Claude Skills by HolobiomicsLab

Use when when preparing library compound spectra and observed mixture spectra for mcfNMR analysis, normalize both spectra to cumulative distribution functions (CDFs) scaled by total intensity.

Use when you have Bruker NMR spectral files (raw instrumental output) and need to prepare them for automated metabolite identification and quantification.

Use when you have a loaded INADEQUATE NMR spectrum file (after referencing via basic shifting) and need to identify individual peak positions and intensities as input to downstream peak network clustering or metabolite matching.

Use when you have acquired raw 1D NMR spectra (FID or processed spectra in NMRPipe format) that exhibit phase distortion—where peaks are not in pure absorption mode—and you need to prepare the spectrum for automatic deconvolution into peak tables.

Use when you have raw or semi-processed 1D NMR spectra (¹H and/or ¹³C) from routine laboratory instruments and need to feed them into a CNN–transformer architecture for end-to-end structure elucidation.

Use when you have an NMR spectrum of a mixture sample and a library of reference spectra for individual compounds.

Use when when loading or creating an NMR spectral dataset (Dataset.createDataFile) and the system must decide between multiple storage backends (SubMatrixFile, BigMappedMatrixFile, MappedSubMatrixFile, MappedMatrixFile). Triggers include: (1) dataset metadata specifies a cache-file flag;

Use when you have 1D NMR spectra (¹H or ¹³C or both) for an unknown organic compound with ≤19 heavy atoms and need to rapidly predict its molecular formula and connectivity graph without manual peak interpretation or exhaustive combinatorial search.

Use when when you have known metabolite concentrations and their spin-system coupling constants (J-values) and need to generate synthetic ¹H NMR spectra for method validation, metabolite library construction, or as a training dataset.

Use when you have raw 1D NMR spectra (FID or processed spectrum files) and need to extract peak parameters (chemical shift, intensity, linewidth) in a tabular format for downstream metabolomic or structural analysis.

Use when when you need to deploy SAND for 1D NMR spectrum deconvolution on a new machine or cluster, or when testing code changes before merging into production.

Use when you have a raw NV file from NMRViewJ or compatible NMR acquisition software and need to extract header metadata before processing spectroscopic data.

Use when when you have an observed NMR mixture spectrum and one or more candidate reconstructed spectra (each formed by combining single-compound reference spectra from a library), and you need a scalar similarity score to rank or select the approximation.

Use when after peak annotation when you have: (1) a peak intensity matrix (rows=peaks with KEGG/ChEBI/UniProt IDs, columns=samples) with group labels; (2) a pathway database (KEGG, Reactome, or user-defined metabolite sets); (3) a comparative experimental design (case vs. control groups).

Use when you have extracted a raw peak list (chemical shifts in a TXT file, one per row) from a 1D 1H NMR spectrum and intend to pass it to NMRformer or a similar deep learning model for metabolite identification.

Use when you have picked peaks (coordinates and intensities) from INADEQUATE spectra and need to distinguish which peaks likely originate from the same metabolite before matching to a reference database.

Use when after peak clustering has produced peak network groups (ideally from the same compound) and you need to assign chemical identities.

Use when you have a set of picked peaks from INADEQUATE NMR spectra and need to group them into networks to identify which peaks co-originate from the same metabolite compound.

Use when you have raw or converted spectral data (jcamp, RAW, or mzML format) from NMR, IR, or MS instruments and need to identify individual peaks, extract their properties (chemical shift, m/z, intensity, width), and generate a structured peak table for annotation, comparison, or publication.

Use when you have extracted m/z and retention time (m/z-RT) information for peaks from untargeted LC/HRMS data (using tools like IDSL.IPA) and need to assign molecular formula identities to those peaks.

Use when you are attempting to run a web application (such as COLMARvista) locally by opening index.html directly in a browser, the application uses WebWorker and/or WebAssembly components, and these fail to load due to default file-access policies.

Use when you have a pretrained model with documented performance on a bounded input domain (e.g., molecules ≤19 heavy atoms, sequences <1000 bp) and you need to establish whether and how much accuracy drops on held-out test cases outside that domain boundary.

Use when you have centroided MS2 spectra (ddMS2 data in mzML format) from HRMS analysis and need to identify potential PFAS compounds among thousands of features.

Use when you have NMR-based metabolomics measurements from a cohort containing both plasma and serum samples with associated processing delay metadata (pre- and post-centrifugation times), and you need to determine whether metabolic parameter stability differs between the two sample types or to.

Use when you have uploaded a pre-analytical data table containing sample metadata, processing delay annotations (pre- and post-centrifugation timestamps or duration), and paired NMR metabolomic measurements for a sample cohort, and you need to quantify how delays at different time-points affect.

Use when when you have NMR metabolite measurements paired with documented pre-centrifugation and post-centrifugation delay times, and need to assess how processing delays affect metabolic parameter stability within a plasma or serum sample cohort.

