Claude Skills by HolobiomicsLab

Use when you have two peak-picked, conventionally aligned untargeted LC-MS metabolomics datasets (metabData objects) acquired under different conditions and need to identify overlapping <m/z, retention time> features across them.

Use when you have a raw peak-picked untargeted LC-MS dataframe with columns containing mass-to-charge (m/z), retention time (rt), feature identifiers, adduct annotations, and sample measurements in non-standard column names or mixed column sets.

Use when you have Nightingale Health 1H-NMR metabolomics data (feature matrix with named metabolite columns) and need to compute predicted metabolic age for each sample, typically to assess whether individuals' metabolic profiles align with or diverge from age-expected trajectories.

Use when when processing data with available MS2 spectra (DDA acquisition) after MS1 peak picking has been completed, and you seek to identify additional metabolic features or validate existing peak picking results through MS2 recognition.

Use when you have a generic constraint-based metabolic model (SBML format) and cross-sectional omics data (RNA-seq, intracellular metabolomics, YSI or bioanalyzer extracellular flux measurements) for multiple biological samples and need to create sample-specific models that discriminate whether.

Use when you have an untargeted metabolomics feature table (with m/z, retention time, and statistical significance values) and want to predict which metabolic pathways and functional modules are active in your sample, but you lack confident metabolite identifications or wish to bypass the.

Use when you have a ranked list of metabolite identifiers (PubChemCIDs, KEGG IDs, or chemical names) from differential abundance or ANOVA testing and need to determine which metabolic pathways are overrepresented or enriched among the most significant features.

Use when you have integrated transcriptomics, intracellular metabolomics, and extracellular flux ratio data from multiple cell lines or conditions, and need to determine whether observed differences in metabolic fluxes originate from gene expression changes, substrate availability changes, or both.

Use when you have a MetaboLights dataset identifier (e.g., MTBLS1124) and need to download a specific mzML file (e.g., QC07.mzML) from the public repository for visualization, quality control assessment, or integration into a metabolomics workflow. The USI format mzspec:MTBLS1124:QC07.

Use when you have a raw LCMS nontargeted metabolomics abundance table spanning multiple injections with embedded pooled technical replicate (PREF) or internal standard injections distributed across the run sequence.

Use when you have intracellular metabolomics abundance data (measured metabolite concentrations) from multiple biological replicates collected from two or more cell lines or conditions, and you need to create a normalized, cell-line-level metabolite dataset before computing Reaction Propensity.

Use when you have intracellular metabolomics data paired with constraint-based metabolic model predictions and need to identify metabolically controlled reactions.

Use when you have extracted a large feature set of m/z values (hundreds to tens of thousands) from a Cardinal MSImagingExperiment object or similar MS dataset and need to assign putative metabolite identities using public structural databases.

Use when when you have an untargeted metabolomics dataset from HPLC–MS (e.g., mzML, NetCDF) with detected peaks of unknown identity, and you need to disambiguate or validate tentative compound annotations by comparing observed retention time against machine-learning predicted retention time.

Use when you have a feature intensity table with feature metadata (m/z, retention time) extracted from XCMS or MS-Dial, and you want to cross-reference each feature against a known-compound database to assign standardized confidence levels.

Use when you have ESI/LC-MS test spectra requiring candidate metabolite ranking, pre-trained MLP (NEIMS) and GNN baseline models are available or can be trained, you seek quantified improvement over single-model average rank performance (baseline MLP shows ~339 average rank), and your evaluation.

Use when you have MZmine-aligned features with m/z and retention time, and you have generated spectral annotations from two or more database sources (e.g., GNPS/ISDB spectral matching and SIRIUS in silico structure elucidation).

Use when you have ESI/LC-MS test spectra requiring metabolite annotation and need to compare ensemble-based neural network predictions (ESP) against a baseline MLP model to quantify performance gains.

Use when when annotating large-scale untargeted metabolomics datasets where reference library coverage is incomplete and you need to infer metabolite identities for unannotated compounds by propagating annotations from seed metabolites (database matches or prior curation) across both.

Use when after running the annotateRC function on LC-MS All-ion fragmentation (AIF) features and obtaining a populated annotations object with ranked candidate matches, use this skill when you need to persist results to disk for archival, sharing, or downstream interpretation (e.

Use when you have a feature table with candidate metabolite annotations (m/z, retention time, chemical identifiers) from MS/MS spectra or external tools (SIRIUS, GNPS), sample metadata linking samples to organisms, and you need to prioritize candidates by both annotation quality AND biological.

Use when you have paired metabolomics data (MS/MS spectra and feature quantification) linked to organismal or tissue taxonomy, and you want to reduce false positive annotations and improve annotation rank by filtering candidate metabolites to those chemically plausible within the given taxon.

Use when you have a set of candidate metabolites for an unknown compound detected in a liquid chromatography–mass spectrometry (LC-MS) experiment, predicted RTs from a trained DNN model, and access to calibration molecules (minimum 10) that connect your observed chromatographic method to a source.

