AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

database-lookup

Queries documented public database APIs with explicit endpoints, filters, pagination, and provenance. Used when a scientific, regulatory, financial, or other database-backed fact must be retrieved reproducibly from a named source rather than inferred from general knowledge.

Original source
K-Dense-AI
License
MIT
Pinned revision
5f2bc0341354fd5590aef76a533ada1e1e5b4e62

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

Ask your assistant

Use Agentic Memory skills to read k-dense-ai--database-lookup at revision 5f2bc0341354fd5590aef76a533ada1e1e5b4e62, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 83

references/ena.md

European Nucleotide Archive (ENA) API Reference

Overview

The ENA is Europe's primary nucleotide sequence repository, part of the International Nucleotide Sequence Database Collaboration (INSDC) alongside NCBI GenBank and DDBJ. It stores raw sequencing reads, assembled sequences, genome assemblies, and associated metadata. ENA provides five complementary APIs for different access patterns.

1. ENA Portal API (Advanced Search)

Base URL

https://www.ebi.ac.uk/ena/portal/api

No authentication required. All endpoints are public.

Key Endpoints

Search records

GET /search?result={result_type}&query={query}&fields={fields}&limit={N}&format={format}

| Parameter | Type | Description | |-----------|------|-------------| | result | string | Required. Data type to search. See Result Types below. | | query | string | Search query using ENA query syntax. | | fields | string | Comma-separated list of fields to return. Use /returnFields to see available fields per result type. | | limit | int | Bound results explicitly; 0 or omission can request the entire result set. | | format | string | json, tsv (default). |

Query syntax:

tax_id=9606 AND description="*hemoglobin*"
tax_id=9606 AND library_strategy="RNA-Seq"
accession="PRJEB40665"
scientific_name="Escherichia coli" AND dataclass="STD"
country="United Kingdom" AND first_public>2024-01-01

Operators: =, !=, >, <, >=, <=. Use AND, OR, NOT. Wildcards: *. Enclose values with spaces in double quotes.

Example -- search human RNA-Seq runs:

https://www.ebi.ac.uk/ena/portal/api/search?result=read_run&query=tax_id%3D9606%20AND%20library_strategy%3D%22RNA-Seq%22&fields=run_accession,experiment_accession,sample_accession,study_accession,instrument_platform,library_strategy,read_count,base_count&limit=5&format=json

Example -- search nucleotide sequences by organism:

https://www.ebi.ac.uk/ena/portal/api/search?result=sequence&query=tax_id%3D9606%20AND%20description%3D%22*hemoglobin*%22&fields=accession,description,tax_id,scientific_name,base_count&limit=5&format=json

Response:

[
  {
    "accession": "AA126503",
    "description": "zk94h05.s1 Soares_pregnant_uterus_NbHPU Homo sapiens cDNA clone ...",
    "tax_id": "9606"
  }
]

Count records

GET /count?result={result_type}&query={query}

Returns a plain integer count.

Example:

https://www.ebi.ac.uk/ena/portal/api/count?result=read_run&query=tax_id%3D9606%20AND%20library_strategy%3D%22RNA-Seq%22

List available result types

GET /results?format=json

List searchable fields for a result type

GET /searchFields?result={result_type}

List returnable fields for a result type

GET /returnFields?result={result_type}

Result Types

| Result Type | Description | |-------------|-------------| | sequence | Nucleotide sequences | | coding | Coding sequences (CDS) | | noncoding | Non-coding sequences | | read_run | Raw sequencing reads (runs) | | read_experiment | Sequencing experiments | | read_study | Studies for raw reads | | analysis | Analyses | | analysis_study | Studies for analyses | | assembly | Genome assemblies | | sample | Samples | | study | Studies | | taxon | Taxonomic classification | | wgs_set | Genome assembly contig sets (WGS) | | tsa_set | Transcriptome assembly contig sets (TSA) | | tls_set | Targeted locus study contig sets (TLS) |


2. ENA Browser API (Record Retrieval)

Base URL

https://www.ebi.ac.uk/ena/browser/api

Use this for direct retrieval of records by accession number.

Key Endpoints

Retrieve record in XML format

GET /xml/{accession}

Example:

https://www.ebi.ac.uk/ena/browser/api/xml/PRJEB40665
https://www.ebi.ac.uk/ena/browser/api/xml/SRR12345678
https://www.ebi.ac.uk/ena/browser/api/xml/ERS1234567

Retrieve record in EMBL flat file format

GET /embl/{accession}

Example:

https://www.ebi.ac.uk/ena/browser/api/embl/AY585947

Supports ?lineLimit=N to truncate long records.