Use when you have Nightingale Health 1H-NMR metabolomics measurements for a new cohort and wish to compute one or more established metabolic risk scores (mortality, MetaboAge, cardiovascular event, type-2 diabetes, COVID-19 severity) without recalibration.

Use when when you have a pre-trained deep learning model checkpoint (saved in PyTorch format) and new 1D 1H NMR spectral data in CSV and peak-list TXT formats, and you need to generate peak-to-metabolite assignments or other structured outputs from that model without modification of model weights.

Use when when you have cloned a scientific Python repository (e.g., ROIAL-NMR) and need to verify that the documented dependencies can be installed and the main entrypoint is invokable without errors. Use this skill before attempting to run the application's core analysis workflows.

Use when when you need to extract both local spatial patterns and global long-range dependencies from sequential or spectral data (e.g., 1H NMR spectra), and neither CNNs nor Transformers alone are sufficient.

Use when after extracting NMR spectra and designating replicate QC samples (typically 10 samples run throughout the study), calculate CV for each metabolite feature to assess which signals are reproducible enough for downstream metabolite-phenotype association testing.

Use when after running QC analysis on NMR or MS metabolomic data and obtaining per-feature CV values, use this skill to validate that the dataset meets FDA thresholds (CV < 0.30 for discovery, CV < 0.15 for quantification) and to report the proportion of features meeting each threshold.

Use when you have raw Bruker NMR spectral data files (1D 1H format) stored in a directory structure and need to prepare them for automated metabolite identification and quantification in ASICS.

Use when you have a published predictive model with known coefficients and feature requirements (e.g., MetaboAge from a peer-reviewed study), a target R package with an established data pipeline (e.

Use when you have multiple rows in a table that describe the same logical entity (e.g., multiple measurements from the same sample, or repeated attributes under a common identifier) and need to merge them into a single dictionary entry keyed by that shared field.

Use when after NMR or MS data acquisition and preprocessing (phasing, baseline correction) when you have a SummarizedExperiment object containing assay intensity matrix with QC sample columns designated.

Use when you have a GitHub repository URL, a documented Python version requirement, and a list of pinned package versions, and you need to verify that the application will initialize without import or runtime errors before proceeding to data analysis or method replication.

Use when when you need to deploy a containerized scientific tool (e.g., CloMet) for the first time on a local machine or CI/CD environment, and the project provides a Dockerfile and GitHub repository.

Use when you have NMR peak data (1H and 13C chemical shift values) and need to obtain SMART 3 classification predictions from the DeepSAT service.

Use when you have 2D NMR spectral data (HSQC, HMBC, COSY) from multiple samples in a natural products screening campaign and need to rank samples by structural novelty or identify which samples share common molecular scaffolds.

Use when when preparing a 1D 1H NMR spectral peak list for input to the NMRformer metabolite identification model, and you have access to peak intensity measurements and noise level estimates.

Use when you have observed compounds (from LC-MS, GC-MS, or spectroscopy) and a set of predicted metabolite structures from BioTransformer, and need to assign identities to the observed compounds by matching their experimental features (mass-to-charge ratio, retention time, spectral signature) to.

Use when you have LC–QTOF mass spectra from real environmental or biological samples (e.g., wastewater, complex mixtures) and need to assign molecular structures to spectra where traditional library matching fails or reference compounds are unavailable.

Use when when annotating matrix-related peaks in MSI datasets where candidate peaks have identical or near-identical m/z values (isobaric ions), or when multiple peaks exhibit overlapping spatial distributions across the tissue image that could confound downstream annotation filtering.

Use when : (1) you have metabolomic data (NMR or MS-derived) and a continuous phenotype variable; (2) you need to quantify associations while controlling for known confounders (age, gender, disease status);

Use when you have one or more spectroscopic datasets (IR, Raman, UV-Vis, mass spectra, NMR) from an unknown compound and need to generate candidate molecular structures ranked by likelihood. Use this when retrieval-based approaches are infeasible (e.

Use when you have picked peaks (coordinates and intensities) from an INADEQUATE NMR spectrum and need to cluster them into networks representing individual compounds.

Use when you have preprocessed 1H NMR spectral data (e.g., from plasma or biological samples acquired on a 600 MHz instrument) and need to identify the chemical composition of a prominent but structurally ambiguous peak.

Use when when you have downloaded raw spectral datasets from multiple spectroscopic modalities (NMR, HSQC, COSY, IR) in their native or proprietary formats and need to convert them into aligned, standardized tensors for multimodal machine learning.

Use when when you have MS2 .mzML format data files from untargeted metabolomics or proteomics experiments and need to perform an initial annotation step by matching experimental spectra against known reference libraries (GNPS, HMDB, MassBank) with a defined precursor mass tolerance (e.g., 15 ppm).