Claude Skills by HolobiomicsLab

Use when you have a tab-delimited feature table (m/z, retention time, intensities) from LC-MS preprocessing and need to group related ions (isotopologues, adducts, in-source fragments) into compound-level annotations with inferred neutral mass.

Use when after generating a feature table from mzML data (via Asari) and before performing MS1 or MS2 annotation.

Use when after feature detection and quality control have produced a feature table in TSV format from Asari or equivalent preprocessing.

Use when after feature table normalization and imputation are complete, immediately before MS1 and MS2 annotation.

Use when after feature detection from mzML files (e.g., via Asari) when you have a feature table with m/z, retention time, and intensity columns, and before MS1 or MS2 annotation.

Use when you have a pretrained encoder that produces fixed-size embeddings from MS/MS spectra (or similar spectral data), a tokenized target dataset of canonical SMILES strings representing molecular structures, and you need to learn a decoder that reliably reconstructs the molecular structure from.

Use when you have an unknown MS/MS spectrum (precursor m/z and fragment peak list) and a set of candidate molecular formulae, and you need to rank them by likelihood without access to a spectrum database or precomputed fragmentation trees.

Use when when you have an unknown MS/MS spectrum (m/z and intensity pairs) and need to assign a chemical formula and ionization adduct to the precursor mass, particularly when spectrum database lookups are unavailable or when you want to exploit learned patterns in fragmentation rather than.

Use when after running Enrichment() on a configured EnrichParam object (via KEGG_Enrich_PlotPanel or similar), when you have a full enrichment result table and need to reduce it to pathway hits meeting a specific significance threshold before visualization or export.

Use when you have a set of compounds (as SMILES strings or molecular structures) that need retention order predictions in a reversed-phase liquid chromatography (RPLC) system at eluent pH ~2.7, and you want to quantify prediction uncertainty rather than relying on a single model's output.

Use when you have pre-trained MLP and GNN models that generate different spectral predictions for the same metabolite candidates, and you want to combine them to improve ranking performance (average rank, Rank@K metrics) without retraining the base models.

Use when when you have retention order predictions from multiple independently trained models (e.g., ROASMI_1 through ROASMI_5) for the same set of compounds and need to estimate prediction confidence or identify compounds with high model disagreement.

Use when when comparing two preprocessed MS/MS spectra for compound identification and you need higher accuracy than dot product similarity provides.

Use when you need to search one or more query MS/MS spectra against large spectral libraries (hundreds of thousands to millions of spectra) and require real-time or near-real-time compound identification.

Use when you need to quantify the degree of match between two MS/MS spectra—either to validate that a denoised spectrum remains faithful to a reference ground-truth spectrum, or to rank candidate library matches for a query spectrum.

Use when when deploying Galaxy-M or similar multi-component metabolomics platforms that depend on heterogeneous runtime environments (Python, R, MATLAB, WINE) across multiple operating systems (Ubuntu 14.

Use when when a user uploads a JSON project document to the Pairing Omics Data Platform and you need to determine whether it satisfies the platform's data structure requirements, including all mandatory fields, proper data types, and constraint satisfaction for paired omics metadata (e.

Use when you are developing or comparing new data-dependent acquisition (DDA) strategies in ViMMS and need to evaluate how well each strategy fragments sampled compounds from the HMDB database.

Use when after completing an Environment simulation run with save_eval flag enabled, when you need to preserve the EvaluationData object containing scan provenance, chemical source definitions, and fragmentation events for later inspection, validation, or reanalysis without re-running the full.

Use when you have feature pairs identified by temporal correlation in direct-injection MS data and need to confirm their relationship is consistent with known adduct/fragment mass shifts.

Use when after feature detection and alignment on raw MS data, when you have a list of unknown feature m/z values and need to assign them to known xenobiotic metabolites or their predicted biotransformation products.

Use when you have a GC-MS dataset in CSV format with retention times, base peak m/z values, component areas, and compound names, and you need to identify whether specific query chemicals are present in your samples and retrieve their -match factors (scoring the confidence of the spectral match) and.

Use when you have a preprocessed LC-MS peak table exported from peak-picking software (e.g., MS-DIAL) in Excel format with three logical compartments: sample annotation (rows), feature annotation (columns), and abundance matrix (numeric values).

Use when when preparing to run QCxMS2 or similar multi-tool orchestration software that depends on five or more external programs with strict version floors.

