Claude Skills by HolobiomicsLab

Use when immediately after chromatographic peak detection (findChromPeaks) when you have detected peaks across multiple samples and need to identify which peaks represent the same feature across the sample cohort.

Use when you have preprocessed LC-MS intensity data (e.

Use when you have generated feature tables from LC-MS data using different parameter combinations (e.g., varying Centwave, FeatureFinderMetabo, or ADAP peak picking settings) and need to objectively compare their outputs to select the -performing configuration for your dataset.

Use when immediately after peak detection and feature table generation from LC-MS data, when you need to rank or filter features by confidence before annotation or statistical analysis.

Use when when you have a set of candidate LC gradients (parameter combinations) that you wish to evaluate with a Gaussian process model, or when you need to convert raw gradient specifications into a standardized numerical format for Bayesian optimization acquisition function computation.

Use when at the start of an untargeted LC-MS annotation pipeline when you have a KEGG database with exact masses and need to prepare a mass-matching reference.

Use when after executing a Nextflow-based LC-HRMS metabolomics workflow with Docker or Singularity containerization on .mzML LC-MS data, before proceeding to downstream statistical or visualization analyses.

Use when you have LC-MS feature tables with m/z and retention time coordinates paired with raw .mzXML or .mzML data files, and you need to systematically assess which features correspond to genuine chromatographic peaks versus noise or artifacts.

Use when after a CNN-Transformer peak detection model has been run on LC-MS ROI images and has output predicted peak locations with confidence scores.

Use when you have raw profile (not centroided) LC-MS data in .mzML format and need to prepare it for automated peak detection using a trained object detection network.

Use when after performing peak detection on centroided .mzML LC-MS data with screening_mode=FALSE in TARDIS.

Use when you have centroided .mzML LC–MS data with multiple sample runs and a preliminary compound target table with theoretical or measured retention times, but you suspect the expected RT values may not align with actual retention windows in your dataset.

Use when when you have extracted LC-MS ROI windows (m/z × retention time snippets) from raw mzML data and need to distinguish genuine metabolite peaks from noise or artifacts.

Use when when you have a curated list of chemical compounds (real or virtual), a defined fragmentation strategy (e.g., Top-N, exclusion lists), and need to simulate how that strategy will acquire MS1 and MS2 scans over a defined retention-time window.

Use when you have raw LC-MS/MS spectral data in .mgf format (or vendor-specific raw data that can be converted to .mgf via MZmine or similar tools) and need to prepare it for interactive exploration using the specXplore dashboard.

Use when you have loaded mzML.gz or HDF5-formatted raw LC-IMS-MS/MS data and need to identify discrete peaks before feature alignment. Use it if your goal is to reduce noise, increase signal-to-noise ratio, and prepare multi-dimensional data for cross-sample feature matching and CCS calibration.

Use when you have raw LC-MS data from a vendor instrument or in netCDF format and need to ingest it into SLAW or similar untargeted LC-MS workflows. Use this skill when raw data must be converted to mzML, validated for centroiding and polarity uniformity, and prepared for peak-picking dispatch.

Use when when you have raw LC/MS data in mzML format and need to execute the LAGF non-targeted screening pipeline. Use this skill as the first step before applying the LAGF algorithm workflow to extract and annotate features from mass spectrometry data.

Use when you have an LCMS feature table annotated with MS2 spectral data and need to distinguish in-source fragments (mass loss patterns, same retention time, MS2 spectral relationships) from true metabolite features.

Use when you have raw LC-MS data (mzML or equivalent format) from a metabolomics experiment and need to extract a reproducible, quantified feature table with intensity measurements before conducting metabolite identification or statistical analysis.

Use when when you have raw LC-MS chromatographic data (mzML or vendor format) and need to identify and characterize all detectable peaks across the full retention time range for untargeted metabolomics or discovery workflows.

Use when you have an untargeted LC/MS feature table (m/z, retention time, intensity columns) and need to move beyond single-hit matching to probabilistic annotation.

Use when after ISFrag has completed identification of in-source fragment features (Part 4 output), when you need to serialize and inspect the hierarchical fragmentation relationships among identified ISF features, or when preparing data for visualization or external analysis of fragment lineage and.

Use when you have centroided, single-polarity mzML files from DDA LC-MS experiments and need to generate a quantitative feature table with aligned m/z and retention time coordinates, isotopic annotations, and MS2 spectra for downstream statistical or annotation analysis.

Use when you have raw nontargeted LCMS feature tables from one or more analytical methods in tabular format (with m/z, RT, and intensity columns) that need to be aligned or clustered, or when integrating multiple feature tables into a shared BMXP processing pipeline that requires standardized.

