Claude Skills by HolobiomicsLab

Use when you have a baseline GNN model for predicting a continuous molecular property (e.

Use when you have ion-mobility mass spectrometry metabolomics data with putative metabolite identifications (e.g., from database matching) and want to reduce false positives by cross-validating compound identities against machine-learned CCS predictions.

Use when when you have MS/MS-annotated features from a natural extract (via SIRIUS, CANOPUS, or ISDB) and need to compute the Literature Component or Class Component of a priority rank—i.

Use when you have raw LC-MS data in mzML or equivalent binary format from a public repository (MetaboLights, MassIVE) or instrument vendor output, and need to ingest it into MetaboAnalystR 4.0 for unified LC-MS1 feature detection and MS/MS spectra processing.

Use when you have multiple CSV feature lists from different acquisition methods (e.g., LC-MS vs LC-IMS-MS) or processing software, each using different naming conventions, retention time scales, or m/z precision;

Use when working with raw multiplexed IM-MS data (UIMF or Agilent MassHunter .

Use when you have a pre-trained GNN model for CCS prediction and need to verify that it generalizes to test data that was held out during training. Use it specifically when comparing model performance across different molecular datasets (e.

Use when you have a pre-trained or newly retrained graph neural network for collision cross section prediction and need to measure whether its performance generalizes across different molecular datasets (e.g., training on METLIN but evaluating on CCSBase).

Use when you have a trained baseline GNN model with established hyperparameters (dropout rate, learning rate, epochs, optimizer settings) and want to evaluate whether alternative message-passing GNN architectures (Graph Attention Networks, Message-Passing Neural Networks) achieve comparable or.

Use when evaluating whether a mass spectrometry data analysis platform (such as mzmine) provides complete module coverage across all advertised separation and ionisation techniques.

Use when when you need to understand how a multi-instrument mass spectrometry platform (like mzmine) decides which processing module receives a given dataset based on its declared data type (LC vs. GC vs. IMS vs. MS imaging).

Use when you have a curated dataset of small molecules with SMILES, optional 3D coordinates, adduct information, and experimentally measured CCS values (in Ångströms or similar units), and you want to train a GNN model to predict CCS on held-out test molecules.

Use when you have SMILES strings or 2D molecular structures of N-Me derived unsaturated sterol lipids (or other C=C-containing molecules) and need to generate 3D conformational and electronic structure data as input to a machine-learning CCS prediction model.

Use when you have a set of molecular SMILES strings and need to prepare them as inputs to a deep-learning CCS prediction model.

Use when when setting up a ViMMS chemical sampling environment and you need to restrict the chemical search space to a specific m/z range (e.g., 100–1000) and MS level (e.g., MS1 only) before generating virtual LC-MS/MS data.

Use when you have raw molecular datasets (e.g., METLIN-CCS, CCSBase) with SMILES strings, 3D coordinates, adduct information, and ground-truth collision cross section labels, and you need to format them for GNN training and held-out test evaluation with consistent standardization.

Use when when you need to train or evaluate a graph neural network for molecular property prediction (collision cross section, ion mobility, or related descriptors) and have access to SMILES strings and/or 3D conformer coordinates.

Use when when you have a collection of molecular structures (SMILES or SDF format) and need to generate a queryable database of collision cross section values for high-throughput mass spectrometry workflows, or when you want to organize pre-computed CCS predictions with conformer metadata into a.

Use when you need to represent, validate, and manipulate molecular compositions in MS analysis—specifically when annotating precursor or product ions with elemental formulas, computing monoisotopic masses for formula-to-charge assignments, predicting isotope patterns ([M]+, [M+H]+, [M+Na]+, etc.).

Use when after RDKit has generated multiple conformations for a molecule in an SDF or XYZ format, and you need to reduce the conformational ensemble to a tractable size (by energy-based ranking) before submitting to expensive quantum-chemical calculations (e.g., QUICK).

Use when you have raw SMILES strings from a chemical database (e.g., CCSBase, METLIN, or custom sources) and need to feed them into a graph neural network model.

Use when you have a trained GNN model predicting CCS values from molecular graphs and need to understand which structural features (node and edge attributes) are most influential for specific predictions or across a test set.

Use when you have 2D molecular structures (SMILES or SDF format) and need to create 3D conformer geometries as input to subsequent computational chemistry workflows, such as CCS prediction, molecular graph construction, or voxel-based property calculations.

Use when you have a collection of molecular structures (with SMILES strings, InChI, or similar identifiers) and corresponding experimentally determined or reference CCS values, and you need to format and validate them as input to a machine learning CCS prediction model.

Use when you have molecular structures (SMILES strings or molecular graphs) that need to be input to a transformer model for property prediction (e.g., Collision Cross Section), and the model requires tokenized or embedded representations rather than raw chemical notation.

