MonoIsotopic mass for DIANN peptides with unknown AA

assets/multiqc_config.yml

@@ -1,7 +1,7 @@
 report_comment: >
-  This report has been generated by the <a href="https://github.com/nf-core/quantms/releases/tag/1.3.0" target="_blank">nf-core/quantms</a>
+  This report has been generated by the <a href="https://github.com/nf-core/quantms/tree/dev" target="_blank">nf-core/quantms</a>
   analysis pipeline. For information about how to interpret these results, please see the
-  <a href="https://nf-co.re/quantms/1.3.0/docs/output" target="_blank">documentation</a>.
+  <a href="https://nf-co.re/quantms/dev/docs/output" target="_blank">documentation</a>.
 report_section_order:
   pmultiqc:
     order: 1

bin/diann_convert.py

@@ -194,6 +194,42 @@ def get_exp_design_dfs(exp_design_file):
 
     return s_DataFrame, f_table
 
+def compute_mass_modified_peptide(peptide_seq: str) -> float:
+    """
+    Function that takes a peptide sequence including modifications and compute the mass using the AASequence class from
+    pyopenms. The notation of a peptidoform for pyopenms is the following:
+
+    if not modifications is present:
+      AVQVHQDTLRTMYFAXR -> AVQVHQDTLRTMYFAX[178.995499]R
+    if modification is present in Methionine:
+      AVQVHQDTLRTM(Oxidation)YFAXR -> AVQVHQDTLRTM(Oxidation)YFAX[178.995499]R
+
+    @param peptide_seq: str, peptide sequence
+    @return: float, mass of the peptide
+    """
+    peptide_parts: List[str] = []
+    not_mod = True
+    aa_mass = {
+        "X": "X[178.98493453312]",  # 196.995499 - 17.003288 - 1.00727646688
+        "U": "X[132.94306553312]",  # 150.95363  - 17.003288 - 1.00727646688
+        "O": "X[237.14773053312]",  # 255.158295 - 17.003288 - 1.00727646688
+    }
+    for aa in peptide_seq:
+        # Check if the letter is in aminoacid
+        if aa == "(":
+            not_mod = False
+        elif aa == ")":
+            not_mod = True
+        # Check aminoacid letter
+        if aa in aa_mass and not_mod:
+            aa = aa_mass[aa]
+        elif aa not in ['G','A','V','L','I','F','M','P','W','S','C','T','Y','N','Q','D','E','K','R','H'] and not_mod and aa != ")":
+            logger.info(f"Unknown amino acid with mass not known:{aa}")
+        peptide_parts.append(aa)
+    new_peptide_seq = ''.join(peptide_parts)
+    mass = AASequence.fromString(new_peptide_seq).getMonoWeight()
+    logger.debug(new_peptide_seq + ":" + str(mass))
+    return mass
 
 class DiannDirectory:
     def __init__(self, base_path, diann_version_file):
@@ -348,7 +384,7 @@ def main_report_df(self, qvalue_threshold: float) -> pd.DataFrame:
         logger.debug("Calculating Precursor.Mz")
         # Making the map is 10x faster, and includes the mass of
         # the modification. with respect to the previous implementation.
-        uniq_masses = {k: AASequence.fromString(k).getMonoWeight() for k in report["Modified.Sequence"].unique()}
+        uniq_masses = {k: compute_mass_modified_peptide(k) for k in report["Modified.Sequence"].unique()}
         mass_vector = report["Modified.Sequence"].map(uniq_masses)
         report["Calculate.Precursor.Mz"] = (mass_vector + (PROTON_MASS_U * report["Precursor.Charge"])) / report[
             "Precursor.Charge"

conf/modules.config

@@ -195,12 +195,12 @@ process {
 
     // PROTEINQUANTIFIER
     withName: '.*:TMT:PROTEINQUANT:PROTEINQUANTIFIER' {
-        ext.args    = "-debug 0"
+        ext.args    = "-debug $params.protein_quant_debug"
     }
 
     // MSSTATSCONVERTER
     withName: '.*:TMT:PROTEINQUANT:MSSTATSCONVERTER' {
-        ext.args    = "-debug 0"
+        ext.args    = "-debug $params.protein_quant_debug"
     }
 }
 

docs/usage.md

@@ -11,7 +11,7 @@
 The typical command for running the pipeline is as follows:
 
 ```bash
-nextflow run nf-core/quantms --input '/url/path/to/your/experimentX_design.tsv' --database '/url/path/to/your/proteindatabase.fasta' --outdir './results' -profile docker
+nextflow run nf-core/quantms --input '/url/path/to/your/experiment_design.sdrf.tsv' --database '/url/path/to/your/proteindatabase.fasta' --outdir './results' -profile docker
 ```
 
 where the experimental design file has to be one of:

nextflow.config

@@ -44,6 +44,7 @@ params {
     luciphor_debug          = 0
     protein_inference_debug = 0
     plfq_debug              = 0
+    protein_quant_debug     = 0
 
     // decoys
     decoy_string                        = 'DECOY_'
@@ -462,7 +463,7 @@ manifest {
     description     = """Quantitative Mass Spectrometry nf-core workflow"""
     mainScript      = 'main.nf'
     nextflowVersion = '!>=23.04.0'
-    version         = '1.3.0'
+    version         = '1.3.0dev'
     doi             = '10.5281/zenodo.7754148'
 }
 

nextflow_schema.json

@@ -828,6 +828,13 @@
                     "description": "Export the results in mzTab format.",
                     "default": true,
                     "fa_icon": "fas fa-check-square"
+                },
+                "protein_quant_debug": {
+                    "type": "integer",
+                    "description": "Debug level for the protein quantification step. Increase for verbose logging",
+                    "fa_icon": "fas fa-bug",
+                    "default": 0,
+                    "hidden": true
                 }
             },
             "fa_icon": "fas fa-braille"