Claude Skills by HolobiomicsLab

Use when after converting raw Bruker .d/.baf or other proprietary mass spectrometry imaging formats using pyBaf2Sql or ProteoWizard MSConvert, or after running MSIGen's get_image_data() function.

Use when building or auditing a multi-instrument MS data processing system that must route different chromatography modes (LC, GC), ion mobility, or imaging modalities (MALDI) to distinct processing workflows.

Use when you have extracted molecular features (voxel projected areas, molecular graphs, m/z values, adduct encodings) and need to build a predictive model to map these features to a continuous molecular property (CCS).

Use when you have transformer encoder components and a prediction head specification, and need to wire them into a single trainable model that maps molecular structure inputs (SMILES, molecular graphs, or feature vectors) to scalar or vector molecular property predictions.

Use when when you have a TransitionGroup structure containing normalized intensity traces (1D signal data from chromatograms or mobilograms) and you want to automatically detect peak regions with high precision by leveraging a conformer-based deep learning model.

Use when you have raw DIA mass spectrometry files in timsTOF (.d), TripleTOF (.wiff), or Orbitrap (.raw) format and need to extract precursor ion chromatogram (PIC) data as input for quality metric computation or machine learning-based file quality prediction.

Use when you have a batch of DIA mass spectrometry raw files (.raw, .d, or .wiff formats) from known instrument types (timsTOF, TripleTOF, or Orbitrap) and need to classify each file as pass/fail or assign a quality label.

Use when when you have detected feature tables from two or more LC-IMS-MS/MS samples and need to establish correspondence between features across samples (i.e., which features in sample A match which features in sample B).

Use when when you have ionized adduct structures (in SMILES or MOL format) from an upstream ionization-state determination step and need to produce multiple low-energy 3D conformations for collision cross section prediction, metabolite annotation, or structure-property modeling.

Use when you have GCIMS samples exhibiting misalignment across drift time (typically 5–16 ms range) and retention time (typically 0–1100 s range) caused by pressure/temperature fluctuations or chromatographic column degradation, and you need to normalize sample positions before downstream peak.

Use when you have loaded a raw GCIMS dataset and need to isolate the region of interest in drift time (typically 5–16 ms for small organic molecules) to exclude low-drift-time chemical noise, high-drift-time tail artifacts, or off-scale ion signals that would degrade subsequent alignment and peak.

Use when after peak detection in GCIMS when you need to group detected peaks across multiple samples into reproducible clusters.

Use when you have tunemix or other reference standards with known m/z, drift-time, charge state, and CCS values, and you need to establish a predictive calibration model for your ion-mobility mass spectrometry instrument.

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

Use when you are reconstructing targeted ion chromatograms (XIC) and ion mobilograms (IM) from raw diaPASEF data and need to balance sensitivity (wide extraction windows) against specificity (narrow windows that reject interference).

Use when when you have a trained GNN model for molecular property prediction (e.g., collision cross section) and need to identify which graph structural features—atomic properties, bond types, or higher-order graph descriptors—are driving the model's predictions.

Use when you have feature lists in CSV format originating from different acquisition methods (e.

Use when you have detected feature tables from multiple LC-IMS-MS/MS samples and need to establish a unified feature catalog in which each row represents a distinct molecular entity observed across one or more samples, with harmonized m/z, drift time, and retention time coordinates.

Use when when you have raw mzML or mzXML mass spectrometry data files that need to be archived or transmitted with minimal storage footprint, and you must verify that the decompressed output exactly reproduces the original input at the byte level.

Use when you have raw MS data files from one or more instrument vendors (Agilent, Bruker, Thermo Fisher, or mzML-formatted) and need to convert them to a vendor-agnostic HDF5-based storage format for downstream software development, machine learning, or cross-platform data access.

Use when you have aligned and baseline-corrected GC-IMS data with detected and clustered peaks, and you want to extract peak intensities using a consistent integration window.

Use when after RDKit generates multiple 3D conformers from ionized molecular structures using distance-geometry embedding, before filtering with ASE-ANI or submitting to quantum calculations.

Use when when you have mass spectrometry data in multiple formats (mzML, mzXML) that must be fed into a compression, analysis, or transformation pipeline that operates on a single canonical spectral data model.

Use when you have generated a lipid spectral library (with lipid identities, adducts, m/z values, and fragmentation patterns) and need to export it for downstream mass spectrometry analysis on either an Orbitrap (via Excalibur DDA) or via Skyline's transition-based workflow.

Use when when preparing tandem MS/MS data for spectral alignment and similarity comparison, particularly when you have loaded raw fragmentation spectra and need to extract and quantify mass difference patterns that capture the fragmentation process.

