argument('--output', required=True, help='Path to the output TSV file.')
+    parser.add_argument('--ppm', type=int, default=5, help='PPM tolerance for mz matching.')
+    parser.add_argument('--rt_tol', type=int, default=10, help='RT tolerance for rt matching.')
+    args = parser.parse_args()
+
+    peaks = pd.read_parquet(args.peaks)
+    markers = pd.read_csv(args.markers, sep='\t')
+
+    hits = find_matches(peaks, markers, args.ppm, args.rt_tol)
+
+    hits.to_csv(args.output, sep='\t', index=False)
+
+
+if __name__ == "__main__":
+    main()

tools/misc/target_screen.xml

@@ -0,0 +1,47 @@
+<tool id="target_screen" name="MS target screening" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="20.01" license="MIT">
+    <description>Extract peaks from recetox-aplcms tables using a list of target ions</description>
+    <macros>
+        <token name="@TOOL_VERSION@">0.1.0</token>
+        <token name="@VERSION_SUFFIX@">0</token>
+    </macros>
+
+    <requirements>
+        <requirement type="package" version="2.2.3">pandas</requirement>
+        <requirement type="package" version="17.0.0">pyarrow</requirement>
+    </requirements>
+
+    <command detect_errors="exit_code"><![CDATA[
+        python3 '${__tool_directory__}/target_screen.py' --peaks '$peaks' --markers '$markers' --output '$hits' --ppm $ppm --rt_tol $rt
+    ]]></command>
+
+    <inputs>
+        <param name="peaks" type="data" format="parquet"/>
+        <param name="markers" type="data" format="tabular"/>
+        <param name="ppm" type="integer" min="0" max="1000" value="10" label="ppm" help="Tolerance for peak filtering in ppm." />
+        <param name="rt" type="integer" min="0" max="100" value="10" label="rt tolerance" help="Toelrance regarding retention time to filter out peaks" />
+    </inputs>
+
+    <outputs>
+        <data name="hits" format="tabular" label="${tool.name} on ${on_string}" />
+    </outputs>
+
+    <tests>
+        <test>
+            <param name="peaks" value="target_screen/peaks.parquet"/>
+            <param name="markers" value="target_screen/markers.tsv"/>
+            <output name="hits" value="target_screen/out.tsv"/>
+        </test>
+    </tests>
+    <help><![CDATA[
+
+.. class:: infomark
+
+**What it does**
+
+This tool pulls out peaks from a table given a list of markers.
+The markers are matched based on m/z values with a specified ppm tolerance and matched based on retention time with a tolerance in units of retention time.
+    ]]></help>
+    <citations>
+        <citation type="doi">10.25080/Majora-92bf1922-00a</citation>
+    </citations>
+</tool>

tools/misc/test-data/target_screen/markers.tsv

@@ -0,0 +1,23 @@
+formula	mz	rt
+C8H6Cl2O3	218.9621	474.6
+C9H15N3O1	180.1142	458.23
+C5H2Cl3N1O1	195.9129	488.1
+C13H10O3	213.0557	508.40
+C13H9FO3	231.0463	521.48
+C6H5NO3	138.0197	166.56
+C6H4Cl1N1O2	155.9858	176.62
+C19H28N2O5S	395.1646	598.96
+C10H12N2O3S1	239.0496	312.55
+C4H11O3P1S1	169.0094	168.08
+C14H17Cl2NO2	300.0564	689.79
+C11H13ClO2	167.0633	572.93
+C12H4Cl2F6N4OS	434.9314	767.86
+C12H4Cl2F6N4O2S	450.9263	791.29
+C16H22ClN3O2	322.1328	706.5
+C16H11ClF6N2O	395.0391	741.93
+C10H11Cl1O3	213.0324	533.9
+C7H9NO2S	170.0281	363.59
+C12H7Cl3O2	286.9439	830.97
+C18H15Cl3O8	462.976	662.99
+C12H7Cl3O5S	366.9007	700.52
+C9H9N4Cl	207.0443	403.37

tools/misc/test-data/target_screen/out.tsv

@@ -0,0 +1,18 @@
+formula	mz	rt	sd1	sd2	area	mz_diff	rt_diff
+C8H6Cl2O3	218.9619738108278	473.4840709352675	0.6057217022739683	2.7706017478506073	1239147.63695882	0.00012618917219242576	1.1159290647325406
+C9H15N3O1	180.11422341297595	450.9460162486645	0.4692965104502825	4.727634916193644	1100073.3285644436	2.341297593488889e-05	7.2839837513355405
+C5H2Cl3N1O1	195.91267599889463	487.37949630118806	0.8695685392506757	2.8811688054510127	734461.0596300099	0.00022400110538001172	0.7205036988119673
+C13H10O3	213.05556658306853	508.4123751384482	2.9585968043814983	3.226731392934289	1787580.264815322	0.00013341693147594924	0.01237513844824889
+C13H9FO3	231.04576243564085	521.2436784813573	0.9930695671903609	2.469013097815558	1316270.081622402	0.0005375643591492008	0.23632151864273965
+C10H12N2O3S1	239.04945126090132	311.8317362000094	0.5578277726641567	3.57063615115722	3042462.634739455	0.0001487390986767423	0.7182637999906092
+C14H17Cl2NO2	300.0561299922103	685.3731548839577	0.8491884774374224	2.8491999009146074	1021277.4141378121	0.0002700077897088704	4.416845116042282
+C12H4Cl2F6N4OS	434.93037227267905	766.6610671335172	0.6265405149641161	3.55175113250731	43923382.478327975	0.0010277273209453597	1.198932866482778
+C12H4Cl2F6N4O2S	450.9259113906124	789.7479646306683	0.5765707513162325	3.4834377486718897	35843894.74749327	0.00038860938764173625	1.5420353693316429
+C16H22ClN3O2	322.13274143359513	705.9176130811956	0.765497607933695	2.9798451004946203	7686414.229962895	5.8566404845805664e-05	0.5823869188044455
+C16H11ClF6N2O	395.0387483584033	741.1840034426168	0.9150873601266857	2.396923077539685	692605.613740076	0.0003516415966942077	0.7459965573831369
+C10H11Cl1O3	213.03219616261535	532.8368925687558	0.8335128693984499	2.548404631638127	1231177.7029795102	0.00020383738464602175	1.0631074312441342
+C7H9NO2S	170.0280487596005	363.28514725405876	0.8844811055327363	2.7876246329523737	915161.3987675996	5.124039950032966e-05	0.3048527459412185
+C12H7Cl3O2	286.9434413572324	831.0018611928409	0.32058179843066653	1.7667251294853705	19934.364712896095	0.00045864276756901745	0.03186119284089273
+C18H15Cl3O8	462.97625391610677	662.6552310211961	0.9093786171678189	2.128435471267278	1209160.0005544876	0.00025391610677161225	0.3347689788039361
+C12H7Cl3O5S	366.90097256680355	699.9403505546061	0.8393755187990459	2.354260942300286	9578789.63215569	0.0002725668035736817	0.5796494453938976
+C9H9N4Cl	207.04420254367005	402.95120970553893	1.2647033563807812	2.594410018631832	40475158.16355405	9.74563299394049e-05	0.4187902944610755