Claude Skills by HolobiomicsLab

Use when you have a trained predictor (like BitterPredict) that accepts structured descriptors and want to understand feature importance without retraining.

Use when you have a log2-normalized, zero-mean and unit-variance standardized intensity matrix of metabolite features (rows=metabolites, columns=samples) and need to compute a single activity score per pathway that reflects the coordinated expression behavior of all metabolites assigned to that.

Use when when you have computed pairwise distances between MS2 fingerprint vectors from multiple metabolomics samples and need to visualize sample similarity relationships in low dimensions.

Use when you have a Thermo Fisher Scientific .raw file containing PRM data and need to verify that acquisition of a specific precursor ion (e.g., LGGNEQVTR++ at m/z 487.2567) is happening at regular intervals consistent with your instrument method design.

Use when you have preprocessed tandem mass spectrometry spectra converted into a bag-of-fragments representation (with fragments and neutral losses extracted and noise filtered) and your goal is to discover recurring fragmentation patterns or substructures across a large spectral dataset without.

Use when after training or loading a NeatMS neural network model, apply this skill when you have a labelled validation dataset and need to determine the optimal probability threshold that maximizes classification performance (true positives minus false positives).

Use when you have a trained NeatMS neural network model (.h5 format) and need to select an operating threshold for peak classification on your LCMS dataset.

Use when you have a ProForma 2.0 peptidoform string (e.

Use when you have peptide sequences with chemical modifications encoded in ProForma notation (e.g., '[Phospho]-PEPTIDE[Carbamidomethyl]-C') and need to validate their syntax, extract modification positions, or prepare them for downstream physico-chemical property calculations (mass, charge, pI).

Use when when a user uploads a JSON project document to the Pairing Omics Data Platform, before accepting it into the repository or indexing it for search. Use this skill to catch missing required fields, incorrect field types, and constraint violations early in the submission workflow.

Use when after initial peak detection on composite mass tracks via local maxima and smoothing, when you have unfiltered peak lists (JSON or structured format) containing prominence values and need to reduce the number of detected features while maintaining signal quality.

Use when after executing annotateRC to match six or more lipidomics/metabolomics features against ion fragment databases (e.

Use when investigating how a specific pull request (e.g., PR #72 introducing MS2Query''s two-branch workflow split) modified the codebase architecture, control flow, or data routing.

Use when when you have mzML spectrum XML already parsed (either from a file, a database query, or an in-memory representation) and need to construct Spectrum or Chromatogram objects that expose methods like accessing MS level, retention time, m/z and intensity arrays, and other metadata.

Use when after installing a package in development mode (e.g., via `pip install -e .[dev]`) to verify the package functions as intended, or before submitting pull requests to confirm no regressions were introduced by code changes.

Use when when extending MSMetaEnhancer with a new local chemical transformation (e.g., SMILES to InChI) that should execute non-blockingly within an asynchronous annotation workflow.

Use when when contributing code changes to a Python project (fork, feature branch, or pull request) that uses a setup.py-based test suite, before pushing changes to the remote repository or merging into the main branch.

Use when when you need to enable bracket notation (e.g., handler[5]) for random access to blocks within a large compressed or remote data source, or when you want to support both random access and sequential iteration over data without materializing all records upfront.

Use when when building an extensible converter framework where new converter implementations (e.g., WebConverters or ComputeConverters for external chemical services) should be automatically discovered and registered without modifying a central registry.

Use when when setting up a multi-stage bioinformatics workflow (e.g., ENPKG) that calls external tools (MZmine, Sirius, SPARQL engines) and depends on specific Python libraries; when reproducibility across team members or cloud infrastructure is required;

Use when when you have installed a Python package (e.g., via pip or conda) and need to confirm that the installation succeeded and all expected submodules can be imported.

Use when when you need to execute a multi-backend visualization library (e.g., pyOpenMS-Viz with matplotlib, Bokeh, and Plotly) and must measure or validate execution times, memory usage, and output consistency across runs or team members.

Use when you have cloned a scientific repository containing Python code (scripts, Jupyter notebooks, or module imports) and need to execute it locally or on new hardware.

Use when when you have completed spectral data preprocessing in Jupyter notebooks and generated a specXplore session data object saved to disk, use this skill to restore that object into a live specXplore dashboard session instance for interactive LC-MS/MS spectral data exploration, without.

Use when you need to enable optional modules in Pyteomics that depend on external libraries not bundled with the core package—such as h5py and hdf5plugin for mzMLb format access, sqlalchemy for Unimod database queries, or psims for ProForma parsing.

