Claude Skills by HolobiomicsLab

Use when after performing assignment operations (assign_ri, assign_smiles) or combining multiple library objects (e.

Use when evaluating alternative implementations of data storage or retrieval strategies in R objects—specifically when deciding whether to eagerly populate all columns in a data frame slot (e.

Use when you have tabular metabolomics data (tab-delimited or Sciex OS format) and need to apply a specialized R package's analysis pipeline—such as mzQuality—that requires sequential function calls (readData → buildExperiment → doAnalysis) to construct, validate, and transform experiment objects.

Use when when you have a new or updated R package available via a non-CRAN repository (such as r-universe) and need to verify it installs cleanly, passes R-CMD-check compliance, and is ready for downstream workflow execution.

Use when before running any R function that wraps compiled C# methods or system executables (e.g., rawrr::readSpectrum), especially when the package depends on language runtimes (.NET, Mono) or proprietary third-party assemblies that must be downloaded and configured separately.

Use when when deploying an R package from a non-CRAN repository (e.g., r-universe, Bioconductor, GitHub), or when verifying that a package build is reproducible and meets CRAN submission standards prior to integration into a larger analysis pipeline.

Use when you have multiple MSP (mass spectrum) library files to read and merge in R, and your computational task is time-consuming (e.g., structure extraction, SMILES assignment, or RI assignment) and you have a multi-core system available.

Use when you have a Spectra object in R and need to apply a specialized Python MS algorithm (e.

Use when when working with large metabolomics peak tables (e.

Use when you need to understand or validate whether calling filter_mispicked_ions() (or similar R6 filter methods) with different copy_object settings will mutate your original data object in memory or preserve it.

Use when when you need to support fast, non-sequential lookups (e.

Use when you have a large compressed scientific data file (e.g., indexed gzip mzML) where you need to retrieve specific spectra, chapters, or records by integer ID without loading the entire file into memory.

Use when you have a large mzML file or text corpus (e.g., Moby Dick, proteomics run) stored in compressed or database format and need to retrieve specific spectra or chapters by ID without sequential iteration or full decompression.

Use when when you have run a pathway ranking method (such as PALS/PLAGE) on clean metabolomics data and wish to assess how sensitive the resulting pathway activity rankings are to realistic data quality issues—specifically Gaussian noise and random peak dropout—which are prevalent in untargeted.

Use when after running annotateRC on LC–MS AIF data when you need to inspect whether a feature has multiple plausible annotations (e.g., isobaric lipids, isomers with the same fatty-acyl chain) or when the rank-1 annotation confidence is borderline and alternatives should be evaluated.

Use when after generating ranked predictions of chemical formulas or subformulas for MS/MS spectra (from a neural network model like MIST-CF's formula transformer), compare predicted assignments against ground-truth reference labels to measure ranking quality.

Use when you have raw GC-MS output files (vendor formats or netCDF) from a chromatography instrument and need to prepare them for automated peak deconvolution and spectral analysis.

Use when when you have a raw mass spectrometry file (e.g., Thermo Orbitrap .raw) and need to establish the measured throughput of a spectral reading function (e.

Use when you have a Thermo Fisher Scientific .raw file (e.g., Q Exactive HF, Orbitrap) and need to extract specific spectral scans, chromatographic traces, scan-level metadata, or file-level headers programmatically—e.

Use when when annotating .msp mass spectrometry files with chemical structure metadata and you need fast, offline molecular transformations (SMILES↔InChI, canonical SMILES generation) without network latency or service availability constraints;

Use when you have RNA-seq read count data and a metabolic model with GPR rules, and you need to assess how differential gene expression translates into differential metabolic reaction capacity across multiple biological conditions or cell lines.

Use when you have computed RAS (Reaction Activity Scores) from transcriptomics and GPR rules, RPS (Reaction Propensity Scores) from intracellular metabolomics via mass-action kinetics, and flux distribution differences (FFD) from constraint-based sampling across multiple biological samples.

Use when you have measured intracellular metabolite abundances (LC-MS or similar) across multiple cell lines or conditions and a stoichiometric metabolic model (with reaction-metabolite associations) to estimate how differences in substrate availability—independent of gene expression—translate into.

Use when you have a trained NeatMS neural network model and labelled peak validation data, and need to select an operational classification threshold or understand how TPR and FPR vary across probability thresholds (e.g., 0.00–0.99).

