Claude Skills by HolobiomicsLab

Use when you have centroided LC-MS/MS spectral data (in MGF, mzXML, mzML, or mzData format) and want to identify known or predicted natural product structures present in your sample.

Use when you have an unknown MS/MS spectrum (m/z and intensity pairs) from positive-mode ionization and need to identify the most likely molecular formula and adduct type (e.g., [M+H]+, [M+Na]+) when no reference library match is available or desirable.

Use when you have cleaned MS/MS spectra (in formats like .mgf, .msp, .

Use when you have raw GC-MS data in netCDF or vendor-specific binary format and need to separate co-eluting compounds and extract clean mass spectra for each individual chemical component prior to molecular networking, spectral matching, or metabolite identification workflows.

Use when you have an experimental MS/MS spectrum (e.g., from MassBank or acquired data) for a single metabolite with known accurate precursor m/z and adduct type, and you need to generate a library entry with scored fragments for use in metabolite annotation pipelines.

Use when when working with multidimensional MS data (LC–IM–MS/MS) converted to MZA format where spectra are stored in jagged arrays with m/z values distributed across individual HDF5 datasets per scan, and you need to ensure m/z consistency for downstream peak detection, isotope analysis, or.

Use when you have raw tandem mass spectra data (mz/intensity pairs and precursor m/z values) and need to train interpretable machine learning models (regression or tree-based) where feature interpretability and direct chemical meaning are required.

Use when you have raw MS/MS spectra in MSP format or as numpy arrays and need to standardize them for comparison or library matching. Specifically, use it before performing electronic or chemical denoising, or before computing entropy-similarity metrics between query and reference spectra.

Use when after generating electronic noise (uniformly sampled m/z with Poisson-distributed intensities) and chemical noise (formula database-sampled m/z with Poisson intensities) and you need to combine both noise types with a clean baseline spectrum into a single unified peak array.

Use when you have a collection of molecular structures (SMILES or chemical graphs) with paired experimental tandem mass spectra and want to build or benchmark a predictive model.

Use when when you have molecular structures (SMILES, InChI, or chemical formula) and need to predict their tandem mass spectra for structural elucidation or compound ranking against databases. Use SCARF when operating at the chemical formula level;

Use when you have raw MS/MS spectra in MGF format with variable peak quality, mixed charge states, or instrument artifacts that could confound clustering or similarity measures.

Use when you have raw mass spectrometry data files from a Thermo instrument (e.

Use when when you have an unknown compound's mass spectrum (m/z peaks and intensities in .mgf or equivalent format) and need to identify structurally related metabolites from a reference database by computing similarity in learned semantic space rather than direct spectral matching.

Use when when you have a query MS/MS spectrum (m/z and intensity pairs) that you need to match against a library of reference spectra, and you want to identify the -matching library entry while accounting for unmatched peaks that may indicate spectral contamination or chimerism.

Use when you have an experimental tandem mass spectrum (collision-induced dissociation, CID) and a known chemical formula (or narrow set of candidate formulas), and you need to identify the most likely structure(s) by ranking against a large candidate library such as PubChem.

Use when after filtering and cleaning MS/MS spectra (positive/negative ion mode) but before applying Latent Dirichlet Allocation for Mass2Motif discovery.

Use when you have an annotated MsmsSpectrum object (with fragment assignments via ProForma 2.0) and need to produce a high-resolution, static PNG figure showing both the observed spectrum and color-highlighted fragment ion matches for inclusion in a manuscript or supplementary materials.

Use when when you have extracted m/z and intensity arrays from an MZA file (or similar HDF5-backed MS data structure) and need to visually inspect a single MS1 or MS2 spectrum, verify peak characteristics, or diagnose data quality issues before downstream analysis (peak fitting, isotope pattern.

Use when after generating theoretical B/Y ion spectra or after importing experimental MS/MS scans when your analysis goal requires restricting the ion population to a specific m/z window (e.g., m/z < 2000).

Use when you are preparing to align two or more nontargeted LCMS datasets from the same analytical method using Eclipse and need to determine the m/z tolerance window.

Use when you have an LC-MS peak-intensity matrix (rows = peaks with m/z and intensity; columns = samples) and need to assign KEGG compound identifiers to observed peaks.

Use when you have centroided data-dependent acquisition (DDA) mzML files from LC- or GC-HRMS measurements and need to convert continuous raw mass spectrometric signals into discrete, quantifiable chromatographic features (m/z, RT, intensity, charge, isotope) before PFAS-specific prioritization or.

Use when you have statistically significant LC-MS features and need to group them into structural clusters. Specifically, use it after selecting features by p-value threshold (e.g., p < 0.

Use when when linking statistically significant LC-MS features into structural clusters based on adduct signatures and cross-assay references (e.g., [M+H]+/[M-H]−), and you need to specify the maximum allowed deviation (in ppm) between observed m/z values and calculated neutral masses.

Use when when you have observed fragment peak m/z values from tandem mass spectra and need to assign chemical subformulae to them.

