Skill UI

Builds molecular property prediction and MoleculeNet workflows with DeepChem: SMILES and graph featurization, scaffold or grouped holdouts, masked labels, graph models, and pretrained encoder transfer for ADMET, toxicity and solubility ML.

Generate professional infographics with Nano Banana Pro AI, Gemini quality review, and optional Perplexity research. Supports 10 infographic types, 8 industry styles, colorblind-safe palettes, and smart iterative refinement.

Design, review, migrate, and safely plan MATLAB or GNU Octave numerical workflows covering arrays, tabular/time data, functions, tests, projects, graphics, MAT files, and Python interoperability, with license and execution safety gates.

Matplotlib is a foundational Python library for creating static, animated, and interactive plots with full customization. It supports various plot types like line, scatter, bar, and heatmaps, suitable for scientific and statistical visualizations. Users can customize appearance, create multi-panel figures, and export to PNG, PDF, or SVG. Ideal for publication-quality graphics and integrating with Jupyter notebooks.

Create, analyze, and visualize complex networks in Python with NetworkX 3.x. Build graphs from data, run algorithms like shortest paths, PageRank and centrality, detect communities, generate synthetic networks, and read/write GraphML, JSON or edge-list files.

Build local, research-only computational pathology workflows with PathML: load and tile whole-slide images, run preprocessing and QC pipelines, manage h5path data, quantify multiplex images, construct spatial graphs, and plan bounded model inference.

Query the PrimeKG precision medicine knowledge graph to search genes, drugs, diseases, and phenotypes, retrieve associated neighbors, analyze disease context, and explore drug repurposing paths.

Seaborn is a Python visualization library built on matplotlib for creating statistical graphics. It integrates with pandas for dataset-oriented plotting, offering publication-quality defaults for distributions, relationships, and categorical comparisons. Ideal for exploratory data analysis, multivariate analysis, and complex multi-panel figures. Supports various plot types like box plots, violin plots, and heatmaps.

PyTorch Geometric is a standard library for Graph Neural Networks built on PyTorch. It supports node, link, and graph classification, message passing algorithms like GCN and GAT, heterogeneous graphs, and neighbor sampling. Ideal for building scalable GNN models with custom datasets and transforms.

TorchDrug is a modular PyTorch library for graph learning tasks including molecular property prediction, molecule generation, and knowledge graph reasoning. It provides datasets, models, and training engines for building workflows in computational chemistry and biology. Ensure compatibility with PyTorch 1.8-2.0 and Python 3.7-3.10 before implementation.

This comprehensive guide outlines essential security best practices for integrating with Adobe APIs. It covers securing OAuth credentials, implementing least-privilege scoping, managing zero-downtime credential rotation, and verifying I/O Events webhook signatures using cryptographic methods like RSA-SHA256. Essential for building secure, production-ready applications connected to Adobe services.

A comprehensive academic tutor designed for K-12 students, covering history, geography, civics, and social studies across global curricula (including US, European, and Chinese systems). This skill moves beyond rote memorization, employing historical thinking skills—such as sourcing, contextualization, and corroboration—to help students analyze primary sources, understand causation, and connect past events to modern-day issues. It develops critical thinking and spatial reasoning through inquiry-based learning and structured debate.