Use when you have raw LC-HRMS metabolomics data in .mzML or .

Use when you have an XCMS-processed feature set (XCMSet object) from replicate LC/MS runs comparing labeled (e.g., 13C-glucose) and unlabeled (e.g., 12C-glucose) conditions, and need to identify which features show isotope incorporation (fold-change ≥1.5, p-value <0.

Use when you have one or more raw mzXML or mzML LCMS data files (from DDA, DIA, or fullscan acquisition) and need to extract quantitative metabolite features for multi-sample comparative analysis.

Use when after XCMS peak picking, alignment, and grouping when you have identified putative incorporations (via PuInc_seeker) or base-peak isotopologue candidates and need to exclude low-intensity peaks that are likely noise or instrument artifacts.

Use when you have raw LC/MS data in mzML or mzXML format and need to initiate untargeted metabolomics analysis.

Use when after completing Part 4 (Identification of ISF Features) in the ISFrag workflow, when you have an analysis results object containing identified ISF features and need to generate a shareable, tabular export that documents feature annotations, hierarchical parent–fragment relationships, and.

Use when when XCMS-aligned LC-MS data shows coefficient of variation (CV) above expected thresholds for known features, or when analyzing long-duration experiments (>1 week) or large cohorts (>100 samples) where global warping functions are known to fail due to compound-specific RT drift structures.

Use when you have raw LC-MS spectral peak data (in the format provided by DOI 10.25345/C5FD2F) and need to build a classifier that can distinguish valid peaks from false positives or background noise without manual feature engineering.

Use when after loading centroided .mzML LC–MS runs and before executing full peak detection and integration.

Use when starting from raw LC-MS spectral files (mzML or mzXML format) in a global metabolomics study and you need to produce a complete, validated feature table with m/z, retention time, and intensity values across all samples.

Use when after configuring LDA hyperparameters (alpha, beta, number of topics, iteration budget) and loading a preprocessed bag-of-fragments corpus with neutral losses extracted and noise filtered, initiate LDA training and apply convergence monitoring to determine when topic-fragment probability.

Use when you have a preprocessed bag-of-fragments corpus derived from tandem mass spectrometry spectra and need to discover recurring fragmentation motifs without prior compound identification.

Use when after running gensim LDA on a corpus of MS2 fragmentation features, when you need to store the LDA results (topics, document-topic assignments, term-topic distributions) in a PostgreSQL database so they can be queried and visualized by a Django web application or other downstream consumers.

Use when you have a metabolomics dataset (LC/MS or GC/MS) with missing values and need to determine which are below the limit of detection (LOD) or limit of quantification (LOQ). Left-censored classification is necessary when the missingness is informative—i.

Use when when you have a complete metabolomics abundance table (e.g., targeted LC/MS or untargeted GC/MS counts) and need to generate synthetic left-censored missingness for evaluating imputation algorithm performance.

Use when after khipu has grouped LC-MS features into empirical compounds with inferred molecular formulas and adduct assignments.

Use when when you need to reconstruct or validate the control-flow architecture of a spectral search system that must handle both exact-match library lookups and analogue discovery in a single pass, particularly when the system uses pre-computed embeddings for efficiency and machine learning for.

Use when you have raw .msp spectral library files (e.g., from MassBank or custom sources) and need to convert them into a structured CSV library format for use in metabolite annotation.

Use when you have custom lipid entries (e.g., synthetic lipids, rare natural variants, or isotopically labeled standards) not covered by LipidMatch's default in-silico library, and you want to include them as matching candidates in your UHPLC-HRMS/MS fragment m/z matching workflow without modifying.

Use when you have experimental MS/MS spectra (from mzML, mzXML, or raw instrument formats) and wish to match them against a reference spectral library provided in MSP or CSV format. The skill is required as the first step before similarity scoring and candidate ranking.

Use when you have extracted retention times at peak maxima (rtFittedAPEX) from extracted-ion chromatograms (XICs) of known internal RT calibrants (e.

Use when when you have 1H-NMR metabolite measurements from Nightingale Health assayed on a new cohort and wish to compute risk scores (e.g., all-cause mortality, cardiovascular event, type 2 diabetes) using published metabolic biomarker weights from a reference study.

Use when after running a scoring algorithm (e.g., MetcalfScoring) on paired genomic and metabolomic datasets.

Use when you have loaded GCFs (from AntiSMASH via BigScape clustering), GNPS spectra, and molecular families (from GNPS molecular networking), and need to compute pairwise scoring between genomic and metabolomic entities to identify putative gene cluster–metabolite associations.

Use when you have vendor-specific raw mass spectrometry data (ThermoFisher .raw, Agilent .