Use when when setting up a new computational environment for tandem MS/MS spectrum clustering or other proteomics analysis, and you need to install a tool (like falcon) that depends on specific versions of auxiliary packages (like spectrum-utils==0.3.5).

Use when when you need to validate that a Python package (or update to it) is accessible to end users through official distribution channels, or when you are preparing a release and need to confirm that installation from PyPI and/or Bioconda does not introduce import failures or missing.

Use when when you need to verify a Python package installs successfully from a cloned or local repository, validate that all tests pass after installation, or prepare a development environment for contributing to the package.

Use when you have a mature Python package (e.g., Mummichog 2.x) that needs to be relocated to a new GitHub organization (e.

Use when you have a Python webservice codebase (e.g., a Flask, Django, or FastAPI application) and need to document its HTTP API surface (endpoints, methods, parameters, schemas, authentication) for integration, testing, or specification generation.

Use when you need to serialize heterogeneous data (e.g., index integers and file offsets of different byte widths) into a compact binary format, or parse a binary file format specification that includes fixed-width field encodings.

Use when you have raw mass spectrometry spectral data in common formats (MGF, MSP, mzML, mzXML, JSON) that requires standardized metadata cleaning, validation, and peak filtering before comparative analysis. Use this skill when you need to encode data quality constraints (e.

Use when after defining a transformer encoder architecture with multi-head self-attention and positional encoding, and before training or inference on mass spectrometry data.

Use when after batch correction of a metabolomics dataset using pooled study quality control (SQC) samples, when you have multiple candidate internal standards and need to systematically evaluate which one produces the most stable compound quantification (lowest QC variability) for each compound.

Use when after applying one or more mpactr filters (filter_mispicked_ions, filter_group, filter_cv, filter_insource_ions) to an mpactr object, call qc_summary() to obtain a structured report of which ions passed or failed each filter.

Use when you have acquired a versioned QC workflow definition file (YAML or JSON) from a metabolomics QC system release (e.g., v1.0.

Use when you have Sciex Multiquant txt exports containing signal intensities from QCpool samples injected at regular intervals (e.g., every 10–20 samples) during one or more analytical sequences in a metabolomics or lipidomics study.

Use when you have Sciex Multiquant (≥v3.0.3) TXT export files containing metabolomics or lipidomics analytical sequences that include pooled QC samples, and you need to verify that QCpool samples were injected at the designed regular intervals and extract their positional metadata for quality.

Use when you need to verify that a QIIME 2 artifact (e.g., a Chemical Feature Tree from q2-qemistree, a FeatureTable[Frequency], or a Phylogeny[Rooted] object) has been correctly produced, before using it as input to downstream analyses.

Use when when you have Sciex Multiquant TXT export files containing injection sequences from metabolomics or lipidomics studies where pooled QC samples were deliberately injected at regular intervals to monitor analytical quality.

Use when after composite-map peak detection has produced an unfiltered peak list with SNR, peakshape (goodness_fitting), peak_height, and prominence values.

Use when you have measured IM-MS lipidomics data spiked with U13C labeled internal standards and need to assess whether CCS bias remains within acceptable limits for each lipid class-adduct combination.

Use when you have a SummarizedExperiment object containing pooled quality control samples with measured compound and internal standard peak areas.

Use when you have a feature table with intensity values across study (unknown) and blank control samples, and you need to remove features that may represent instrument artifacts, contamination, or noise rather than true biological signal.

Use when you have sequential QCpool (pooled quality control) samples analyzed with Sciex Multiquant (≥v3.0.

Use when you have raw MS intensity tables showing systematic drift during a measurement sequence (e.g., declining or variable ion counts across a run), particularly in targeted quantification workflows where external calibration standards are available.

Use when when converting raw metabolomics data (tab-delimited text files, Sciex OS exports) into a structured object for batch processing, or when you need to organize compound-level measurements (assays), sample metadata (colData), and feature annotations (rowData) in a single container that.

Use when after installing an R package from a non-CRAN repository (such as r-universe) to confirm the package build is sound, dependencies resolve correctly, and no warnings or errors are introduced. Use it as a gate before relying on the package for downstream analysis or distribution.

Use when a Shiny application or R-based tool is known to run on only one operating system (e.g., Windows-only), and you need to identify the root causes preventing execution on Linux or macOS before undertaking cross-platform porting.

Use when when you need to verify or retrieve package-internal metadata about compiled .NET assembly location and version before processing raw mass spectrometry files, or when testing the R↔C# dispatch mechanism in isolation without loading actual Orbitrap .raw data files.