Claude Skills by HolobiomicsLab

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 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 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 you have run ORA on simulated metabolite sets with known null conditions (no true pathway enrichment) and need to measure how detection coverage, pathway database size, or other experimental parameters inflate Type I error rates.

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 you have computed empirical p-values from randomized sampling (e.

Use when after integrating transcriptomics-derived (RAS), metabolomics-derived (RPS), and extracellular flux constraints into cell-relative metabolic models, sample the feasible flux region when you need to: (1) visualize and compare the metabolic phenotype distributions across biological samples.

Use when when you have a trained GNN model for molecular property prediction (e.g., collision cross section) and need to identify which graph structural features—atomic properties, bond types, or higher-order graph descriptors—are driving the model's predictions.

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 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 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.

Use when you have a feature list with assigned molecular formulas and m/z values from non-target HRMS analysis, and you need to identify and rank potential PFAS compounds among thousands of detected features.

Use when after peak detection and feature extraction have produced a composite feature table with m/z, retention time, and intensity values for individual samples.

Use when after training a neural network model (e.

Use when when you have molecular structures, a regression target (e.g. retention time), and want to establish whether one class of molecular features (e.g., fingerprints) outperforms another (e.g., descriptors) or whether combining them yields marginal gains.

Use when after initial retention-time-based feature grouping (e.g., ±20 s window) when you need to separate co-eluting features that are chemically distinct. Triggers include: (1) large feature groups (>2–3 members) suspected to contain multiple compounds;

Use when after sample alignment and isotopologue/adduct grouping are complete, when you need to associate MS2 spectral data (DDA-acquired) with the consolidated feature groups to enable MS/MS-based compound annotation or to bundle MS1 quantification with MS2 evidence.

Use when you have a feature list with m/z values from HRMS data and need to identify homologous PFAS series to prioritize suspect screening.

Use when after peak picking and sample alignment have produced an aligned feature table with m/z and retention time coordinates.

Use when when you have high-resolution tandem MS/MS spectra in mzML, mzXML, or MGF format and need to cluster or search millions of spectra efficiently.

Use when after blank masking and sample dropping, when you have a feature table with intensity values that exhibit systematic variation across sample collection batches or instrument runs.

Use when after sample alignment in untargeted LC-MS workflows, when the aligned feature table contains missing (NA or zero) intensity entries for features that are detected in some samples but fall below the instrument detection limit or are absent in others.

Use when a user supplies a custom feature list from external feature-finding software (vendor tools, alternative open-source pipelines) instead of using pyOpenMS automatic detection, or wishes to augment/replace pyOpenMS results with pre-processed features.

Use when after extracting and encoding molecular descriptors and structural features (atom types, bond connectivity, graph topology) from SMILES strings into a fixed-size numerical tensor, before passing the feature matrix to the PS2MS deep learning model for NPS prediction.

Use when after initial retention-time-based feature grouping (e.

Use when you have aligned feature tables (CSV format) paired with MS2 spectral data (MGF or mzML files) and need to compare chemodiverse samples with poor feature overlap or strong retention-time shifts across different LC methods or mass spectrometer technologies (e.g., Orbitrap vs. Q-ToF).

Use when you have an aligned feature table from untargeted LC-MS with missing intensity values (NA or zero entries) for features that are present in some samples but fell below detection threshold in others.

Use when after retention-time and m/z-based peak alignment has been completed across a cohort of LC-MS samples, and you need to create a unified quantitative matrix for statistical testing, multivariate analysis, or annotation workflows.

Use when when processing a metabolomics feature table through multiple sequential transformations (e.g., imputation, normalization, batch correction, annotation) and you need to track which version of the table is being used at each step.

Use when after peak quality filtering has been applied to a composite map peak detection output using SNR (>2), goodness-of-fit (peakshape > 0.5), minimum peak height (default 1e5), and prominence (≥20% of peak_height) thresholds.

Use when you have an untargeted metabolomics feature table (m/z, retention time, p-value from statistical test) but lack comprehensive metabolite identifications or MS/MS annotations.

