tandem-mass-spectra-peptide-matching
Summary
Match centroided tandem mass spectra (MS/MS) against peptidic natural product (PNP) structure databases using Dereplicator, VarQuest, or Dereplicator+ to identify known or novel peptidic metabolites with quantified confidence scores and false discovery rates. This skill enables dereplication and variant detection across multiple search modes.
When to use
You have centroided LC-MS/MS spectra (in MGF, mzXML, mzML, or mzData format) and wish to identify peptidic natural products or ribosomally synthesized and post-translationally modified peptides (RiPPs) against a known structure database or custom RiPP structure database. Use this skill when you need to compare hit patterns across multiple database search pipelines, detect modification-tolerant variants, or generate significant_matches.tsv outputs with scan identifiers, match scores, and p-values for downstream cross-validation.
When NOT to use
- Input spectra are not centroided or are in unsupported binary formats without conversion capability
- Your goal is to identify non-peptidic metabolites and you do not have access to Dereplicator+ (use Dereplicator+ instead of Dereplicator or VarQuest alone)
- You lack a suitable structure database or RiPP reference set tailored to your organism/sample type
Inputs
- centroided LC-MS/MS spectra files (MGF, mzXML, mzML, or mzData format)
- chemical structure database of known natural products (for Dereplicator, VarQuest, Dereplicator+)
- RiPP structure database (for MetaMiner)
- optional: RiPP sequence file (FASTA format) for MetaMiner
Outputs
- significant_matches.tsv (tab-separated file with scan identifiers, match scores, p-values, and false discovery rates)
- match results including tool-specific hit sets and common identifications
- spectral networking graphs (optional, for RiPP propagation visualization)
How to apply
Prepare centroided spectra in MGF format (convert non-native formats using ProteoWizard msconvert if needed). Select the appropriate NPDtools pipeline: Dereplicator for standard PNP database search, VarQuest for modification-tolerant variant detection, or Dereplicator+ for non-peptidic metabolite identification. Execute the chosen pipeline with the spectra and structure database as inputs, using standard parameters (or --blind flag for arbitrary post-translational modification search in MetaMiner). Parse the resulting significant_matches.tsv output to extract scan identifiers, match scores, p-values, and false discovery rates. For comparative analysis, run all three pipelines on identical test spectra and cross-tabulate results to identify tool-specific detections, common hits, and mode-dependent sensitivity differences.
Related tools
- Dereplicator (database search pipeline for identification of peptidic natural products through tandem mass spectra matching against known PNP structures) — https://github.com/ablab/npdtools
- VarQuest (modification-tolerant database search pipeline for identification of novel variants of PNPs with PTM tolerance) — https://github.com/ablab/npdtools
- Dereplicator+ (extended database search pipeline for identification of both peptidic and non-peptidic metabolites through tandem mass spectra matching) — https://github.com/ablab/npdtools
- MetaMiner (metabologenomic pipeline that integrates tandem mass spectra matching with genomic data to identify RiPPs; executes Dereplicator internally with optional --blind flag for arbitrary PTM search) — https://github.com/mohimanilab/MetaMiner
- ProteoWizard msconvert (format conversion utility to convert non-native spectrum formats (e.g., .mzML) to MGF for NPDtools compatibility)
Examples
python dereplicator.py test_spectra.mgf -d RiPP_structure_db.txt -o dereplicator_output/ && python varquest.py test_spectra.mgf -d RiPP_structure_db.txt -o varquest_output/
Evaluation signals
- significant_matches.tsv contains non-empty rows with valid scan identifiers, numeric match scores, and p-values within expected ranges (e.g., p < 0.05 for significant matches)
- false discovery rate (FDR) values are computed and reported for each match, indicating statistical significance filtering was applied
- cross-pipeline comparison identifies expected overlaps and tool-specific detections; Dereplicator+ should detect at least as many hits as Dereplicator alone on the same inputs
- spectral network output (if generated) shows connected components with appropriate cosine similarity thresholds, validating propagation-based RiPP enlargement
- no parsing errors or missing columns in significant_matches.tsv; all spectra from input files are represented in output with no truncation
Limitations
- Requires spectra to be centroided; profile mode spectra must be preprocessed or conversion may fail or produce unreliable results
- antiSMASH output (.final.gbk) may fail to detect certain biosynthetic gene clusters (e.g., AmfS) where raw contig.fasta succeeds; genome input format selection affects MetaMiner sensitivity
- No changelog available in NPDtools 2.5.0 documentation; version-specific behavior changes between releases are not formally documented
- Spectral network visualization requires matplotlib and networkx Python libraries; plain-text output only if these dependencies are absent
- VarQuest and MetaMiner --blind flag increase computational cost and may reduce specificity due to expanded PTM search space
Evidence
- [other] NPDtools version 2.5.0 includes multiple database search pipelines (Dereplicator, VarQuest, Dereplicator+) within its toolkit for natural product mass spectrometry analysis.: "NPDtools version 2.5.0 includes multiple database search pipelines (Dereplicator, VarQuest, Dereplicator+) within its toolkit"
- [readme] Spectra files must be centroided and be in an open spectrum format (MGF, mzXML, mzML or mzData). NPDtools natively supports MGF and mzXML/mzData and uses msconvert utility from ProteoWizard to convert spectra in other formats to MGF.: "Spectra files must be centroided and be in an open spectrum format (MGF, mzXML, mzML or mzData). NPDtools natively supports MGF"
- [other] Execute Dereplicator pipeline on the same test spectra and RiPP structure database, output match results. Execute VarQuest pipeline on the same inputs with standard parameters. Parse and tabulate significant_matches.tsv outputs from all three tools, recording scan identifiers, match scores, p-values, and false discovery rates.: "Parse and tabulate significant_matches.tsv outputs from all three tools, recording scan identifiers, match scores, p-values, and false discovery rates."
