Claude Skills by HolobiomicsLab

Use when you have an untargeted metabolomics dataset with partial metabolite annotations (from database matching or prior curation) and need to extend annotation coverage to unannotated metabolites.

Use when implementing data replacement methods (such as `[<-`, `$<-`, `mz<-`, `intensity<-`, `peaksData<-`) in a writable MsBackend subclass.

Use when you have peak-picked UHPLC-HRMS/MS data (from Q-Exactive orbitrap, Agilent, Bruker, or SCIEX Q-TOF instruments) with both MS1 and MS/MS fragment spectra, and you need to assign lipid identities to detected features using in-silico fragmentation patterns.

Use when you have aligned feature tables from LC–MS/MS, corresponding in silico annotations (from GNPS/ISDB or SIRIUS), and metadata describing sample origin. Use it to rank extracts by the proportion of sample-specific, unannotated features—a proxy for structural novelty.

Use when when you have multiple independently trained models (e.g., ROASMI_1 through ROASMI_5) making predictions on the same set of compounds or samples, and you need to assess confidence in individual predictions.

Use when when you have a trained neural network (e.g., a Siamese model predicting molecular structural similarity) and need to identify and filter unreliable predictions before using them in downstream analysis. Apply this skill when raw model performance is suboptimal (~0.

Use when when refactoring or adding workflow branching logic that routes spectral results into separate processing paths (e.

Use when you have extended a neural network model class (e.

Use when when implementing new scoring components (inchikey score, neighbourhood score, or similar structural/spectral similarity metrics) that are integrated into an MS/MS candidate re-ranking pipeline.

Use when you have a preprocessed feature table from non-targeted LC-MS/MS metabolomics data (after data merging, cleanup, blank removal, and batch correction) and need to test whether individual metabolomic features show statistically significant differences between experimental groups or.

Use when when you have an existing real mzML file from a metabolomics LC-MS/MS acquisition (e.g., beer or urine samples) and need to populate a virtual mass spectrometer with the actual chemicals that were measured, so that you can replay the acquisition with alternative fragmentation strategies (e.

Use when you have UPLC-HRMS raw data (ThermoFisher, Agilent, or compatible vendor format) from water samples or environmental matrices containing unknown organic pollutants, a Windows environment with ≥16 GB RAM and ≥2 GB NVIDIA GPU, and you need compound identification with confidence scores and.

Use when when annotating MS/MS spectra against spectral libraries and chimeric spectra (spectra containing fragments from multiple precursor ions) are suspected or known to be present in your dataset.

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 high-dimensional embedding vectors from pretrained models (e.g., MSBERT) and need to verify that the learned representation space groups spectra by chemical similarity without labeled training data.

Use when when you have raw untargeted LC-MS metabolomics data and need to detect low-quality or mis-integrated peaks in an XCMS-processed xcmsSet object before performing metabolite annotation, statistical analysis, or biomarker discovery.

Use when you have a peak-intensity matrix from untargeted LC-MS analysis (raw detected peaks with m/z and intensity values) and need to assign putative metabolite identities.

Use when you have authentic metabolite standards analyzed by LC-MS in both positive and negative ESI modes (converted to .

Use when you have untargeted MS2 spectral data (from LC-MS/MS or similar instruments) and need to assign metabolic pathway context to detected compounds when standard spectral library matching is unavailable or insufficient.

Use when after imputation and signal drift correction (OUKS steps 3–4), when you have a QC-annotated feature intensity table with replicated QC samples and need to assess which features have stable, reproducible signals before filtering and statistical testing.

Use when you have two LC-MS feature tables (each with m/z, retention time, and intensity columns) from independent untargeted metabolomic experiments or replicates and need to establish one-to-one feature correspondence across them to compare abundances, detect shared metabolites, or merge datasets.

Use when you have a feature table from untargeted metabolomics (with m/z, retention time, and p-values from differential abundance testing) but lack or wish to bypass metabolite annotation.

Use when you have an untargeted metabolomics feature table (m/z values, retention times, p-values from statistical testing) and need to infer which metabolic pathways are active without performing metabolite identification.

Use when you have LC-MS/MS data acquired in DDA mode from untargeted metabolomics experiments and need to remove chimeric (co-fragmented) MS/MS spectra that result from multiple precursor ions fragmented simultaneously.

Use when after loading and filtering search results from two or more DIA-MS analysis tools at a specified Q-value cutoff, when you need to summarize which analytes are identified by all tools, by specific subsets, or uniquely by individual tools.

Use when when you have a multi-step computational workflow (e.g., peak detection, filtering, manual review) implemented in R and need to expose it to end-users who lack R expertise.

Use when when you need to retrieve mass spectrometry spectrum data from a metabolomics repository but only have a USI string (e.g., 'mzspec:GNPS:TASK-c95481f0c53d42e78a61bf899e9f9adb-spectra/specs_ms.mgf:scan:1943' or 'mzspec:MASSBANK::accession:SM858102').