Use when you have experimental MS/MS spectra matched against a reference library (via cosine similarity or dot-product scoring) and need to map individual fragment peaks in the experimental spectrum to their corresponding m/z and intensity values in the matched library entry.

Use when when you have detected m/z values from LC-IM-MS/MS that match a candidate CCS prediction database but require structural confirmation. Apply this skill to disambiguate isomers (e.

Use when after completing multidimensional smoothing and saturation repair on Agilent MassHunter (.

Use when after integratePeaks has been executed with a chosen integration method (e.g., fixed_size with RIP saturation threshold of 0.1) on a clustered, baseline-corrected GC-IMS dataset.

Use when after peak detection in GC-IMS preprocessing, when you need to assess whether detected peaks from multiple samples align to the same chemical entities (clusters) using hierarchical clustering.

Use when you have raw GCIMS sample files (from a GC–IMS instrument) and an annotations table (Excel, CSV, or TSV) with sample metadata, and you need to begin the GCIMS preprocessing pipeline.

Use when when you have converted multidimensional MS data to MZA HDF5 format and need to programmatically read spectra, metadata, and ion mobility or mass-to-charge arrays in Python or R without vendor-specific dependencies.

Use when you have a pre-trained GNN model checkpoint, a test dataset with molecular representations (SMILES, 3D coordinates, adducts) and ground-truth labels, and need to quantify how well the model generalizes to held-out data.

Use when you have a trained graph neural network model for CCS prediction and need to identify which molecular structural features drive individual predictions or systematic biases.

Use when after constructing a spectral similarity network from pairwise cosine similarity scores between deconvolved GC-MS spectra.

Use when you have a baseline GNN model trained on a molecular property prediction task (e.

Use when when your input includes molecular structures (SMILES, conformers) and you need to predict a continuous property (e.g., CCS, binding affinity, solubility) that depends on molecular connectivity and spatial relationships.

Use when you have a trained GNN model (stored as .h5 weights) and molecular graph representations (SMILES strings and/or 3D coordinates), and you need to compute predicted CCS values or perform feature importance analysis via ablation or gradient-based saliency mapping.

Use when when you need to evaluate GNN performance on collision cross section prediction using the enveda/ccs-prediction repository, either by loading an existing pre-trained model checkpoint or by retraining from scratch using deposited datasets and published hyperparameters.

Use when after constructing feature tensors encoding atom adjacency matrices, bond types, and chemical properties from canonical SMILES—and before feeding graphs into a GNN training loop—to enable reproducible, portable, and memory-efficient storage of graph objects that will be loaded in batches.

Use when you have mass spectrometry feature data stored in HDF5 format (.h5 files) and need to load specific dimensional columns (m/z, drift time, retention time, intensity) for multi-dimensional alignment, CCS calibration, or isotope detection.

Use when you have raw or peak-picked mass spectrometry data in HDF5 format that needs to be loaded into memory for downstream processing (feature alignment, isotope detection, CCS calibration), or when you need to export annotated feature tables with isotopologue metadata back to HDF5 for archival.

Use when when you have converted multidimensional mass spectrometry data (or other complex data) into MZA HDF5 format and need to understand the file hierarchy, validate the conversion output, extract metadata tables and array structure, or access spectra across different programming environments.

Use when you have multidimensional MS data converted to MZA HDF5 format (from Agilent .d, Bruker .d with ion mobility, Thermo .

Use when after invoking the DEIMoS CLI with a configuration file and allowing the Snakemake workflow to execute, use this skill to confirm successful completion of all workflow rules.

Use when after converting proprietary vendor mass spectrometry files (Thermo .raw, Agilent .d, Bruker .d, or mzML) to MZA HDF5 format using the MZA executable.

Use when after peak detection in GC-IMS preprocessing, when you need to group peaks across multiple samples and must decide whether euclidean distance is appropriate for your drift time and retention time coordinate space, and when you need to validate that your chosen dt_cluster_spread_ms and.

Use when when you have received an MZA-format HDF5 file (or other hierarchical HDF5-based scientific data) and need to understand its internal structure before downstream processing, share it across programming languages (Python, R), or document its schema for reproducibility.

Use when you have a filtered set of conformers (100s–1000s) from ASE-ANI that each require independent quantum calculations via QUICK, and you have access to HPC resources with multiple cores or nodes. Parallelization is necessary when serial execution would exceed practical time budgets (e.

Use when when you have IM-MS lipidomics data acquired on samples spiked with U13C-labeled internal standards (fully labeled yeast extract) and need to quantify systematic CCS bias and apply lipid class-specific bias correction to all measured CCS values, particularly when multiple lipids per lipid.