> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openbio.tech/llms.txt
> Use this file to discover all available pages before exploring further.

# Database Tools

> Query biological and chemical databases

OpenBio provides access to major biological and chemical databases through natural language queries.

## Protein Data Bank (PDB)

**Purpose**: 3D protein structures

**Available queries**:

* Text search by protein names or descriptions
* Search by resolution, method, organism
* Sequence similarity search
* Structure similarity search
* Get details for specific PDB IDs
* Extract sequences from structures

**Example**: "Find structures of human hemoglobin"

## UniProt

**Purpose**: Protein sequences and functional annotations

**Available queries**:

* Search by protein name, gene name, or keywords
* Search by protein function or activity
* Find proteins with specific domains
* Filter by organism
* Get detailed feature annotations

**Example**: "Find proteins with kinase activity in humans"

## AlphaFold DB

**Purpose**: Access AlphaFold protein structure predictions and annotations

**Available queries**:

* Get AlphaFold predictions for UniProt accessions
* View structure quality metrics (pLDDT scores)
* Access structure file URLs (PDB, CIF formats)
* Get structure summaries with coverage information
* Retrieve AlphaFold Missense (AM) annotations
* View predicted aligned error (PAE) data

**Features**:

* **Quality metrics**: Global confidence scores and per-residue pLDDT distributions
* **Structure files**: Direct links to download PDB, CIF, and PAE files
* **Coverage information**: Sequence coverage and confidence scores
* **Annotations**: Missense variant predictions and functional annotations

**Example queries**:

* "Get AlphaFold structure for P01308"
* "Show me AlphaFold predictions for human insulin"
* "What's the AlphaFold structure quality for P12345?"
* "Get AlphaFold annotations for P01308"

<Info>
  AlphaFold DB contains millions of predicted protein structures. Structures are identified by UniProt accession numbers. Use UniProt search first to find the accession, then query AlphaFold DB for the structure.
</Info>

## ChEMBL

**Purpose**: Bioactive drug-like compounds and bioactivity data

**Available queries**:

* Search compounds by name, structure, or properties
* Search by biological target
* Filter by bioactivity (IC50, Ki, EC50)
* Filter by molecular properties
* Filter by clinical development stage

**Example**: "Find kinase inhibitors with IC50 \< 100nM"

## PubChem

**Purpose**: Chemical compounds and substances

**Available queries**:

* Find compounds by name or identifier
* Find structurally similar compounds
* Find compounds containing specific substructures
* Calculate molecular properties
* Access bioactivity assay data

**Example**: "Show me the structure of aspirin"

## PubMed / Literature

**Purpose**: Scientific literature database

**Available queries**:

* Search papers by keywords, topics, or research areas
* Find papers by specific authors
* Filter by journal or publication date
* Filter by article type (review, clinical trial, etc.)
* Find citing papers or references

**Example**: "Find papers about CRISPR-Cas9 from 2024"

## Other Databases

**PDBe**: Enhanced PDB data with additional annotations

* Search with enhanced metadata and ligand interactions

**MyGene**: Gene annotation and information

* Search genes by name or symbol
* Get genomic location and cross-references

**MyDisease**: Disease annotations and associations

* Search diseases and associated genes/variants

**MyVariant**: Genetic variant annotation

* Search variants and get clinical interpretations
* Get population frequency data

**MyTaxon**: Taxonomic information

* Search organisms and get taxonomic lineage

**MyChemInfo**: Chemical annotation and drug information

* Get drug annotations and chemical properties

**ArXiv / BioRxiv**: Preprint servers

* Search preprints before peer review

**Web Search**: General scientific web search

* Search scientific websites and documentation

## Query Tips

* Use specific identifiers (PDB: 1ABC, UniProt: P12345, PubMed: PMID:12345678)
* Always specify the organism (e.g., "human insulin" not "insulin")
* Start broad and narrow down with filters
* OpenBio can query multiple databases in one conversation