Use when before initiating raw file conversion or feature extraction, when you have a heterogeneous collection of raw LC-MS files (.raw or .mzML) and sample information scattered across instrument logs, sequence files, or spreadsheets.

Use when when you have LC-MS raw data and need to process it through a feature detection and quantification pipeline in KNIME, but lack a structured mapping between sample identifiers, experimental conditions, and the raw LC-MS runs.

Use when you are in the Bayesian optimization loop after fitting a Gaussian Process model to observed LC gradient runs, and you need to propose the next gradient to evaluate.

Use when after validating a peak table (either standardized format or software tool–generated format) and its corresponding label file, before applying any NOREVA assessment functions (normulticlassqcall, nortimecourseqcall, etc.).

Use when your project JSON document contains genome identifiers but lacks organism name or taxonomic annotations. The platform needs to auto-populate these fields to enable browsing and cross-linking with public genomic databases. Trigger this skill when you have genome IDs (e.

Use when you have constraint-based metabolic models of multiple cell lines, experimental measurements of extracellular metabolite concentrations at two timepoints (e.

Use when after running MS1 extraction and prescreening on mzML files with assigned adducts and tags, when you need to inspect detected compounds visually to verify peak shape, confirm retention time consistency across samples (e.

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

Use when after running tardisPeaks() in screening mode or peak detection mode, when you need to visually confirm that target compounds are visible in the expected m/z and retention time windows, verify that peak integration boundaries are correct, diagnose whether sawtooth artefacts are present.

Use when after features have been grouped by retention time similarity and abundance correlation across samples, but before downstream annotation or compound identification.

Use when you have Thermo Fisher Orbitrap .raw files and need to locate and quantify specific peptide precursor ions (e.g., iRT calibrants, synthetic standards, or putative identifications).

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 you have generated candidate peptide-spectrum matches from a spectral library search (especially open modification searches using cascade strategies) and need to assign statistical confidence to those matches. Use it whenever the scoring metric (e.

Use when when you have executed database search pipelines (Dereplicator, VarQuest, or Dereplicator+) on centroided LC-MS/MS spectra in MGF format and obtained match results with associated p-values and false discovery rates.

Use when after high-scoring spectral library matching (e.g., EQ module output) of LC-MS/MS data yields candidate lipid annotations; when spectral similarity alone produces false positives and you have computed relative retention time intervals across species cohorts;

Use when you have implemented an automated feature annotation or adduct detection module and need to verify that assigned labels (e.g., [M+NH4]+, [M+K]+, [M+H2O+H]+, [M-H2O+H]+) are accurate and do not produce erroneous assignments on a reference feature set.

Use when after running Paramounter's peak-height optimization on XCMS CentWave-extracted metabolomic features, if the downstream analysis or feature validation reveals an unacceptable rate of false positives, or if the extraction workflow is experiencing software crashes or timeout failures due to.

Use when when a metabolite feature has been assigned a top-rank lipid annotation (e.g., LPC(14:0)) but you need to assess whether related lipid species containing the same fatty acyl chain(s) (e.g., PC fragments with 14:0 acyl chains) also match the observed spectrum with lower scores.

Use when after initial retention-time-based feature grouping (e.g., using SimilarRtimeParam with a 20-second window), apply this skill when you need to split large feature groups into more homogeneous sub-groups.

Use when after peak picking (e.g., via MS-DIAL) and quality control filtering, when you have a raw feature abundance matrix with intensity values across multiple samples and need to make intensities comparable before statistical testing or multivariate analysis.

Use when after initial retention-time-based feature grouping has been performed on LC-MS data but before final EIC similarity refinement.

Use when after initial retention-time-based feature grouping when you have groups of multiple features at similar m/z and retention time but need to determine which features actually arise from the same compound.

Use when after loading an MZmine3-exported feature quantification table and identifying blank sample columns, when you need to remove features with significant intensity in procedural blanks before proceeding to batch correction and statistical analysis.

Use when you have completed peak detection and feature alignment in metabolomic LC-MS processing and suspect systematic errors in peak integration or feature misalignment across your sample cohort.

Use when when a traditional peak extraction pipeline (e.g., XCMS) has generated a feature table from LC-MS data but fails to detect known or suspected compounds present in your sample. Specifically, when you have a suspect database (e.

Use when after chromatographic peak detection and feature detection in LC-MS preprocessing, when you have a set of detected features (m/z, retention time, intensity) and need to consolidate redundant or related ion signals into compound-level feature groups before downstream statistical or.