Claude Skills by HolobiomicsLab

Use when you have (1) a peaklist from untargeted LC/HRMS analysis with m/z and intensity values (typically output from IDSL.IPA or similar peak-picking tools), (2) a need to assign molecular formulas to detected peaks, and (3) only MS1 spectral data available (MS/MS is not required).

Use when when you need to generate synthetic LC-MS/MS data to test fragmentation strategies, validate acquisition controllers, or benchmark peak-picking and spectral matching algorithms before deployment on real mass spectrometers.

Use when after generating a scan index from a Thermo Fisher Orbitrap raw file using readIndex(), apply this skill when your analysis goal requires working exclusively with MS1 (precursor) scans rather than tandem MS (MS2/MS3) spectra.

Use when you have MS2 tandem mass spectrometry data in .mzML format and need to match unknown spectra against a reference library (GNPS, HMDB, MassBank) to identify which known compounds are present in your sample.

Use when you have centroided MS2 spectra from data-dependent LC- or GC-HRMS measurements and need to rapidly prioritize potential PFAS features within a larger feature set.

Use when after aligning MS2 fingerprints across samples (generating a sample-by-feature occurrence matrix) and before applying visualization techniques (MDS/PCoA, TMAP, Heatmap) or statistical comparisons.

Use when after MS2 fingerprints have been generated by counting MS2 peaks and neutral losses in each sample, and you have aligned them into a MemoMatrix (sample-by-fingerprint matrix).

Use when you have two MemoMatrix objects generated from separate sample cohorts (e.

Use when you have LC-MS/MS data in mzML, mzXML, or MGF format from one or more metabolomics samples and need to compare samples that may have poor overlap in detected features, strong retention time shifts between runs, or were acquired on different LC methods or MS technologies (e.

Use when you have extracted MS/MS spectra for a given metabolomic feature across multiple replicates (e.g., after top-TIC filtering) and need to identify which fragments are reproducible across replicates.

Use when you have data-dependent acquisition (DDA) MS2 spectra from HRMS measurements (ESI or APCI ionization) and need to prioritize potential PFAS features from a large pool of detected ions.

Use when you have raw MS2 spectra in common formats (mzML, mzXML, msp, MGF, JSON) and need to convert them into normalized, queryable spectral objects for downstream analysis such as MS2 fingerprinting, spectral similarity scoring, or sample comparison.

Use when you have raw MS2 spectral data (MGF, mzML, or msp format) and need to generate a sample-level fingerprint for comparison across metabolomics samples, especially when samples were acquired using different LC methods, mass spectrometer technologies, or exhibit poor feature overlap or large.

Use when when you have a resolved spectrum file in mzML or mzXML format and need to identify where MS2 (tandem mass spectrometry) scans occur within an LC-MS run, particularly to annotate XIC displays with MS2 precursor positions or validate that MS2 events align with expected chromatographic.

Use when when beginning a MEMO analysis workflow with raw or unaligned MS2 spectra files and needing to extract fragmentation data and precursor information before counting MS2 peaks and neutral losses to generate sample fingerprints.

Use when after concatenating replicate MS/MS spectra for each precursor feature (m/z and retention time), use this skill when you have multiple replicate scans per feature and need to reduce spectral count while preserving the highest-intensity, most-reliable spectra.

Use when you have acquired LC-IMS-MS/MS data (or equivalent multidimensional MS/MS acquisition) in mzML or mzML.gz format and need to disambiguate overlapping fragmentation spectra arising from co-eluting or co-drifting precursor ions.

Use when you have MS2 fragmentation spectra from multiple metabolomics samples and need to identify samples with unusual spectral profiles that may indicate novel chemistry, independent of feature abundance or annotation status.

Use when you have LC-MS/MS metabolomics data in MGF format and need to prepare it for Latent Dirichlet Allocation (LDA) topic modeling.

Use when you have extracted an MS1 feature table (from XCMS or custom CSV with m/z, retention time, and intensity columns) and one or more DDA mzXML files from the same or related LC-MS runs, and you need to annotate features with MS2 spectral data and match them against a standard spectral library.

Use when you have two or more LC-MS/MS datasets (in mzML, mzXML, or MGF format) and need to quantify their overall spectral content similarity—typical scenarios include data quality control, species identification, molecular phylogenetics, or proteome comparison across samples, cell lines, or.

Use when after temporal intensity profile correlation and exact mass difference refinement have identified candidate ion-species pairs in direct-injection plasma ionization MS data (e.g., DBDI-MS, DBDI-FT-ICR-MS).

Use when you have DDA LC-MS/MS raw data (mzML format) with detected chromatographic peaks at a target m/z value and retention time window, and you need to build a high-confidence MS2 consensus spectrum for that peak to match against reference spectra (e.g., Metlin, GNPS).

Use when after sample alignment and feature grouping steps in untargeted LC-MS workflows, when you have DDA-mode raw files with both MS1 and MS2 scans and need to link tandem mass spectra to quantified features for annotation and structural characterization.

Use when when you have DDA LC-MS/MS data (mzML format) with identified chromatographic peaks at a specific m/z (e.g., 304.1131) and multiple MS2 spectra fragmented from that precursor, and you need to produce a single high-confidence MS2 spectrum for comparison against reference databases (e.

