Claude Skills by HolobiomicsLab

Use when use when the workflow requires background-distribution-significance-thresholding.

Use when when you have trained a predictive model (e.g., neural network or regression model) that outputs continuous scores (such as Spearman correlation coefficients) for individual features (e.

Use when you have a feature table from LC-MS data alongside blank (solvent-only) sample runs, and you want to remove features whose intensity in study samples is not substantially higher than their intensity in blanks.

Use when when you need to verify the current operational status of a software project across multiple dimensions (CI/CD, code quality, test coverage, containerization, archival) and those status indicators are exposed as badge endpoints in the project's README.

Use when you have raw MS/MS spectral data in standard mass spectrometry formats and need to prepare it for unsupervised substructure discovery via topic modeling.

Use when before peak detection on a composite or individual mass track when you need to filter out low-intensity noise and baseline drift without removing true signal.

Use when preparing labeled LC-MS peak data for neural network training and you need to decide whether class imbalance in your dataset should be preserved or corrected in batch construction. Use it particularly when your annotated peak dataset has unequal class distributions (e.

Use when after batch correction has been applied to metabolomics data using pooled SQC samples, and you need to retrieve the corrected ratios (compound / internal standard) for quality metrics calculation, internal standard recommendation, concentration estimation, or statistical modelling.

Use when after applying pycombat-based batch correction to multi-batch interpolated feature tables in LC-MS metabolomics workflows, when you need to verify that batch effects have been attenuated without loss of data dimensionality or sample information.

Use when when you have deposited a collection of JSON project documents in a platform and need to verify that all conform to the published schema before public release or after schema updates.

Use when when you have processed metabolomics LC-MS/MS data organized by batch and sample type (including pooled QC replicates), and you need to quantify whether batch-to-batch and matrix effects are acceptable for downstream reporting.

Use when after running pycombat batch correction on multi-batch metabolomics feature tables when you need to validate that batch correction has successfully attenuated inter-batch intensity variance without altering the structural integrity (sample and feature counts) of the corrected table.

Use when you have raw mzML files and a feature table (CSV) from LCMS data processed by tools like mzMine, and you need to create train/test/validation batches with specific matrix dimensions (120 × 2) and verified margin/peak signal separation before training or evaluating a neural network.

Use when apply batch normalization after dense hidden layers (but not the final embedding layer) in a deep neural network trained on MS/MS spectral data, particularly when the network processes high-dimensional binned spectra (9948-dimensional vectors) and you need to stabilize gradient flow across.

Use when when you have a trained molecular classifier (like BitterPredict) that accepts structured descriptor input, and you need to understand which chemical descriptor subgroups drive prediction outcomes.

Use when when you have raw mzML files and a feature table (CSV from mzMine or XCMS) with labeled peaks of unequal class sizes (e.g., fewer false positives than true positives) and plan to train a CNN classifier on the LCMS data.

Use when you have generated a peak table or feature list output file from an external peak-picking tool (MZmine, XCMS, MS-DIAL, or Compound Discoverer) in its native export format and need to ingest it into LipidMatch for lipid identification without manual column remapping or format conversion.

Use when you have a collection of N scripts (e.g., 19 gallery examples) that must run on multiple backends or configurations, and you need to produce a reproducible benchmark table showing per-script execution times and aggregate totals.

Use when you have a pre-trained DNN RT predictor (e.g., trained on METLIN SMRT with 80,038 experimental RTs) and need to adapt it to predict retention times in a new or external chromatographic method for which you have only 10–50 calibration molecules with known RTs.

Use when you need to transfer retention time predictions from one chromatographic method to another, but have access to only a small number (≥10) of molecules with known retention times in both methods.

Use when you have trained a new machine learning model for chemical formula or adduct assignment from MS/MS spectra and need to assess whether it offers genuine performance gains over established baselines. Use it specifically when you have access to ground-truth annotations (e.

Use when you have assembled genome sequences (contigs or scaffolds in FASTA format) and want to identify putative BGCs and their precursor peptides before constructing a RiPP structure database for spectral matching.

Use when you have computed raw strain correlation scores and IOKR scores for the same set of GCF–MF (gene cluster family–molecular feature) pairs, and you want to compare or combine them fairly without one score dominating due to scale differences.

Use when you have: (1) a trained IOKR model mapping from spectrum kernels to molecular fingerprints, (2) MS2 spectra from your sample, (3) a set of candidate BGCs with known or predicted structures (e.

