Claude Skills by HolobiomicsLab

Use when when training or validating a deep learning model for molecular formula prediction from tandem MS/MS spectra, use this metric to track whether the model's predicted formula (including hydrogen atoms) exactly matches the annotated ground-truth formula.

Use when when preparing multi-formula MS/MS training data for a rescore model, if the raw positive examples show extreme imbalance (some formulas represented by hundreds of spectra while others have only a few).

Use when use this skill after training or fine-tuning a chemical formula transformer model on annotated tandem MS/MS spectra, when you need to measure whether the model's ranked formula candidates match ground truth.

Use when you need to generate a set of candidate chemical formulas for LC-MS/MS simulation—specifically when you want to populate a virtual mass spectrometer with realistic chemical structures drawn from a reference database (HMDB) or a uniform m/z distribution, and you need to apply m/z filtering.

Use when you have tandem mass spectra (MS/MS) with unknown precursor formulas and need to rank chemical formula candidates conditioned on observed fragment m/z values and precursor mass. Use this skill when fragmentation tree computation (e.

Use when when you have 1D NMR spectra (1H and/or 13C) of an unknown compound with up to ~19 heavy atoms and need to predict both molecular formula and connectivity without manual structure hypothesis generation.

Use when you have a collection of molecular fragments (e.g., from molecular decomposition, retrosynthesis, or synthetic planning) that must be matched to known fragment libraries or standardized representations before feeding them into a transformer assembly model.

Use when you have centroid-mode LC-MS AIF chromatograms processed through xcms and RAMClustR, a feature table with target m/z and retention time values, and access to fragment libraries (e.g., LipidPos for lipids).

Use when you have a tandem mass spectrum (MSMS) of a known or hypothesized peptide, along with its ProForma 2.

Use when immediately after loading raw MS/MS spectra from .mgf, .msp, or .mzML files, before generating the bag-of-fragments corpus or extracting neutral losses.

Use when when implementing fragment ion annotation in proteomics workflows and needing to determine whether neutral loss annotation (e.g., H2O: -18.010565, NH3: -17.026549) should be enabled to maximize peak interpretation.

Use when you have MS/MS spectra with assigned precursor formulas and need to annotate the chemical composition of individual fragment peaks for metabolite structure elucidation or fragmentation pathway analysis. Apply this skill when you want to avoid external fragmentation tree computation (e.

Use when you have preprocessed mass spectrometry fragmentation data (neutral losses and fragment masses extracted and noise-filtered) and want to discover hidden structural motifs across a spectral dataset in an unsupervised manner.

Use when you have custom lipid species (not covered by the 500,000+ built-in LipidMatch entries) that you need to match against experimental MS/MS data, or you are extending LipidMatch's coverage for specialized lipid classes.

Use when you have a collection of MS/MS spectra (≥2 spectra) and wish to identify fragmentation signatures common to subsets of those spectra.

Use when you have extracted a chemical mixture from a real mzML acquisition (e.g., Beer1pos), simulated the same chemicals through ViMMS using a chosen controller (e.

Use when you have loaded a table of entity–attribute pairs (e.

Use when you are evaluating or selecting FT-ICR MS software for a specific metabolomics workflow and need to assess which tools support your required analytical dimensions (e.g., Van Krevelen diagrams, PERMANOVA, thermodynamic indices, chemodiversity metrics, transformation networks).

Use when when you have raw or processed FT-ICR MS peak-abundance .

Use when you have FTICR-MS direct injection (mzML) data with identified chromatographic peaks and need to correct systematic m/z bias.

Use when you have loaded an FT-ICR raw spectrum (e.g., ESI_NEG_SRFA.d in Bruker or ThermoFisher .raw format) and need to identify the m/z positions and intensities of individual mass spectral peaks.

Use when you have received raw FT-ICR transient data from Bruker Solarix or ThermoFisher instruments and need to perform signal processing, apodization, calibration, or molecular formula assignment in CoreMS. The data must be in native vendor format (.d directory with ser/fid files, or .

Use when when you have Spectra objects in R and need to apply Python MS library functionality (spectral similarity scoring, filtering, normalization) without leaving the R environment, or when you want to create custom hybrid workflows that leverage both R and Python MS packages within a single.

Use when you have a set of query chemicals (chemical names or structures) and need to match them against a reference chemical library organized by type or category, with the goal of identifying structural similarity, functional group membership, or categorical assignment.

Use when you have an untargeted metabolomics feature table (m/z and retention time columns) and a statistical test result (p-value) per feature, but lack confident metabolite identifications.