Use when after running annotateRC() on LC-MS AIF features, when you need to validate whether a feature's rank-1 annotation is reliable or when you suspect that structurally similar metabolites (e.

Use when you have an untargeted mass spectrometry spectrum (MS/MS data) and a set of candidate molecules from PubChem or similar databases, and you need to rank candidates by likelihood of being the true molecular annotation.

Use when after running RAMClustR clustering on XCMS-detected LC-MS features in positive ionization mode, when you need to assign molecular weights to compound clusters and want to cross-validate the two available scoring methods (findMain and RAMClustR internal scoring) to identify cases where they.

Use when after preprocessing, imputation, and batch correction of LC-MS peak tables when you need to group redundant or related feature measurements (e.g., [M+H]+ and [M+Na]+ adducts, or isotope peaks) into metabolite-level clusters before statistical testing or identification.

Use when after normalizing a metabolomic feature matrix when you have both non-QC (study) samples and QC (quality-control) replicates in the same experiment. Use it to remove features that are poorly reproducible or show inconsistent signal across samples relative to instrument/technical variation.

Use when when performing untargeted metabolomics annotation at scale and you need to estimate false discovery rates for candidate metabolite identifications.

Use when when you have paired tandem MS spectra and known molecular structures (SMILES or fingerprints) and want to annotate novel spectra by retrieving similar structures from a reference database without relying on spectral database matching.

Use when after generating or filtering transformation products using generateTPs() or filter(), when you need to annotate MS/MS spectra using MetFrag and require a database of candidate structures (parent compounds and/or their TPs) in a format MetFrag can read.

Use when you have simulated or experimental mzML data from two or more fragmentation controllers (e.

Use when you have a measured m/z value from spatially-resolved metabolomics or mass spectrometry imaging and need to assign a molecular formula with high confidence. Use it specifically when you have access to a pre-constructed formula network (KnownSet database) linking 2.

Use when after selecting statistically significant features from multi-assay LC-MS metabolomics datasets (e.g., via MB-VIP and permutation testing with p < 0.01).

Use when after MS1 feature extraction from mzXML files when you have: (1) a feature table with m/z, retention time, and intensity values; (2) DDA (Data-Dependent Acquisition) mzXML files containing MS2 fragmentation spectra; and (3) a reference spectral library in MSP format.

Use when you have normalized abundance data from LC-MS/MS for multiple samples classified into three or more discrete groups (e.

Use when after PuInc_seeker has identified putative incorporations (m/z features showing significant fold-change and p-value signals between labeled and unlabeled sample groups) and you need to assign base peaks—the most intense isotopologue signals—and validate isotope-pair mass gaps match.

Use when after XCMS feature detection, grouping, retention time correction, regrouping, and missing value filling on LC-MS or GC-MS data, when you have an aligned feature table with retention times and intensity profiles across multiple samples and need to collapse redundant features into.

Use when you have peak-picked LC-MS metabolomics data in a tabular format (R data frame) with columns for mass-to-charge ratio, retention time, feature identifiers, adduct annotations, and sample measurements, but the column names do not follow a standard naming convention.

Use when after m/z grouping and pairwise alignment detection when you have a metabCombiner object containing candidate feature pair alignments and need to select a subset of mutually abundant, high-confidence anchors to anchor a nonlinear retention time mapping spline.

Use when after retention-time correction and data alignment have been completed on centroided LC-MS data (mzML or mzXML format).

Use when you have one or multiple raw mzXML/mzML files from DDA, DIA, or full-scan LCMS analyses and need to detect, align, and quantify metabolite features across samples to create a unified feature matrix before MS2 annotation or in-source fragment analysis.

Use when you have a raw metabolite abundance matrix (e.g., from MSPrep or another LC-MS/MS pipeline) with many features and samples, and you observe that a substantial fraction of metabolites are missing (NA or zero-valued) across replicates.

Use when after drift correction and before imputation, when you have a MetaboSet object with LC-MS peak abundances and need to remove features with insufficient detection consistency across QC samples.

Use when immediately after generating a feature table (m/z, retention time, intensity) from centroided mzML data when you need to collapse multiple feature detections of the same compound (arising from different ionization states, charge states, or isotope patterns) into unified empirical compound.

Use when after imputation and batch-effect correction (OUKS steps 3–4) have been completed on your LC-MS feature-intensity table, and before statistical hypothesis testing.

Use when you have two independent LC-MS untargeted metabolomic feature tables (e.

Use when after data merging and cleanup (blank removal) and before univariate or multivariate statistical analysis, when your merged feature table (samples as columns, metabolite features as rows) contains samples processed in different MS batches or instrumental runs that may introduce systematic.

Use when you have loaded two or more nontargeted LCMS feature tables from the same analytical method that contain m/z, retention time, and intensity values, and these datasets exhibit differences in metadata scale, distribution, or format that could confound cross-dataset feature matching or.

Use when when you have completed feature detection in MZmine3 or similar tools and produced a feature quantification table (rows = features, columns = samples with intensity values), and you possess a separate sample metadata file (sample identifiers, treatment groups, batch information).