Use when you have a normalized matrix of feature attribution scores (microbes × metabolites) derived from a trained neural network, and you want to partition both microbes and metabolites simultaneously into co-clusters that share similar interaction patterns.

Use when you have executed a binary classifier (such as BitterPredict.m) on a set of molecules with chemical structure descriptors and need to translate the raw predictions into a structured CSV output file that maps molecule identifiers to their predicted class labels (bitter or not-bitter).

Use when you have a trained binary molecular classifier (like BitterPredict) and want to understand which groups of chemical descriptors drive its predictions.

Use when you have a pre-trained TCN spectrum encoder, annotated MS/MS spectra paired with ground-truth molecular formulas, and you want to train only the formula ranking and rescoring components without retraining the spectrum feature extractor.

Use when you have a compressed file format (e.g., igzip) with a custom binary header structure that encodes metadata (index-to-offset mappings) in a fixed layout, and you need to parse or generate compliant files.

Use when you have a large gzip-compressed file (e.g., mzML.gz) with an embedded index structure in the gzip header comment field, and you need to retrieve specific blocks (e.g., mass spectra by scan number, chapters by ID) without decompressing the entire file.

Use when you have a filtered FT-ICR MS peak list (m/z values and assigned molecular formulas per sample) and wish to reconstruct biochemical transformation networks ab initio to characterize how microbial or environmental metabolic pathways differ across conditions.

Use when you have identified a package available in the Bioconda channel (indicated by a conda version badge or Bioconda recipe URL) and need to verify that installation succeeds and that the package's critical modules are importable in the target Python environment, especially before integrating.

Use when after constructing a SummarizedExperiment object from raw metabolomics data via buildExperiment, or after batch correction and ratio computation steps, inspect rowData, colData, and assays slots to verify that compound identities, sample annotations, and computed assay values (primary.

Use when after biomolecular class labels have been assigned to features in a TWIM-MS dataset and you have raw ion mobility arrival time measurements. Use it when you need to convert arrival times to standardized CCS values where calibration accuracy depends critically on the biomolecular class (e.

Use when you have raw or processed TWIM-MS data with arrival time and m/z values for multiple features, but lack prior structural identification (e.g., from spectral libraries or databases).

Use when you have assembled microbial genomes (nucleotide FASTA files) and want to identify biosynthetic potential and group related BGCs for downstream linking with metabolomic data;

Use when you have LC-MS/MS spectra (MGF, mzXML, mzML, or mzData format) from a bacterial or fungal strain and corresponding genomic sequence data (FASTA, antiSMASH .final.gbk, or BOA .annotated.

Use when you have output from a biotransformation rules module (candidate transformed structures linked to anchor molecules) and untargeted MS/MS spectral data, and you want to identify molecular families and annotate features with predicted structures by leveraging spectral similarity and network.

Use when you have untargeted metabolomics data with unknown metabolite structures and need to generate plausible candidate products by systematically applying known enzymatic or chemical transformation rules.

Use when you need to represent natural product molecules as fixed-length bit vectors for downstream machine learning (e.

Use when your peak table includes features flagged in blank control samples (e.g., solvent blanks, media blanks) at relative abundance above a project-specific threshold.

Use when apply this filter after feature detection and before downstream statistical analysis when your experimental design includes blank samples (e.

Use when when you have tandem mass spectra from ribosomally synthesized peptides (RiPPs) and suspect the presence of unknown or non-standard post-translational modifications that would be missed by standard database search modes constrained to known modification classes.

Use when when you need to verify that a GitHub Actions workflow (such as dev_build_release.

Use when you need to validate that a repository's automated build and publish pipeline is functioning correctly on a release or target branch, particularly when assessing the reliability of release artifacts or the health of a CI/CD workflow.

Use when when you need to validate that a development build release workflow (such as dev_build_release.yml for a mass spectrometry data processing project) executes without failure and generates artifacts for downstream testing or distribution.

Use when when you need to capture a snapshot of a research software project's health metrics from multiple CI/CD and repository services (Travis CI, Landscape.

Use when when you have implemented a binary file format encoder (such as igzip header construction) and need to verify that the binary output is correct before deploying it to read or write real files.

Use when when you need to read proprietary or binary data formats (e.g., Thermo Fisher .raw files) from R but the native implementation is in .NET/C#, and direct language bindings are unavailable or impractical. Use this when the target assembly requires Windows/.

Use when when integrating an R package that wraps a compiled .NET assembly (such as rawrr), you need to verify that the internal dispatch mechanism between the R layer and the C# layer is operational before attempting to read actual raw data files.