Use when you have an MS/MS spectrum with observed peaks and a peptidoform specification (e.g., ProForma 2.0 notation such as 'EM[Oxidation]EVEES[Phospho]PEK'), and you need to annotate which observed peaks correspond to known fragment ions.

Use when you have experimental peak lists (m/z, retention time, intensity) from peak-picking software (MZmine, XCMS, MS-DIAL, or Compound Discoverer) and need to match them against a simulated lipid fragment library (500,000+ lipid species).

Use when when implementing adduct detection in LC-MS metabolomics workflows, after defining theoretical adduct mass offsets (e.g., [M+NH4]+ at +17.0266 Da, [M+K]+ at +38.9815 Da), and before assigning adduct labels to a feature table.

Use when after constructing initial data bins from mzTree (indexed by int(mz × 1000)), determine whether a single bin contains one or multiple mass tracks. Apply clustering when the m/z range of points in a bin exceeds 2 × ppm tolerance (e.

Use when after mass tracks have been aligned across all samples (either via pairwise alignment for ≤10 samples or nearest-neighbor clustering for larger cohorts), and you need to generate a single representative m/z per aligned bin for downstream feature extraction and annotation.

Use when when you have centroided mzML files from LC-MS metabolomics and need to construct high-mass-resolution mass tracks for each sample before alignment. Apply this skill at the start of an untargeted metabolomics workflow, before building a cross-sample MassGrid.

Use when when you have centroid mzML files from LC-MS metabolomics acquisition and need to construct sample-level mass tracks before cross-sample alignment. Specifically: you are starting fresh with vendor-converted or pre-processed mzML input;

Use when after discovering Mass2Motifs through LDA topic modeling of MS/MS fragmentation data, when you need to assign chemical meaning (substructure classes, candidate annotations) to motifs by leveraging pre-trained spectral embeddings and a reference motif database.

Use when when you have a preprocessed bag-of-fragments corpus from tandem mass spectrometry spectra and need to train an MS2LDA model to discover Mass2Motifs.

Use when when you have preprocessed MS/MS spectra converted to a bag-of-fragments format (fragments and neutral losses extracted, noise filtered) and you seek to discover recurring fragmentation patterns without prior compound identification.

Use when you have created a GNPS molecular network (classical or feature-based workflow) and run an MS2LDA experiment on the corresponding MGF spectra, and you want to annotate network nodes with shared Mass2Motifs and chemical class information to interpret the structural basis of network.

Use when after individual mass tracks (EICs) have been extracted from each sample''s mzML file and you need to create a unified, cross-sample m/z reference structure. Triggered when: (1) you have ≥2 samples in a cohort; (2) mass tracks have been binned at 0.

Use when you have trained a shallow decision tree on ChemEcho feature vectors (representing unique peak or neutral loss formulas from tandem MS spectra) and need to deploy the learned splitting logic as a queryable, inspectable artifact.

Use when you need to search for specific mass spectrometry patterns (e.g., precursor ion m/z, product ion presence, retention time windows, intensity constraints, neutral loss patterns) across one or more mzML files.

Use when you have completed one or more domain-specific MASST searches (microbeMASST, plantMASST, tissueMASST, microbiomeMASST, foodMASST) and have aggregated search outputs (matches.tsv, library.tsv, datasets.

Use when you have two separate LC-MS untargeted metabolomic feature datasets (each with retention time and m/z values) and need to establish feature-to-feature correspondence between them.

Use when you have mass spectrometry data in mzXML or mzML format and need to systematically extract regions of interest (ROIs) from multi-dimensional m/z-intensity-time arrays, normalize feature values, and augment datasets for untargeted metabolomics workflows.

Use when when you have a confusion matrix (predicted vs. ground-truth labels) or similarity matrix (pairwise scores between spectra) and need to communicate classification accuracy or chemical similarity patterns through a visual heatmap.

Use when after executing memo_from_unaligned or memo_from_aligned functions to generate a MemoMatrix object from MS2 spectra or aligned feature tables.

Use when you have computed pairwise similarity or mass difference scores between all fragment ions across two tandem mass spectra and need to select the non-overlapping set of ion pair matches.

Use when after fitting a Multi-Block PLS (MB-PLS) discriminant or regression model on multi-assay LC-MS intensity data (e.g., HPOS, LPOS, LNEG blocks), and you need to identify which features drive model performance and warrant further statistical validation or biological interpretation.

Use when you have a feature table from LC- or GC-HRMS data (either detected via pyOpenMS or imported as a custom feature list) containing m/z, retention time, and intensity values, and you want to rapidly filter to candidate PFAS features that exhibit the elevated mass defects typical of.

Use when you have aligned feature tables (CSV format) with corresponding MS2 spectra data (MGF or mzML files), and need to construct a sample-level vectorization matrix where each row represents a sample and columns encode the occurrence counts of MS2 peaks and neutral losses observed in that.

Use when you have generated one or more MemoMatrix objects (MS2 fingerprint matrices from separate sample sets) and need to combine them for cross-cohort alignment, validate structural consistency after merging, or prepare merged matrices for downstream filtering and visualization.