- [other] Cross-compare hit sets to identify tool-specific detections, common identifications, and mode-dependent sensitivity differences.: "Cross-compare hit sets to identify tool-specific detections, common identifications, and mode-dependent sensitivity differences."
- [readme] MetaMiner is a metabologenomic pipeline which integrates metabolomic (tandem mass spectra) and genomic data to identify novel RiPPs: "MetaMiner is a metabologenomic pipeline which integrates metabolomic (tandem mass spectra) and genomic data to identify novel RiPPs"
- [other] While
MetaMiner successfully detect AmfS using the contigs.fasta file, it fails with antiSMASH result as input: "While MetaMiner successfully detect AmfS using the contigs.fasta file, it fails with antiSMASH result as input"
1---2name: tandem-mass-spectra-peptide-matching3description: Use when you have centroided LC-MS/MS spectra (in MGF, mzXML, mzML, or mzData format) and wish to identify peptidic natural products or ribosomally synthesized and post-translationally modified peptides (RiPPs) against a known structure database or custom RiPP structure database.4license: CC-BY-4.05---67# tandem-mass-spectra-peptide-matching89## Summary1011Match centroided tandem mass spectra (MS/MS) against peptidic natural product (PNP) structure databases using Dereplicator, VarQuest, or Dereplicator+ to identify known or novel peptidic metabolites with quantified confidence scores and false discovery rates. This skill enables dereplication and variant detection across multiple search modes.1213## When to use1415You have centroided LC-MS/MS spectra (in MGF, mzXML, mzML, or mzData format) and wish to identify peptidic natural products or ribosomally synthesized and post-translationally modified peptides (RiPPs) against a known structure database or custom RiPP structure database. Use this skill when you need to compare hit patterns across multiple database search pipelines, detect modification-tolerant variants, or generate significant_matches.tsv outputs with scan identifiers, match scores, and p-values for downstream cross-validation.1617## When NOT to use1819- Input spectra are not centroided or are in unsupported binary formats without conversion capability20- Your goal is to identify non-peptidic metabolites and you do not have access to Dereplicator+ (use Dereplicator+ instead of Dereplicator or VarQuest alone)21- You lack a suitable structure database or RiPP reference set tailored to your organism/sample type2223## Inputs2425- centroided LC-MS/MS spectra files (MGF, mzXML, mzML, or mzData format)26- chemical structure database of known natural products (for Dereplicator, VarQuest, Dereplicator+)27- RiPP structure database (for MetaMiner)28- optional: RiPP sequence file (FASTA format) for MetaMiner2930## Outputs3132- significant_matches.tsv (tab-separated file with scan identifiers, match scores, p-values, and false discovery rates)33- match results including tool-specific hit sets and common identifications34- spectral networking graphs (optional, for RiPP propagation visualization)3536## How to apply3738Prepare centroided spectra in MGF format (convert non-native formats using ProteoWizard msconvert if needed). Select the appropriate NPDtools pipeline: Dereplicator for standard PNP database search, VarQuest for modification-tolerant variant detection, or Dereplicator+ for non-peptidic metabolite identification. Execute the chosen pipeline with the spectra and structure database as inputs, using standard parameters (or --blind flag for arbitrary post-translational modification search in MetaMiner). Parse the resulting significant_matches.tsv output to extract scan identifiers, match scores, p-values, and false discovery rates. For comparative analysis, run all three pipelines on identical test spectra and cross-tabulate results to identify tool-specific detections, common hits, and mode-dependent sensitivity differences.3940## Related tools4142- **Dereplicator** (database search pipeline for identification of peptidic natural products through tandem mass spectra matching against known PNP structures) — https://github.com/ablab/npdtools43- **VarQuest** (modification-tolerant database search pipeline for identification of novel variants of PNPs with PTM tolerance) — https://github.com/ablab/npdtools44- **Dereplicator+** (extended database