Use when you have a Universal Spectrum Identifier (USI) string referencing a spectrum in a supported metabolomics repository (GNPS, MassBank, MetaboLights, Metabolomics Workbench, MassIVE, or MS2LDA) and need to create an embeddable, scannable reference for publication or data integration that.

Use when when you have a USI string (e.g., mzspec:MTBLS1124:QC07.

Use when you have a USI accession (e.g., 'mzspec:MSV000082283:f07074:scan:5475' or 'mzspec:PXD000561:Adult_Frontalcortex_bRP_Elite_85_f09:scan:17555') pointing to a publicly deposited tandem mass spectrometry scan in a GNPS or ProteomeXchange repository, and you need to retrieve and instantiate.

Use when you have a USI string referencing a spectrum in an online public repository (PRIDE, MassIVE, etc.) and need to load its raw spectral data without downloading the entire dataset file.

Use when when you have a USI string (e.g., mzspec:GNPS:TASK-d93bdbb5cdda40e48975e6e18a45c3ce-f.mwang87/data/Yao_Streptomyces/roseosporus/0518_s_BuOH.

Use when when you have scored GCF-MF (gene cluster family–molecular family) links using two or more complementary scoring approaches (e.g., standardised strain correlation and IOKR), and you need to determine whether combining scores improves discrimination of true links from false positives.

Use when you have a QC-annotated LC-MS feature intensity table (CSV or data frame) with replicate QC samples and biological samples from multiple batches or run orders, and you need to assess which features maintain consistent signal intensity across technical replicates (within-group) relative to.

Use when you have raw MS data files directly from a vendor instrument (Thermo .raw, Agilent .d, Waters .ms, etc.) and need to process them through AriumMS or any other metabolomics pipeline that accepts only .mzXML or .mzML formats.

Use when you have mass-spectrometry raw data files from multiple vendors (e.g., Thermo, Waters, Bruker, Sciex) and/or mixed standard formats (mzML, mzXML) that must be processed by a single analysis pipeline.

Use when when Rapid QC-MS receives vendor-format LC-MS acquisition files from instrument data folders and must prepare them for automated QC checks and MS-DIAL processing. Specifically: input files are in proprietary vendor formats (Thermo .raw, Bruker .d, Sciex .

Use when when ingesting raw LC-MS/MS output from a mass spectrometry instrument and you need to prepare it for metabolite identification, fragmentation tree computation, or molecular formula annotation.

Use when you have acquired raw mass spectrometry data in vendor-proprietary formats (ThermoFisher, Agilent, or equivalent) and need to analyze it using MSThunder for unknown organic pollutant identification.

Use when you have one or more vendor mass spectrometry raw files (Thermo .raw, Agilent .d, Sciex .wiff2, or other MSConvert-supported formats) that must be converted to Aird format for batch processing, cloud deployment, or integration with downstream analysis tools.

Use when you have mass spectrometry raw data in a proprietary vendor format (Thermo .raw, Agilent .d with or without ion mobility, Bruker ion mobility .d, or mzML) and need to enable reproducible, language-agnostic access to multidimensional spectra (e.

Use when you have a directory containing mass spectrometry data files from multiple instrument vendors (Thermo, AB Sciex, Agilent, Bruker, etc.) and need to convert them to a common format (Aird or mzML).

Use when you have a set of chemical compounds (with known retention times and intensities) loaded into a ViMMS IndependentMassSpectrometer and need to simulate a specific MS/MS fragmentation strategy (e.

Use when you need to create a synthetic chemical population for testing data-dependent acquisition (DDA) strategies in a simulation environment before committing to real mass spectrometry analysis.

Use when when you have generated or extracted a chemical mixture (via DatabaseFormulaSampler, ChemicalMixtureCreator, or ChemicalMixtureFromMZML) and need to establish a virtual instrument to simulate scan acquisition.

Use when after applying two or more LC-MS alignment algorithms (such as XCMS and ncGTW) to the same dataset, use this skill to determine which produces warping functions. Specifically, apply this when: (1) you have detected or suspect misaligned features in XCMS output (e.

Use when you have access to a project README or repository documentation (Zenodo deposit, GitHub, or local clone) describing multiple domain-specific web applications, and you need to produce a machine-readable inventory of those applications with verified live URLs and associated publications for.

Use when you have a user-submitted spectrum with associated domain context metadata (e.g., selected as 'microbial origin', 'plant tissue', 'food sample') and need to route that spectrum to the appropriate domain-specific MASST application for searching.

Use when when you have a parsed mass spectrum (precursor m/z, ionization mode, collision energy, and fragment peak list as m/z–intensity pairs) and need to obtain molecular fingerprint predictions, de-novo candidate structures, or chemical class annotations without maintaining local neural network.

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.