Use when when receiving raw MS data files of unknown or mixed acquisition modalities and needing to route each to its corresponding analysis pipeline. Specifically, apply this skill when: (1) input files arrive without documented instrument type or chromatographic/mobility dimensionality;

Use when when evaluating or designing a mass spectrometry data analysis platform, and you need to verify that every supported separation/ionisation technique (LC, GC, IMS, MS Imaging) is covered by at least one processing module.

Use when when you have predicted MS/MS fragments from quantum chemistry calculations on N-Me derived unsaturated sterol structures and need to map each fragment to its precursor lipid, calculate exact m/z values, estimate relative intensities, and produce a machine-readable reference table for.

Use when you have raw or semi-processed MS/MS spectral data from bottom-up tandem mass spectrometry experiments (data-dependent acquisition) that you intend to input to de novo peptide sequencing tools like Casanovo.

Use when you have raw or parsed tandem MS spectra (MGF, mzML, or in-memory Spectrum objects) and need to remove artifacts and normalize intensities prior to spectral matching, library searching, or downstream analysis.

Use when when you have raw LC-MS data files and need to identify which compounds were actually detected at high abundance during a gradient run, prior to evaluating whether the gradient provided good separation across the chemical space.

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 after generating a 3D pixel array (shape n+1, y, x) from raw line-scan MSI data via MSIGen's get_image_data() call, curate and validate the accompanying JSON metadata file before visualization or downstream analysis.

Use when you have raw line-scan MSI data (from Bruker .d/.baf, converted to .mzML, or other supported formats) and need to configure MSIGen before calling msigen() to generate image arrays.

Use when when building a mass spectrometry visualization library that must support multiple plotting backends (matplotlib for static output, bokeh and plotly for interactive exploration) with a uniform DataFrame-based API, and where different plot kinds (spectrum, chromatogram, mobilogram, peakmap).

Use when your TWIM-MS dataset contains ions with multiple charge states (e.g., +1, +2, +3 for the same molecular species) and you need CCS values that correctly account for the relationship between drift time, m/z, and charge state.

Use when you have raw or processed mass spectrometry data in HDF5 (.h5) or mzML format and need to ingest it into DEIMoS for multi-dimensional analysis.

Use when you have DIA mass spectrometry raw files from multiple instrument types (timsTOF, TripleTOF, Orbitrap) in their native formats (.raw, .d, .

Use when you have raw or processed TWIM-MS data with arrival time and m/z dimensions, and you need to assign biomolecular class labels and compute class-specific CCS values across multiple compound classes simultaneously, particularly when you lack pre-identified feature peaks or want to bypass.

Use when when you need to visually compare two or more spectra (MS1, MS2, or extracted ion chromatograms) across different retention times, m/z ranges, or ion mobility bins in a single figure. Apply this skill after extracting spectra arrays from MZA files (e.

Use when when you have MS1 mass spectrometry data from multiple samples (a cohort) acquired across an instrument run or batch, and you need to identify which samples deviate from cohort norms or which ion targets show anomalous behavior.

Use when you have detected feature tables from two or more LC-IMS-MS/MS samples (each containing m/z, drift_time, retention_time, and intensity measurements) and need to identify which features correspond to the same chemical compound across samples.

Use when you have a peak-picked feature table (HDF5 format) from high-dimensional MS data (m/z, drift_time, retention_time, intensity) and need to identify and label isotopic signatures to distinguish monoisotopic peaks from isotopologues, reduce feature redundancy, and support multi-dimensional.

Use when when you have acquired LC-IMS-MS/MS data (mzML or mzML.gz format) from multiple samples and need to detect features that exploit simultaneous separation in m/z, drift time, and retention time to improve detection sensitivity and reduce false positives.

Use when you have acquired untargeted MS data with orthogonal separations (LC, ion mobility) and/or data-independent acquisition (DIA) from Thermo, Agilent, or Bruker instruments, and you need to: (1) store multidimensional spectra in a vendor-neutral, platform-agnostic format;

Use when processing raw IM-MS data (Agilent MassHunter .d or UIMF format) that contains jagged, low-abundance ion peaks or when saturation repair has been applied and the resulting reconstructed signals need artifact removal and enhancement.

Use when you have multidimensional MS data (with LC and/or ion mobility dimensions) converted to MZA HDF5 format and need to retrieve raw spectral intensity and m/z values for specific scans, retention times, drift times, or mass ranges.

Use when you have raw MS data in vendor formats (Agilent .d, Thermo .raw, Bruker .

Use when you have raw IM-MS data in UIMF or Agilent MassHunter .d format acquired from a multiplexed (interleaved) ion mobility experiment, and you need to recover individual, demultiplexed frames to reconstruct conventional IM-MS spectra for downstream omics analysis.

Use when you have mzML-format mass spectrometry data files and need to load them into memory as structured data (pandas DataFrame) to prepare for visualization with pyOpenMS-Viz or other analysis pipelines.