---
title: UniProt Search outputs
description: Read entry_type, function, GO terms and feature positions off each entry, then follow cross_references to PDB, ChEMBL and AlphaFold or fold the sequence.
tags:
  - uniprot
  - outputs
---

The search returns protein entries with annotations, GO terms, features, and cross-references. When the amino acid sequence is included, it feeds directly into ESMFold structure prediction, a BLAST similarity search, or a Target Passport.

## Output structure

UniProt Search returns a JSON object containing matched protein entries with annotations and metadata.

## Top-level fields

| Field | Type | Description |
|-------|------|-------------|
| `total_results` | integer | Total matching entries in UniProt |
| `returned_count` | integer | Number of entries returned in this response |
| `query_info` | object | Search parameters used |
| `entries` | array | List of protein entry objects |

## Protein entry fields

Each entry in the `entries` array contains:

| Field | Type | Description |
|-------|------|-------------|
| `accession` | string | UniProt accession ID (e.g., "P00533") |
| `entry_name` | string | UniProt entry name (e.g., "EGFR_HUMAN") |
| `entry_type` | string | "Swiss-Prot" (reviewed) or "TrEMBL" (unreviewed) |
| `protein_name` | string | Recommended protein name |
| `gene_name` | string | Primary gene symbol |
| `organism` | string | Source organism scientific name |
| `function` | string | Functional annotation summary |
| `disease_associations` | string[] | Known disease links |
| `subcellular_location` | string | Where the protein is found in the cell |
| `go_terms` | object | Gene Ontology annotations (see below) |
| `features` | object[] | Protein features (included unless you ask to leave them out) |
| `sequence` | string | Full amino acid sequence (when you ask for the sequence) |
| `sequence_length` | integer | Number of amino acids |
| `cross_references` | object | Links to external databases |

### GO terms object

| Field | Type | Description |
|-------|------|-------------|
| `molecular_function` | string[] | What the protein does (e.g., "protein tyrosine kinase activity") |
| `biological_process` | string[] | Pathways and processes (e.g., "signal transduction") |
| `cellular_component` | string[] | Where in the cell (e.g., "plasma membrane") |

### Feature object fields

| Field | Type | Description |
|-------|------|-------------|
| `type` | string | Feature type (Domain, Active site, Binding site, Modified residue, etc.) |
| `description` | string | Feature description |
| `start` | integer | Start position in the sequence |
| `end` | integer | End position in the sequence |
| `evidence` | string | Evidence code (experimental, predicted, etc.) |

Common feature types:

| Type | Meaning |
|------|---------|
| Domain | Structural or functional domain (e.g., "Protein kinase") |
| Active site | Catalytically active residue |
| Binding site | Substrate or cofactor binding residue |
| Modified residue | Post-translational modification site (phosphorylation, glycosylation, etc.) |
| Signal peptide | N-terminal signal sequence |
| Transmembrane | Membrane-spanning region |
| Disulfide bond | Cysteine-cysteine bond |
| Mutagenesis | Known mutation with phenotypic effect |

### Cross-references object

| Field | Type | Description |
|-------|------|-------------|
| `pdb_ids` | string[] | Associated PDB structure IDs |
| `chembl_id` | string | ChEMBL target identifier |
| `alphafold_url` | string | AlphaFold predicted structure URL |
| `pfam_domains` | string[] | Pfam domain family IDs |
| `interpro_ids` | string[] | InterPro family identifiers |
| `gene_id` | string | NCBI Gene ID |

## Entry type interpretation

| Entry type | Meaning | Reliability |
|------------|---------|-------------|
| Swiss-Prot | Manually curated by expert biologists | High — experimental evidence reviewed |
| TrEMBL | Computationally annotated | Variable — functional predictions may be incomplete |

Always check the entry type when assessing annotation quality. Swiss-Prot entries have undergone expert review, while TrEMBL entries are automatically generated.

## Example output

```json
{
  "total_results": 1,
  "returned_count": 1,
  "entries": [
    {
      "accession": "P00533",
      "entry_name": "EGFR_HUMAN",
      "entry_type": "Swiss-Prot",
      "protein_name": "Epidermal growth factor receptor",
      "gene_name": "EGFR",
      "organism": "Homo sapiens",
      "function": "Receptor tyrosine kinase binding ligands of the EGF family and activating several signaling cascades to convert extracellular cues into appropriate cellular responses.",
      "disease_associations": ["Lung cancer", "Glioblastoma"],
      "subcellular_location": "Cell membrane; Single-pass type I membrane protein",
      "go_terms": {
        "molecular_function": ["protein tyrosine kinase activity", "ATP binding"],
        "biological_process": ["signal transduction", "cell proliferation"],
        "cellular_component": ["plasma membrane", "receptor complex"]
      },
      "features": [
        {
          "type": "Domain",
          "description": "Protein kinase",
          "start": 712,
          "end": 979
        },
        {
          "type": "Active site",
          "description": "Proton acceptor",
          "start": 837,
          "end": 837
        }
      ],
      "sequence": "MRPSGTAGAALLALLAAL...",
      "sequence_length": 1210,
      "cross_references": {
        "pdb_ids": ["1M17", "1NQL", "2GS6", "3POZ", "4HJO"],
        "chembl_id": "CHEMBL203",
        "alphafold_url": "https://alphafold.ebi.ac.uk/entry/P00533",
        "pfam_domains": ["PF07714", "PF00069"],
        "gene_id": "1956"
      }
    }
  ]
}
```

## Canvas visualization

When UniProt Search results appear on the canvas:

- **Border color**: Blue-grey (#607D8B)
- **Tier 2 metrics**: Entry count, organism, sequence length, annotation count

### Interpreting canvas metrics

| Metric | Meaning |
|--------|---------|
| Swiss-Prot entry type | High-quality, manually reviewed annotations |
| Many PDB cross-references | Well-studied protein with experimental structures available |
| Long feature list | Rich annotation with many known domains and sites |
| Sequence included | Ready for downstream structure prediction or similarity search |

## Downstream data use

### Chaining into ESMFold

When you ask for the sequence, the returned sequence can be passed directly to ESMFold for structure prediction. The orchestrator handles this automatically when you say "fold this protein" after a UniProt search.

### Chaining into BLAST

The protein sequence feeds into BLAST for similarity searches across NCBI databases. Useful for finding homologs, orthologs, or related proteins.

### Chaining into Target Passport

UniProt annotations (function, disease associations, GO terms, druggability indicators) are primary data sources for Target Passport generation.

## Error responses

| Field | Type | Description |
|-------|------|-------------|
| `error` | string | Error message describing the failure |
| `suggestion` | string | Guidance for resolving the issue |

Common error values:
- `"Provide accession, gene, or query"` — No accession, gene, or query was given
- `"No entries found"` — Query matched zero proteins
- `"UniProt API timeout"` — API did not respond in time; retry
- `"Invalid accession format"` — Accession ID does not match expected pattern

## Read next