Use when you have millions of MS/MS spectra in mzML, mzXML, or MGF format and need to identify similar spectra for clustering, but exhaustive pairwise cosine-similarity computation would be prohibitively slow.

Use when converting intermediate JSONized experimental metadata (extracted from tagged tabular data) to a target repository format (e.

Use when when working with JSON project documents that must conform to a schema-defined structure, and you need to apply type-specific validation, sanitization, or transformation rules (e.g., URL whitespace detection, numeric range checking, mandatory field enforcement).

Use when you need to verify claims about algorithm performance, data processing correctness, or workflow outcomes in a scientific article or software repository. Use it when source documents contain figures, tables, or visualization badges (e.

Use when you have generated or received mzPeak files from a Rust, Python, R, or other implementation and need to verify they comply with the published HUPO-PSI specification before integration into a production workflow, data repository, or downstream analysis pipeline.

Use when you have generated a peak table or feature list from MZmine, XCMS, MS-DIAL, or Compound Discoverer and need to ingest it into LipidMatch for lipid identification.

Use when when building a file I/O abstraction layer that must support multiple serialization formats (e.g., uncompressed mzML, gzip-compressed mzML, indexed gzip mzML, or SQLite-backed mzML) and route each to its correct handler class without exposing format logic to downstream consumers.

Use when when you have authored a custom .csv lipid library and need to confirm it adheres to LipidMatch's documented schema before placing it in the designated library directory and running the library integration/loading step.

Use when you need to open an mzML file in pymzML and must automatically select the correct handler based on file extension (.mzML, .mzML.gz, .db) and—for gzip files—indexed vs. non-indexed compression status. Use it whenever FileInterface.

Use when you have mass spectrometry data stored in a non-standard format (SQLite database, custom indexed gzip files, or other database backends) and want to enable pymzML's Reader to access it with both random-access by spectrum ID and sequential iteration capabilities, avoiding the need to.

Use when a Shiny application or R package is confirmed to work on one OS (e.g., Windows only) but fails to initialize or run on others due to unresolved file path conventions, system library calls, or OS-specific package dependencies.

Use when after invoking the saveAnnotations function on a MetaboAnnotatoR annotations object to confirm that all four expected output file types (global results file, ranked results file, per-feature ranked spectra PDFs, and pseudo-MS/MS MGF file) have been written to the output directory without.

Use when after applying a filter function (filter_mispicked_ions(), filter_group(), filter_cv(), filter_insource_ions()) to an mpactr object, use this skill to inspect and document which features were retained versus removed.

Use when after applying a sequence of mpactr filters (filter_mispicked_ions, filter_group, filter_cv, filter_insource_ions) to an LC-MS/MS peak table in Progenesis or MS-DIAL format, call qc_summary() to obtain a structured report of per-ion filtering outcomes and use this skill to understand.

Use when when you have deposited or archived biosynfoni fingerprint vectors (such as from Zenodo 10.5281/zenodo.14822624) and need to ingest them into a Python workflow to compute distributional statistics, bit-frequency profiles, sparsity metrics, or pairwise similarity coefficients.

Use when after loading a specXplore session data object file from the hard drive and instantiating a dashboard session layer with it, validate that the architecture layer has initialized without errors and that the interactive dashboard is responsive to user input.

Use when when you have published LOTUS flat files (TSV or compressed TSV.GZ) containing structure-organism pairs and need to enumerate unique structures, group by organism prevalence, or validate record counts against gold-standard benchmarks.

Use when when you have sampled the feasible flux solution space of constraint-based metabolic models (via optGpSampler or equivalent uniform sampling) and need to normalize flux predictions across reactions and cell lines for concordance analysis with transcriptomics and metabolomics data.

Use when you have paired measurements (e.g., gene expression counts, protein abundance, or sampled flux distributions) from two cell lines or conditions and need to assign a directional sign to each reaction or gene for downstream concordance analysis.