isolib: strip terminal polarity markers from formulas before isotopic parsing (#758)

* Initial plan

* Handle polarity symbols in isolib formulas

* Refine polarity stripping to terminal charge markers

* Cover additional terminal charge notation forms

* Restrict polarity stripping to sign-led suffixes

* Update isolib version from galaxy3 to galaxy4

* Modify output structure for isotope library test

* Change expected column count from 4 to 9

* Update expected line count in isotope_library output

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: Helge Hecht <helge.hecht@recetox.muni.cz>

Author: Copilot

tools/isolib/isolib.R

@@ -46,6 +46,10 @@ parse_args <- function() {
 
     # Extract selected columns
     compound_table <- compound_table_full[, intersect(c("name", "formula", "rt"), colnames(compound_table_full)), drop = FALSE]
+    compound_table$formula <- stringr::str_remove(
+        compound_table$formula,
+        "\\s*[+-]\\d*$"
+    )
 
     # Extract remaining columns
     remaining_columns <- setdiff(colnames(compound_table_full), colnames(compound_table))

tools/isolib/isolib.xml

@@ -1,4 +1,4 @@
-<tool id="isolib" name="isolib" version="2.6+galaxy3" profile="23.0" license="MIT">
+<tool id="isolib" name="isolib" version="2.6+galaxy4" profile="23.0" license="MIT">
     <description>create an isotopic pattern library for given compounds and adducts based on enviPat</description>
     <creator>
         <person
@@ -236,6 +236,21 @@
             </conditional>
             <output name="isotope_library" file="test9.tsv" ftype="tabular"/>
         </test>
+        <!-- Test 11: Formulas with polarity symbols -->
+        <test>
+            <param name="input_file" value="markers_no_rt_polarity.tsv"/>
+            <param name="formatting|out_format" value="tabular"/>
+            <conditional name="rel_to">
+                <param name="value" value="1"/>
+                <param name="threshold" value="1.0"/>
+            </conditional>
+            <output name="isotope_library">
+                <assert_contents>
+                    <has_n_columns n="9"/>
+                    <has_n_lines n="20" delta="2"/>
+                </assert_contents>
+            </output>
+        </test>
     </tests>
     <help><![CDATA[
         This tool computes isotopic patterns for given compounds and adduct forms.

tools/isolib/test-data/markers_no_rt_polarity.tsv

@@ -0,0 +1,6 @@
+formula	name
+C8H6Cl2O3+	2,4-Dichlorophenoxyacetic acid ou 2,4-D
+C9H15N3O1-	2-diethylamino-6-methyl pyrimidin-4-ol/one
+C5H2Cl3N1O1+	3,5,6-Trichloro-2-pyridinol
+C13H10O3-	3-phenoxybenzoic acid
+C13H9FO3+	4-Fluoro-3-phenoxybenzoic acid