3. Data, Science & AI

Provides molecular structure analysis, SMILES/SDF parsing, descriptors, and substructure search for cheminformatics workflows.

Coordinates multiple AI agents for parallel execution, communication, task distribution, and result aggregation in complex workflows.

Enables molecular machine learning with diverse featurizers, pre-built datasets, and MoleculeNet benchmarks for property prediction tasks such as ADMET and toxicity.

Provides access to and analysis of comprehensive drug data from DrugBank, including properties, interactions, targets, and pharmacology.

Provides a comprehensive toolkit for molecular biology data processing, including sequence analysis, file format parsing, and programmatic access to biological databases.

Builds and deploys bioinformatics pipelines using Latch SDK with Nextflow/Snakemake integration and serverless deployment capabilities.

Explains the protocol for AI agents to discover and invoke skills, enabling seamless capability integration in agent-based systems.

Performs AI-powered web searches with real-time information and source citations using Perplexity models via OpenRouter API.

Offers AI-ready drug discovery datasets including ADME, toxicity, and DTI benchmarks for therapeutic machine learning models.

Accesses and retrieves gene expression data from NCBI GEO, including microarray and RNA-seq datasets in SOFT/Matrix formats for transcriptomics analysis.

Enables ML analysis of genomic regions using BED files, supporting region embeddings and scATAC-seq data processing.

Queries Reactome database for pathway analysis, gene-pathway mapping, and molecular interaction studies in systems biology.

Queries NCBI ClinVar database to retrieve variant clinical significance, supporting genomic medicine research through gene/position searches and VCF annotation.

Automates exploratory data analysis for scientific files across 200+ formats, generating markdown reports with quality metrics and domain-specific recommendations.

Provides deep generative models for single-cell omics data, enabling probabilistic batch correction, transfer learning, and multi-modal integration (scVI, TOTALVI, MultiVI).

Enables distributed computing to scale pandas and NumPy operations for larger-than-memory datasets and cluster processing.

Direct REST API access to KEGG database for academic pathway analysis, gene mapping, and metabolic pathway exploration.

AI-driven tool for evaluating task complexity to determine optimal workflow, phases, and validation requirements.

Builds hierarchical Bayesian models using PyMC with MCMC (NUTS), variational inference, and model comparison techniques for probabilistic programming.

Comprehensive toolkit for microbiome analysis including sequence alignment, phylogenetic trees, diversity metrics, and statistical tests.

Efficient cloud storage for large scientific datasets using chunked N-D arrays with S3/GCS integration and compatibility with NumPy/Dask/Xarray.

Queries openFDA API for drug/device data, adverse events, recalls, and regulatory submissions to enable safety research and analysis.

Python library for statistical data visualization with pandas integration, featuring built-in exploratory plots like box plots and heatmaps.

Enables drug discovery research via ZINC database access, supporting compound searches by ID, SMILES, similarity, and 3D structure docking for virtual screening.