Use when you have DDA (data-dependent acquisition) LC-MS/MS data with MS2 spectra and want to discover metabolic features that may be missed by MS1-only peak picking, or when you need an alternative feature extraction workflow that leverages fragmentation patterns to identify true metabolite.

Use when you have raw or unstructured MS2 spectral data (from untargeted tandem mass spectrometry experiments) and plan to run MS2MP inference for KEGG pathway prediction.

Use when when you have raw MS2 spectra (m/z and intensity pairs) and need to compute a Probability Product Kernel score or other fragmentation-based similarity metric against a training dataset of known spectra (e.g., GNPS library spectra with structural annotations).

Use when after creating a GNPS mass spectral molecular network and running an MS2LDA experiment, use this skill when you want to identify and visualize which substructural motifs (Mass2Motifs) are shared across clustered spectra, particularly to highlight fragmentation pattern similarities between.

Use when you have a GNPS-generated molecular network (either classical or feature-based) and corresponding MS2LDA experiment output containing Mass2Motif-to-spectrum assignments, and you want to visualize which substructural motifs are shared across clusters or features and how they distribute.

Use when when you have a GNPS molecular network (classical or feature-based) and corresponding MS2LDA experiment results, and you want to annotate network nodes with discovered substructural motifs to support structural elucidation and chemical family interpretation.

Use when you have a GNPS molecular network (in GML or GraphML format) and a completed MS2LDA experiment on ms2lda.org, and you want to annotate network nodes with detected substructural motifs to identify shared fragmentation patterns and structural classes across molecular families.

Use when when you have vendor-format LC-MS acquisition files (Thermo .raw, Bruker .d, Sciex .ms, Agilent, Waters formats) that need to be ingested into a quality control or data processing pipeline that requires open, standardized spectral formats.

Use when you have acquired raw mass spectrometry data from ThermoFisher, Agilent, or compatible vendors in their native formats (.raw, .d, or equivalent) and need to prepare it for nontargeted analysis using MSThunder on a Linux system.

Use when training embeddings from multi-modal spectral data (peak information + metadata) where you need to ensure both contrastive discriminability AND reconstruction fidelity. Specifically use it in transformer-based architectures that produce embeddings from heterogeneous inputs (e.

Use when you have multiple centroided .mzML LC-MS files that need to be loaded into a unified object for targeted peak integration, and you need to distinguish QC runs from sample runs to compute per-group quality metrics (e.g., average SNR, peak correlation, area under curve per QC cohort).

Use when after MS-DIAL has completed feature detection and peak alignment on .mzML LC-HRMS data, producing an aligned feature table. Use MSFLO when you need to assign metabolite identities to detected features and filter results by significance criteria before downstream interpretation.

Use when you have MSI intensity data exported from MSiReader, SCiLS, or Cardinal as plain-text CSV files or as Cardinal MSProcessedImagingExperiment/MSContinuousImagingExperiment objects, and need to load it into R as a validated msimat object for mass difference and adduct analysis.

Use when when you have a target compound (modified or unmodified) and need to obtain its experimental MS/MS spectrum and metadata to serve as a known reference for ModiFinder analysis, or when benchmarking evaluation methods like average_distance scoring.

Use when removing invalid or malformed entries (e.g., SMILES validation, format errors) from large spectral datasets (GNPS, MoNA, MTBLS1572, MassBank).

Use when you have acquired EI or MS2 library files in MSP format (e.g., from NIST via Lib2NIST export, RIKEN, MoNA, SWGDRUG, or GNPS) and need to read them into R to assign SMILES, retention indices, or combine multiple libraries into a single consolidated MSP file for MS-DIAL.

Use when you have a .msp format MS/MS spectrum library (e.g., from MassBank or similar public databases) and need to convert it into individual CSV entries indexed by positive or negative ionisation mode for use as a custom fragment library in MetaboAnnotatoR annotation pipelines.

Use when you have statistically significant features from multiple LC-MS assays with different ionization modes (e.g., positive and negative ESI) and need to collapse redundant feature representations into single structural entities.

Use when you have independent LC-MS assays (e.g., positive and negative ionization modes, different lipid profiling assays, or different chromatographic methods) analyzed on the same sample cohort and want to integrate them into a single discriminant or regression model without losing assay-level.

Use when your metabolomics experiment includes samples acquired across multiple instrument runs, different preparation dates, or distinct sample cohorts.

Use when when you have LC-MS normalized intracellular metabolite abundance measurements for multiple cell lines and need to estimate reaction activity driven by substrate availability rather than enzyme expression alone.

Use when you have raw LC-MS metabolomics data from multiple disease groups and need to classify new or existing samples into discrete disease categories.

Use when you have (1) metabolite-disease correlation matrices and protein association predictions generated by a deep learning module (e.

Use when you have completed LC–MS/MS feature detection and annotation (via MZmine2/3 and GNPS/SIRIUS/CANOPUS), have compiled a taxonomically annotated metadata table, and need to rank extracts by their likelihood of containing novel chemistry.

Use when you have an LC-HRMS feature table (m/z, retention time, isotope ratios, fragmentation patterns) and a suspect compound database with reference properties (m/z, expected RT windows, theoretical isotope ratios, characteristic neutral losses), and you need to rank which features most likely.