search pipeline for identification of both peptidic and non-peptidic metabolites through tandem mass spectra matching) — https://github.com/ablab/npdtools45- **MetaMiner** (metabologenomic pipeline that integrates tandem mass spectra matching with genomic data to identify RiPPs; executes Dereplicator internally with optional --blind flag for arbitrary PTM search) — https://github.com/mohimanilab/MetaMiner46- **ProteoWizard msconvert** (format conversion utility to convert non-native spectrum formats (e.g., .mzML) to MGF for NPDtools compatibility)4748## Examples4950```51python dereplicator.py test_spectra.mgf -d RiPP_structure_db.txt -o dereplicator_output/ && python varquest.py test_spectra.mgf -d RiPP_structure_db.txt -o varquest_output/52```5354## Evaluation signals5556- significant_matches.tsv contains non-empty rows with valid scan identifiers, numeric match scores, and p-values within expected ranges (e.g., p < 0.05 for significant matches)57- false discovery rate (FDR) values are computed and reported for each match, indicating statistical significance filtering was applied58- cross-pipeline comparison identifies expected overlaps and tool-specific detections; Dereplicator+ should detect at least as many hits as Dereplicator alone on the same inputs59- spectral network output (if generated) shows connected components with appropriate cosine similarity thresholds, validating propagation-based RiPP enlargement60- no parsing errors or missing columns in significant_matches.tsv; all spectra from input files are represented in output with no truncation6162## Limitations6364- Requires spectra to be centroided; profile mode spectra must be preprocessed or conversion may fail or produce unreliable results65- antiSMASH output (.final.gbk) may fail to detect certain biosynthetic gene clusters (e.g., AmfS) where raw contig.fasta succeeds; genome input format selection affects MetaMiner sensitivity66- No changelog available in NPDtools 2.5.0 documentation; version-specific behavior changes between releases are not formally documented67- Spectral network visualization requires matplotlib and networkx Python libraries; plain-text output only if these dependencies are absent68- VarQuest and MetaMiner --blind flag increase computational cost and may reduce specificity due to expanded PTM search space6970## Evidence7172- [other] NPDtools version 2.5.0 includes multiple database search pipelines (Dereplicator, VarQuest, Dereplicator+) within its toolkit for natural product mass spectrometry analysis.: "NPDtools version 2.5.0 includes multiple database search pipelines (Dereplicator, VarQuest, Dereplicator+) within its toolkit"73- [readme] Spectra files must be centroided and be in an open spectrum format (MGF, mzXML, mzML or mzData). NPDtools natively supports MGF and mzXML/mzData and uses msconvert utility from ProteoWizard to convert spectra in other formats to MGF.: "Spectra files must be centroided and be in an open spectrum format (**MGF**, **mzXML**, **mzML** or **mzData**). NPDtools natively supports MGF"74- [other] Execute Dereplicator pipeline on the same test spectra and RiPP structure database, output match results. Execute VarQuest pipeline on the same inputs with standard parameters. Parse and tabulate significant_matches.tsv outputs from all three tools, recording scan identifiers, match scores, p-values, and false discovery rates.: "Parse and tabulate significant_matches.tsv outputs from all three tools, recording scan identifiers, match scores, p-values, and false discovery rates."75- [other] Cross-compare hit sets to identify tool-specific detections, common identifications, and mode-dependent sensitivity differences.: "Cross-compare hit sets to identify tool-specific detections, common identifications, and mode-dependent sensitivity differences."76- [readme] MetaMiner is a metabologenomic pipeline which integrates metabolomic (tandem mass spectra) and genomic data to identify novel RiPPs: "MetaMiner is a metabologenomic pipeline which integrates metabolomic (tandem mass spectra) and genomic data to identify novel RiPPs"77- [other] While `MetaMiner` successfully detect AmfS using the `contigs.fasta` file, it fails with antiSMASH result as input: "While `MetaMiner` successfully detect AmfS using the `contigs.fasta` file, it fails with antiSMASH result as input"
HolobiomicsLab 