Claude Skills by HolobiomicsLab

Use when when you have experimental tandem MS spectra (with peak m/z and intensity values) and need to annotate each peak with its chemical formula (SCARF) or molecular fragment origin (ICEBERG), particularly for structural elucidation campaigns where understanding the fragmentation pathway is.

Use when you have time-resolved direct injection mass spectrometry data (e.

Use when when you have a complete dataset of labeled examples (e.g., 100,000 augmented spectra, chromatograms, or synthetic samples) and need to train a supervised model (such as a Transformer) while preserving a held-out test set to measure generalization performance without bias.

Use when you have electron ionization mass spectrum data (m/z and intensity pairs) and a pre-trained transformer model checkpoint, and you need to predict molecular weight directly from the spectrum without manual feature engineering or rule-based methods.

Use when you have a dataset of augmented simulated overlapped GC-MS peaks and need to train a Transformer model to automatically deconvolve them into pure component mass spectra.

Use when you have raw GCxGC-MS chromatogram data in NetCDF format from multiple samples (e.g., case and control groups) and need to prepare them for multivariate analysis such as multiway principal component analysis (MPCA).

Use when when you have a preprocessed sample chromatogram (smoothed and baseline-corrected) and a preprocessed reference chromatogram, and need to align them using 2D COW.

Use when you have raw GC–MS or LC–MS data represented as a two-dimensional map (m/z axis vs. retention time axis) and need to identify chemo-/biomarker features across multiple analytes simultaneously, especially when conventional peak picking produces high false-positive or false-negative rates.

Use when you have GC–MS or LC–MS data represented as a two-dimensional map with m/z values on one axis and retention time on the other, and you need to identify analyte signals and chemo-/biomarker features while minimizing false positive and false negative peak detections.

Use when immediately after acquiring raw GCxGC-MS data in NetCDF format (.cdf files) and before any signal enhancement (smoothing, baseline correction) or alignment steps.

Use when you have a GC-MS results table with a Match.Factor column (representing identification confidence) and you need to understand how many distinct compounds survive at different quality cutoffs (e.g., ≥65, ≥80, ≥90).

Use when when you have preprocessed mass spectral data (normalized peak intensities or binned m/z representations) and need to discover latent spectral patterns to enhance neural network predictors without labeled spectral classes.

Use when when you have untargeted GC–MS or LC–MS data in the form of a two-dimensional m/z vs retention time map and need to identify marker features without conventional peak picking.

Use when after baseline correction (e.g., via asymmetric least squares) when raw GCxGC-MS chromatograms still contain high-frequency noise that obscures true signal structure. Use it when you need to reduce noise before peak alignment or multivariate analysis (e.

Use when you have already assigned samples to batches (inter-batch balance is fixed) and need to shuffle injection order within each batch to decorrelate sample properties from time-dependent instrumental effects. Use it when your metadata table includes a randomization dimension (e.

Use when when you have pre-processed MS/MS spectra and a pre-trained Word2Vec model, and need to compute fast, scalable similarity scores for library matching or molecular networking that correlate better with structural similarity than cosine-based methods.

Use when when comparing large numbers of MS/MS spectra against spectral libraries or in molecular networking, particularly when molecules differ by multiple structural modifications and cosine-based scores produce excessive false positives.

Use when you have a large collection of preprocessed MS/MS spectra (typically >10,000 spectra) with diverse chemical structures and you need to learn embeddings that capture fragmentation patterns and neutral loss relationships.

Use when when applying a pre-trained Word2Vec model to mass spectra at inference time (e.g., library matching or molecular networking), especially when the query spectra may contain fragment peaks or neutral losses not represented in the model's training vocabulary.

Use when converting MS/MS spectra into Spec2Vec embeddings using a pre-trained Word2Vec model that was trained on reference data (e.g., a subset of GNPS or MassBank).

Use when when you have raw mass spectrometry data files from multiple acquisition modalities (LC-MS, GC-MS, ion mobility, or imaging) and need to automatically route each to the correct downstream analysis module without manual intervention.

Use when you have raw LC-MS or GC-MS data files from a mass spectrometer (in mzML, NetCDF, or mzXML format) and need to detect chromatographic peaks, correct m/z bias via mass calibration (e.

Use when you have raw mzXML LC/MS files from replicated metabolomics experiments (e.g., 12 samples across labeled/unlabeled conditions) and need to extract, align, and group peaks before downstream feature filtering (e.g., fold-change or isotope enrichment analysis).

Use when when you have direct-injection or low-complexity mass spectrometry data (mzML files) and need to detect chromatographic peaks using wavelet-based methods instead of centWave, especially when standard retention-time-dependent peak detection is not suitable or when you need to tune.

