Skill UI

Integrate with the DNAnexus cloud genomics platform to build apps/applets, manage FASTQ/BAM/VCF data, run workflows, and script automations via dxpy for biomedical pipelines.

Conduct rigorous, multi-database literature reviews with systematic workflows, verified citations in styles like APA or Nature, and AI-enhanced schematics for publication-ready markdown and PDF outputs in biomedical and technical domains.

TorchDrug is a PyTorch-native toolbox for building graph neural networks tailored to molecules, proteins, and biomedical knowledge graphs, covering property prediction, generation, retrosynthesis, and knowledge graph reasoning with curated datasets and models.

Leverages 78 public scientific, biomedical, materials, and economic REST APIs to find compounds, genes, pathways, clinical trials, patents, economic indicators, weather, and other structured data, returning raw JSON plus the databases/endpoints used.

Collection of battle-tested PyTorch training recipes covering architecture selection, optimizers, LR scheduling, precision tips, and debugging to train or fine-tune LLMs, vision, diffusion, and biomedical models with high GPU throughput.

A comprehensive cloud platform for biomedical data analysis and genomics research. This skill provides tools for building, deploying, and managing complex bioinformatics pipelines (apps/applets). Users can manage diverse data types (FASTQ, BAM, VCF), execute advanced workflows, and interact with the platform programmatically using the dxpy Python SDK.

This is a powerful Python library built on scikit-learn for conducting advanced survival analysis and time-to-event modeling. It excels at handling censored data (e.g., right-censored) and allows users to fit various models, including Cox proportional hazards, Random Survival Forests, and Gradient Boosting. Users can also evaluate model performance using metrics like the Concordance Index (C-index) and Brier score, making it ideal for biomedical, clinical, or reliability engineering applications.

TorchDrug is a comprehensive PyTorch-based machine learning toolbox specializing in computational chemistry and drug discovery. It enables users to apply sophisticated Graph Neural Networks (GNNs) to tackle complex tasks, including molecular property prediction, protein structure modeling, knowledge graph reasoning, molecular generation, and retrosynthesis planning. It is highly valuable for researchers working with SMILES, PDB, and biomedical datasets.

Access 78 public REST APIs spanning biomedical, materials science, physics, environment, and economics. This skill allows users to query vast datasets for compounds, genes, proteins, patents, and economic indicators by selecting the relevant database and executing precise API calls. Ideal for comprehensive scientific research and data gathering.

A comprehensive research tool providing access to 10 major academic databases, including PubMed, arXiv, Crossref, OpenAlex, and Semantic Scholar. Use it to search for scholarly articles, preprints, citations, full-text PDFs, and author profiles across multiple fields (biomedical, physics, CS). It integrates complex cross-database queries, ensuring users can find the most complete and up-to-date information on any research topic or identifier (DOI, PMID, etc.).

This guide provides a comprehensive workflow for searching and retrieving biomedical literature from PubMed and NCBI. It covers advanced query construction using MeSH terms and specific field tags, advanced filtering (date, publication type), and a repeatable programmatic workflow using NCBI E-utilities (esearch, efetch). Essential for researchers conducting systematic reviews and literature monitoring.

The Brain Imaging Data Structure (BIDS) is a community standard defining a consistent file naming convention and directory hierarchy for organizing diverse neuroscientific and biomedical datasets. This skill helps users manage, validate, query, and convert various modalities (MRI, EEG, PET, behavioral, etc.) into a standardized, shareable format, ensuring data integrity for research and publication.