AI & ML

holobiomicslab Skill

Use when you have collections of chemical formulae (e.g., from SIRIUS decomposition or subformula labeling) derived from MS/MS spectra and need to feed them into a transformer encoder.

holobiomicslab Skill

Use when you have raw tandem MS metabolomics data (in mzML or MGF format) and wish to compare the clustering performance of two or more MS clustering tools on the same dataset.

holobiomicslab Skill

Use when after computing pairwise cosine similarities between all spectra across two LC-MS/MS datasets when you need a single scalar summary of dataset-level resemblance rather than individual spectrum matches.

holobiomicslab Skill

Use when after generating a structured JSON result file from a prior computational step (e.g., gensim LDA model output in myexp.ldaresult.json format) and needing to store it in PostgreSQL for web application access, visualization, or further analysis.

holobiomicslab Skill

Use when when you have parsed EI-MS spectrum data (m/z and intensity values) and need to feed it into a pre-trained MWFormer transformer model for direct molecular weight prediction.

holobiomicslab Skill

Use when you are preparing to run LipoCLEAN on MS-DIAL output and need to create or update a configuration file, or you have switched between MS-DIAL 4 and MS-DIAL 5 data and need to verify that your options file and trained model are compatible with the current version's column naming and scaling.

holobiomicslab Skill

Use when when beginning an untargeted LC-MS analysis and either (1) the dataset characteristics (sample complexity, instrument platform, or polarity) differ from previously optimized cohorts, (2) multiple peak-picking algorithms (Centwave, FeatureFinderMetabo, ADAP) are available and their relative.

holobiomicslab Skill

Use when after embedding MS/MS spectra into 32-dimensional GLEAMS vectors, when you need to group spectra by the same peptide origin.

holobiomicslab Skill

Use when when you have a collection of mass spectra with annotated chemical structures (SMILES/InChI) and need to generate structural similarity labels to train or validate a model that predicts molecular similarity from spectral pairs.

holobiomicslab Skill

Use when when you have an unknown metabolite's predicted structural similarity scores (from a deep learning model such as DeepMASS) against all known metabolites in a reference database, and need to identify which known metabolites are most likely structurally related to the unknown to guide.

holobiomicslab Skill

Use when when you have raw predictions from a trained fragment generation or intensity prediction neural network model and need to convert those predictions into a standard spectrum file format (m/z–intensity pairs) for comparison against experimental spectra or for structural elucidation workflows.

holobiomicslab Skill

Use when you have collected liquid chromatography (LC) spectra and retention time labels for your in-house molecular database, and you want to leverage a pretrained GNN-RT model rather than train from scratch.

holobiomicslab Skill

Use when when you have tokenized mass spectra (peak-mass and peak-intensity pairs from experimental or in-silico libraries such as NIST 2017 or MassBank) and need to perform rapid similarity searches or spectrum matching at scale.

holobiomicslab Skill

Use when you have a pretrained PyTorch model with released weights (e.g., JESTR on NPLIB1), a prepared dataset with input features (spectra m/z–intensity arrays, molecular graphs), a GPU environment with CUDA 11.

holobiomicslab Skill

Use when you have pre-computed Word2vec embeddings of mass spectra and need to retrieve the k most similar spectra from a library of hundreds of thousands to millions of candidates.

holobiomicslab Skill

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.

holobiomicslab Skill

Use when you have a trained multitask machine learning model for structure prediction, test set molecules with paired ¹H and ¹³C NMR spectra, and need to understand the marginal contribution of each NMR modality or justify multimodal input design.

holobiomicslab Skill

Use when processing raw MS/MS spectra (in MGF, mzML, mzXML, JSON, or MSP format) prior to MS2Query library matching or MS2Deepscore embedding calculation.

holobiomicslab Skill

Use when when you have a trained BitterPredict classifier, a dataset of molecules with computed descriptors and known bitter/not-bitter labels, and want to understand which descriptor categories (e.g., molecular weight, lipophilicity, topological, pharmacophoric) drive prediction decisions.

holobiomicslab Skill

Use when you have intracellular metabolomics concentration measurements across multiple cell lines or conditions, a stoichiometric metabolic network model with reaction-metabolite associations, and you need to predict how differences in substrate availability (independent of enzyme expression).

holobiomicslab Skill

Use when you have two LC-MS feature tables (each containing m/z, retention time, and intensity columns) from the same or related biological samples and need to identify which features in dataset A correspond to which features in dataset B.

holobiomicslab Skill

Use when you have a GC-MS dataset with Match.Factor scores for each detected compound (output from Agilent Unknowns Analysis or equivalent), and you want to reduce the number of query chemicals passed to computationally intensive cheminformatics functions (categorate, mzExacto, or exactoThese).

holobiomicslab Skill

Use when you have loaded two or more nontargeted LCMS feature tables from the same analytical method that contain m/z, retention time, and intensity values, and these datasets exhibit differences in metadata scale, distribution, or format that could confound cross-dataset feature matching or.

holobiomicslab Skill

Use when your metabolomics dataset contains metabolite identifiers in multiple formats (e.

holobiomicslab Skill

Use when when preparing to send peak data (1H and 13C NMR measurements) to a machine learning classification endpoint and you need to verify the current model's input/output names and schema, especially before implementing or updating code that constructs JSON payloads for the /api/smart3/search.

holobiomicslab Skill

Use when you have 1D ¹H and/or ¹³C NMR spectra (as preprocessed numerical arrays or peak lists) from an unknown organic molecule with ≤19 heavy atoms, and you need to recover its molecular formula and connectivity graph.

holobiomicslab Skill

Use when when you have a raw or partially processed MS/MS spectra collection (e.g., GNPS-sourced Orbitrap or Q-TOF spectra in MGF format) and need to (1) restrict to a specific instrument type (e.

holobiomicslab Skill

Use when after clustering and filtering KEGG candidates for LC-MS features, when you have a ranked set of candidate metabolites per feature and access to a metabolite interaction network (e.g., from FELLA).

holobiomicslab Skill

Use when when you have raw MS2 spectra (m/z and intensity pairs) and a curated reference peak list from a large training dataset (e.

holobiomicslab Skill

Use when when you have implemented or adopted a new clustering or analysis tool and need to validate that it meets stated runtime claims on a representative production-scale dataset. Particularly important when the tool uses hardware acceleration (GPU) and the claimed speedup is a core contribution;

holobiomicslab Skill

Use when you have pre-trained MLP and GNN spectral prediction models evaluated on the same ESI/LC-MS test dataset, and you seek to improve average rank performance beyond either baseline model alone.

holobiomicslab Skill

Use when you have a TCN-predicted training set of MS/MS spectra with formula annotations and need to prepare it for Siamese rescore model training.

holobiomicslab Skill

Use when when you have paired microbiome (16S rRNA or functional) and metabolome (LC-MS/MS or similar) data and must decide between compositional transformations (CLR, RA, or others) before training a predictive model.

holobiomicslab Skill

Use when you have molecular structures (SMILES strings or molecular graphs) that need to be input to a transformer model for property prediction (e.g., Collision Cross Section), and the model requires tokenized or embedded representations rather than raw chemical notation.

holobiomicslab Skill

Use when after applying polynomial m/z domain recalibration using a reference peak list (e.g., SRFA.ref) to a Bruker FT-ICR dataset. Use this skill to verify that calibration has converged and that mass error statistics support reliable downstream annotation.

holobiomicslab Skill

Use when you have raw mass spectrometry spectra (peak lists or intensity arrays) that must be fed into a pre-trained deep learning model for substance classification (e.g., PS²MS for NPS detection).