Use when when you have computed similarity scores (cosine, modified cosine, Spec2Vec, or other metrics) between a set of query spectra and a reference library with known structural annotations (InChIKey), and you need to evaluate how well each scoring method distinguishes true structural matches.

Use when you have extracted tabular data into an intermediate JSON form and need to restructure records by mapping input fields to output dictionary keys, collating multiple records under a single grouping field, filtering records by test conditions or exclusion rules, or applying custom.

Use when when you have a tagged tabular file (Excel or CSV) with columns marked using export tag syntax (e.g., #study.id, #subject.id, #.

Use when after computing expected adduct ions for a metabolite using a derivatizing matrix ruleset, validate the predicted m/z values and adduct formulas against a curated reference dataset that documents which ions are actually produced by that matrix under standard ionization conditions.

Use when when you have IM-MS lipidomics data with measured CCS values from samples spiked with U13C labeled internal standards, and you need to assess systematic CCS bias or enable CCS correction by comparing measured lipids against known library entries with validated CCS values.

Use when when performing m/z domain calibration on FT-ICR or high-resolution MS data and the initial calibration attempt finds fewer than 5 reference m/z matches within the standard PPM window (typically ±1–5 ppm).

Use when you have executed a structure annotation pipeline (like BAM) on a validation dataset for which ground-truth molecular structure annotations exist, and you need to assess whether the pipeline's predictions match the reference annotations at the required sensitivity and specificity.

Use when when you have trained multiple regression models (e.g., using different feature sets: descriptors-only, fingerprints-only, or combined) on the same training data and need to objectively rank their generalization performance on unseen test data.

Use when you have trained two or more regression models (e.g., original vs. alternative GNN architectures) on the same training set and need to evaluate which generalizes better on held-out test data.

Use when you have model predictions and ground-truth labels for a test set and need to assess how well the trained model generalizes to unseen data. Typical triggers: after executing inference on a held-out test partition, after cross-dataset evaluation (e.

Use when when you have trained a regression model on experimental retention times or similar continuous molecular property predictions and need to quantify its generalization performance on held-out test data.

Use when when you have fitted one or more regression models (linear or polynomial) to external calibration standards in MS data and need to verify model adequacy before applying it to unknown samples.

Use when when a software project has reached a stable milestone (v-tagged commit) and you need to produce official distribution artifacts with verified version metadata, checksums, and release documentation that can be validated against a published GitHub release record.

Use when you need to validate that a software project''s release branch is stable and ready for distribution. Specifically, use it when: (1) a release tag has been created or code pushed to a release branch; (2) you need to confirm that CI/CD pipelines execute without failures;

Use when you need to validate that a scientific software project's continuous integration pipeline is functional and producing reproducible builds—particularly before releasing new versions, after merging changes to release branches, or when troubleshooting build failures that block distribution of.

Use when when starting a fresh ENPKG installation, you have a GitHub URL (e.g., https://github.com/enpkg/enpkg_full or https://github.

Use when you have identified a published method (e.g., MIST-CF for chemical formula ranking from mass spectra) whose source code and trained weights are available in a public repository, and you need to apply that method to new experimental data (e.

Use when when you need to understand how a complex feature or architectural pattern was implemented in a codebase, particularly when the current README or documentation does not fully explain the control flow, decision criteria, or parameter passing between subsystems.

Use when when you need to inventory a collection of related web applications or tools distributed across multiple repositories, discover their live deployment URLs, trace their associated publications, and verify accessibility and metadata completeness.

Use when you need to validate that a published software tool (e.g., MassQL) executes correctly in your environment, reproduce published results, or contribute to development.

Use when when you encounter a multi-module scientific software project (e.

Use when a scientific Python package is being moved to a new GitHub organization with different structural conventions (e.g., from a personal lab account to a community-led organization like metabolomics-cloud or scverse).

Use when training a contrastive learning model on ion image data (mass spectrometry imaging) where augmented pairs of the same ion image must maximize similarity while different images minimize similarity.

Use when after obtaining 512-dimensional representation vectors from the Encoder module, when you need to compress these vectors for visualization, clustering, or downstream classification tasks on mass spectrometry imaging data while maintaining interpretability of ion relationships.

Use when evaluating whether a published computational method can be independently executed: (1) source code is claimed to be available but repository structure, build instructions, or dependency specifications are incomplete;

Use when when you have TCN-predicted candidate formulas with ranked scores and need to train a Siamese rescore model to re-rank those candidates.