AI & ML

holobiomicslab Skill

Use when you have trained two or more graph neural network models on the same CCS dataset split (using identical hyperparameters, loss functions, and optimization settings) and need to rigorously compare their held-out test performance to determine which architecture balances prediction accuracy.

holobiomicslab Skill

Use when you need to determine the complete set of validated instrument/vendor and acquisition mode combinations for a mass spectrometry analysis tool, when assessing whether your specific instrument platform (vendor, model, acquisition method) is supported before committing to a workflow, or when.

holobiomicslab Skill

Use when you have paired microbiome and metabolomic (or similar compositional) data with a new regression model and want to rigorously demonstrate its predictive advantage over alternatives (Elastic Net, Random Forest, CCA).

holobiomicslab Skill

Use when when you have a neural network layer definition (parameters, weight initialization, embedding dimension) from a trained or pretrained model checkpoint and need to generate embeddings or activations for a new batch of chemical formulas or spectrum fragments.

holobiomicslab Skill

Use when preparing tandem MS/MS datasets for cross-dataset similarity analysis or spectral matching, particularly when datasets originate from different instruments, acquisition dates, or sample preparation protocols that may introduce systematic variations in peak intensities.

holobiomicslab Skill

Use when when you have raw MS/MS spectral data in the form of intensity arrays indexed by m/z values and need to feed them into the Spec2Mol encoder neural network. Apply this skill before encoder inference to ensure spectral inputs conform to the encoder's expected dimensionality and value ranges.

holobiomicslab Skill

Use when you have acquired or generated multi-modal spectroscopic data (integrated NMR, HSQC, COSY, IR spectra) in the model's expected input format, a pre-trained MultiModalSpectralTransformer checkpoint is available, and you need to predict molecular structures from these spectra without.

holobiomicslab Skill

Use when when you have a raw GNPS or other spectral library dataset with inconsistent or incomplete instrument annotations, and you need to verify or reproduce reported dataset split counts (e.g., training/test compound ratios). Apply this skill when an instrument allowlist fix (e.

holobiomicslab Skill

Use when use this skill after training or fine-tuning a chemical formula transformer model on annotated tandem MS/MS spectra, when you need to measure whether the model's ranked formula candidates match ground truth.

holobiomicslab Skill

Use when you have paired microbiome (16S rRNA/metagenomic) and metabolome (LC-MS/MS or similar) count data and need to evaluate how well a predictive model (e.g., neural network, Elastic Net) generalizes across samples.

holobiomicslab Skill

Use when you have a formula-assigned FT-ICR MS dataset (molecular formulas already assigned to individual mass features) and seek to understand the chemodiversity landscape and transformation relationships within DOM.

holobiomicslab Skill

Use when when clustering or encoding large MS/MS spectra datasets (>1 million spectra) where CPU-only runtime exceeds practical thresholds (hours to days).

holobiomicslab Skill

Use when when you have positive- or negative-mode ion mobility spectrometry data with tunemix reference standards (known m/z, drift times, and CCS values) and need to verify that the calibration model accurately captures the relationship between drift time, reference m/z, and collision cross.

holobiomicslab Skill

Use when your TWIM-MS dataset contains ions with multiple charge states (e.g., +1, +2, +3 for the same molecular species) and you need CCS values that correctly account for the relationship between drift time, m/z, and charge state.

holobiomicslab Skill

Use when when you receive a mass spectrometry imaging dataset in unknown or mixed vendor formats and need to apply format-specific preprocessing before generating ion images.

holobiomicslab Skill

Use when you have implemented or modified a tandem mass spectrometry formula inference model and need to measure whether a specific architectural change (e.

holobiomicslab Skill

Use when you have a list of fragment peak m/z values and intensities from tandem MS/MS data and need to assign chemical subformulae to each peak for downstream formula ranking or structure inference.

holobiomicslab Skill

Use when you have a trained GNN model (stored as .h5 weights) and molecular graph representations (SMILES strings and/or 3D coordinates), and you need to compute predicted CCS values or perform feature importance analysis via ablation or gradient-based saliency mapping.