Use when you have raw LC-MS data files (mzML, netCDF, or raw vendor formats) from multiple samples and need to extract, align, and quantify chromatographic features across the cohort.

Use when a task needs a skill from ASB Metabolomics — GC-MS — search this unit's 367 evidence-grounded skills, then apply and optionally ground the one that fits.

Use when you have raw LC-IM-MS/MS data files from sterol lipid analysis and need to identify unsaturated sterol isomers by matching experimental collision cross section values against a quantum chemistry calculation-assisted CCS prediction database.

Use when after peak detection and MS1 feature extraction from FIA-MS, GC-MS, LC-MS(/MS), or CE-MS data, when you need to identify unknown metabolites by matching observed m/z values to a reference database and want to recover HMDB identifiers, molecular formulas, and structural annotations for.

Use when you have acquired the same sample(s) using multiple LC-MS, LC-IMS-MS, or direct infusion methods (e.

Use when after temporal correlation has identified candidate feature pairs with matching intensity profiles across time-resolved DBDI-MS experiments.

Use when when you have multiple mass spectrometry data formats (mzML, mzXML, or others) that must be ingested into a single format-agnostic processing engine (e.g., mspack compression), and you need to avoid replicating the core logic for each format.

Use when you have raw or processed TWIM-MS data with arrival time and m/z dimensions, and you need to label each experimental feature by biomolecular class before feature identification or peak detection steps are complete.

Use when when you have raw TWIM-MS arrival-time data and need to transform it into absolute CCS values for downstream biomolecular class assignment or comparative analysis.

Use when when processing raw TWIM-MS experimental data that contains arrival time measurements but you need drift times for CCS calibration or class-specific CCS calculations.

Use when processing raw IM-MS data (UIMF or Agilent MassHunter .d format) that exhibits jagged peaks in low-abundance ions, isolated high-intensity noise spikes, or saturated detector signals that distort elution and mobility profiles.

Use when you have a trained GNN model for molecular property prediction (such as CCS) and need to understand which atomic and bond features are most influential in driving predictions.

Use when you have canonicalized SMILES strings from a chemical database (e.

Use when you have imaging mass spectrometry (IMS) datasets with peak intensity features organized as spatial graphs (nodes = pixels/voxels, edges = spatial adjacency), and you need to discover latent peak patterns for automatic peak picking or marker ion identification without manual feature.

Use when when working with imaging mass spectrometry (IMS) datasets where you need to extract latent low-dimensional peak features from high-dimensional peak intensity data while preserving spatial adjacency information.

Use when you have ion mobility-mass spectrometry data (raw drift times, m/z values, and feature intensities) from DTIMS-MS or SLIM-based IMS-MS platforms and need to compute collision cross section values using a calibration standard (e.g., Agilent tune-mix).

Use when before peak detection on a composite or individual mass track when you need to filter out low-intensity noise and baseline drift without removing true signal.

Use when after implementing a lossless compression–decompression cycle on mzML or mzXML mass spectrometry files, or when validating that a lossy compression pipeline meets acceptable error thresholds.

Use when you have parsed imzML XML metadata and loaded the corresponding .ibd binary intensity file, and need to extract specific ion images at one or more target m/z values.

Use when you have raw or processed TWIM-MS data with arrival time and m/z dimensions, and you need to label experimental features by biomolecular class before performing CCS calibration or validation.

Use when after biomolecular class labels have been assigned to features in a TWIM-MS dataset and you have raw ion mobility arrival time measurements. Use it when you need to convert arrival times to standardized CCS values where calibration accuracy depends critically on the biomolecular class (e.

Use when you have raw or processed TWIM-MS data with arrival time and m/z values for multiple features, but lack prior structural identification (e.g., from spectral libraries or databases).

Use when you have Bruker .d/.baf format mass spectrometry imaging data and need to ingest it into MSIGen for conversion to visualizable ion images. This skill applies when your raw data originates from Bruker TIMSTOF or similar instruments and you lack direct .

Use when you have raw mass spectrometry data in mzML or mzXML format and need to compress it using a format-agnostic compressor that expects a standardized spectral data contract.

Use when after applying deimos.calibration.tunemix() to positive-mode or negative-mode tune mix reference data (containing known CCS values across m/z range 118–1522), assess whether the single-field calibration model's r-squared coefficient meets the expected precision (typically ≥0.

Use when you have IM-MS lipidomics data acquired on samples spiked with fully labeled U13C lipid standards (e.g., U13C yeast extract), and you need to assess whether systematic CCS deviation exists between your instrument's measured values and the DT CCS N2 reference library for U13C labeled lipids.