Use when when you have abundance-normalized FT-ICR MS peak data with assigned molecular formulas and need to distinguish between richness (total number of distinct metabolites) and functional diversity (diversity in metabolic potential).

Use when after peak detection on a composite mass track has identified candidate peaks in a mass chromatogram, and before compiling the final feature table.

Use when when working with raw GC-MS data in NetCDF (ANDI) format that requires peak detection, baseline removal, and retention time alignment before spectral matching against reference libraries such as PNNLMetV20191015.MSL.

Use when you have raw GC-MS data (in netCDF or vendor format) containing overlapping chromatographic peaks from complex mixtures where individual compound spectra cannot be resolved by simple peak picking.

Use when when you have deconvolved GC-MS spectra (post-deconvolution output compatible with GNPS_GC input specification) and need to group them by chemical similarity to construct a molecular network.

Use when after BGC detection and clustering (producing GCFs) and metabolomics profiling (producing MFs with MS/MS spectra), when you have paired genomic and metabolomic data from the same microbial strains and need to rank which GCF–MF pairs are likely to represent true biosynthetic relationships.

Use when you have antiSMASH v5.0.0 BGC predictions from a set of microbial genomes and you need to integrate those predictions with GNPS metabolomic data (MS2 spectra and molecular families).

Use when when running metabologenomic RiPP detection pipelines (MetaMiner) on the same genomic dataset but with different input sequence formats (e.g., contigs.fasta vs. antiSMASH .final.gbk output), or when unexpected null results occur and input format choice is a plausible cause.

Use when a paired omics project JSON document contains genome identifiers (e.g. IMG IDs, NCBI accessions) but lacks corresponding organism names.

Use when you have assembled genomic DNA sequences (contigs in FASTA format, not antiSMASH or BOA output) and corresponding LC-MS/MS data (in MGF, mzXML, mzML, or mzData format) from the same organism, and you want to identify novel RiPPs by linking gene cluster predictions to observed mass spectra.

Use when after running qc_summary() on a filtered mpactr object and aggregating ion counts by filter status category (passed/failed).

Use when when you need to reproduce a computational workflow described in a GitHub repository, validate CI/CD pipeline definitions (e.g., GitHub Actions workflows), inspect source code structure, or execute local versions of automated tests.

Use when you need to obtain source code or computational workflows from a published repository, particularly when the article explicitly provides a GitHub URL and documents that the repository contains code required to regenerate published results (e.g., simulation outputs, figures, or tables).

Use when when you need to reproduce or validate a specific historical release artifact (e.g., a Semantic Release v1.0.

Use when when you need to verify that a GitHub Actions workflow (such as a development build or release pipeline) executes without fatal errors and produces expected artifacts. Use this skill when the workflow is already configured in a repository (e.g., a .yml file in .

Use when you need to verify that a GitHub Actions workflow (such as a development build release pipeline) has completed successfully, capture its build artifacts (installers, portable binaries, or packages), and document the workflow run metadata.

Use when when you have a Python package repository on GitHub and need to automatically verify that pull requests and commits pass unit tests and meet code quality standards before merge.

Use when you need to validate that a repository's automated build, test, or publish pipeline is functioning correctly on a target branch (e.g., release branch); when you want to confirm that workflow status badges in documentation accurately reflect current execution state;

Use when a GitHub repository displays a CI workflow badge (e.g., passing/failing status in README) and you need to verify that the reported status is accurate, reproduce the CI environment locally, or debug workflow failures.

Use when you have reproduced a release artifact locally (e.g., via Semantic Release or a build tool) and need to verify it matches the official GitHub release record.

Use when when you need to verify that a software package (such as MassQL) passes its periodic integration test suite as indicated by CI workflow badges in the project documentation, or when you must reproduce pass/fail results for package-testing workflows distinct from unit tests to establish.

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 computed frequent fragmentation patterns from a collection of MS/MS spectra using mineMS2, and you want to focus pattern interpretation on subsets of spectra that form meaningful network components (connected groups, cliques, or high-similarity pairs) in a GNPS molecular network.

Use when you have generated a GNPS mass spectral molecular network (in classical or feature-based mode) and want to annotate network nodes with substructural motifs from MS2LDA or chemical class information.

Use when when you have run a spectral networking job on GNPS (e.g. ProteoSAFe-METABOLOMICS-SNETS-V2) and need to reuse the network output files locally with MetaMiner or another tool that accepts spectral network input directories.