holobiomicslab Skill

Use when when you have quantified intracellular metabolite abundances (LC-MS normalized values) for multiple cell lines or samples, a metabolic network model with reaction stoichiometry, and you need to predict how substrate availability translates into metabolic flux differences.

holobiomicslab Skill

Use when when adopting a mass spectrometry data processing tool (e.g., LipidMatch) and needing to verify whether your specific instrument platform (vendor + model) and acquisition mode combination (targeted, ddMS2-topN, AIF, direct infusion, imaging) have been formally validated.

holobiomicslab Skill

Use when when you have predictions from multiple independently trained models (e.g., ROASMI_1–ROASMI_5) for the same set of compounds in a reversed-phase liquid chromatography system at eluent pH ~2.7, and you need to estimate prediction reliability without ground-truth labels.

holobiomicslab Skill

Use when you have access to pre-trained MSGO model weights (PFAS or lipid variants) and a set of 300+ real mass spectra (LC–QTOF or similar), and need to verify whether the model can generate correct molecular structures for unknown chemicals.

holobiomicslab Skill

Use when you have a pre-trained neural network model (e.g., MSBERT weights in .

holobiomicslab Skill

Use when you have paired tandem MS spectra and either (1) molecular fingerprints or structures as labels for supervised fingerprint prediction, or (2) both spectra and unpaired structure/SMILES libraries and want to train embeddings for database-free structure lookup.

holobiomicslab Skill

Use when you have a pretrained deep learning model, a reserved test set with ground-truth annotations, and need to evaluate prediction quality or generate embeddings for downstream analysis. Typical triggers: benchmarking a new model against classical baselines (e.

holobiomicslab Skill

Use when when you need to detect and classify peaks in LC-MS regions of interest (ROIs) and simultaneously localize their boundaries for area integration.

holobiomicslab Skill

Use when you have paired tandem MS/MS spectra with known molecular fingerprints, chemical formulae, or SMILES annotations, and you want to learn a generalizable model that can predict molecular properties or annotate unknown spectra by ranking candidates.

holobiomicslab Skill

Use when when you have preprocessed MS/MS spectral data (filtered, noise-reduced, with neutral losses extracted) and need to discover recurring fragmentation patterns across a spectral dataset.

holobiomicslab Skill

Use when after training a neural network or regression model to predict metabolomic profiles from microbiome data.

holobiomicslab Skill

Use when after applying mass calibration functions (LedFord, linear, or quadratic equations) to an FT-ICR transient or magnitude-mode dataset, before running SearchMolecularFormulas. Specifically, validate recalibration when: (1) comparing recalibrated spectra against reference calibration files (e.

holobiomicslab Skill

Use when when you have applied multiple scoring functions (e.g., strain correlation and IOKR) to rank genomic-metabolomic (GCF-MF or BGC-spectrum) links and need to verify that: (1) standardisation produces zero mean and unit variance across all links;

holobiomicslab Skill

Use when you have a preprocessed corpus of mass spectrometry spectra converted to bag-of-fragments format (with neutral losses extracted and noise filtered), and you seek to discover recurring fragmentation patterns or substructures that characterize multiple spectra without prior knowledge of the.

holobiomicslab Skill

Use when you have two LC-MS feature tables (each with m/z, retention time, and intensity columns) from independent untargeted metabolomic experiments or replicates and need to establish one-to-one feature correspondence across them to compare abundances, detect shared metabolites, or merge datasets.

holobiomicslab Skill

Use when when building an end-to-end deep learning model that must predict multiple correlated peptide properties (charge, isotope count, retention time) simultaneously from mass spectrometry data, and you need a principled way to merge learned representations from separate task-specific branches.

holobiomicslab Skill

Use when you have tandem mass spectra (MS/MS) with unknown precursor formulas and need to rank chemical formula candidates conditioned on observed fragment m/z values and precursor mass. Use this skill when fragmentation tree computation (e.

holobiomicslab Skill

Use when you have two or more nontargeted LCMS feature tables from the same analytical method (same column, ionization mode, and acquisition parameters) and need to identify which features in one dataset correspond to features in another.