Retrieve sequence in FASTA format

GET /fasta/{accession}

Example:

https://www.ebi.ac.uk/ena/browser/api/fasta/AY585947

Response Formats

| Endpoint | Format | Use case | |----------|--------|----------| | /xml/{accession} | XML | Full structured metadata for studies, samples, experiments, runs | | /embl/{accession} | EMBL flat file | Annotated sequences with features | | /fasta/{accession} | FASTA | Raw nucleotide/protein sequences |

Accession Types

| Prefix | Entity | Example | |--------|--------|---------| | PRJEB / PRJNA / PRJDB | Study/Project | PRJEB40665 | | ERX / SRX / DRX | Experiment | ERX1234567 | | ERS / SRS / DRS | Sample | ERS1234567 | | ERR / SRR / DRR | Run | ERR1234567 | | GCA | Genome assembly | GCA_000001405.29 | | Standard INSDC | Sequence | AY585947, M10051 |


3. ENA Taxonomy REST API

Base URL

https://www.ebi.ac.uk/ena/taxonomy/rest

Key Endpoints

Lookup by taxonomy ID

GET /tax-id/{taxId}

Example:

https://www.ebi.ac.uk/ena/taxonomy/rest/tax-id/9606

Response:

{
  "taxId": 9606,
  "scientificName": "Homo sapiens",
  "commonName": "human",
  "formalName": true,
  "rank": "species",
  "division": "HUM",
  "lineage": "Eukaryota; Metazoa; Chordata; Craniata; Vertebrata; ...; Homo; ",
  "geneticCode": "1",
  "mitochondrialGeneticCode": "2",
  "submittable": true,
  "binomial": true,
  "metagenome": false,
  "otherNames": [
    {"nameClass": "authority", "name": "Linnaeus, 1758"},
    {"nameClass": "genbank common name", "name": "human"}
  ]
}

Search by scientific name

GET /scientific-name/{name}

Example:

https://www.ebi.ac.uk/ena/taxonomy/rest/scientific-name/Homo%20sapiens

Returns an array of matching taxonomy records.

Search by common name

GET /any-name/{name}

Example:

https://www.ebi.ac.uk/ena/taxonomy/rest/any-name/human

Suggest names (autocomplete)

GET /suggest-for-submission/{partialName}

Example:

https://www.ebi.ac.uk/ena/taxonomy/rest/suggest-for-submission/Homo%20sap

4. ENA Cross Reference Service

Base URL

https://www.ebi.ac.uk/ena/xref/rest

Retrieves links between ENA records and external databases (UniProt, PDB, PubMed, etc.).

Key Endpoints

Search cross-references by accession

GET /json/search?accession={accession}

| Parameter | Type | Description | |-----------|------|-------------| | accession | string | ENA accession to look up cross-references for. | | source | string | Filter by source database (e.g., UniProtKB). | | target | string | Filter by target type. | | limit | int | Max results. | | offset | int | Pagination offset. |

Example:

https://www.ebi.ac.uk/ena/xref/rest/json/search?accession=A00145

Response:

[
  {
    "Source": "EuropePMC",
    "Source Primary Accession": "PMC12345",
    "Source Secondary Accession": "",
    "Source URL": "https://europepmc.org/...",
    "Source Secondary URL": "",
    "Target": "sequence",
    "Target Primary Accession": "A00145",
    "Target Secondary Accession": "",
    "Target URL": "https://www.ebi.ac.uk/ena/...",
    "Has Inferred": "N",
    "Inferred From": ""
  }
]

5. CRAM Reference Registry

Base URL

https://www.ebi.ac.uk/ena/cram

Retrieves reference sequences used in CRAM file compression.

Key Endpoints

Lookup by MD5 checksum

GET /md5/{md5}

Example:

https://www.ebi.ac.uk/ena/cram/md5/b1eba5b6e4440e22e1e02f7e0febd2da

Lookup by SHA1 checksum

GET /sha1/{sha1}

Inspect the CRAM registry response: checksum retrieval can return raw sequence rather than FASTA. Do not pass an unverified body to a FASTA parser. Verify the checksum of the normalized sequence against the requested registry identifier.


Common Search Patterns

# All RNA-Seq runs for a species
result=read_run&query=tax_id=9606 AND library_strategy="RNA-Seq"

# WGS assemblies for an organism
result=assembly&query=tax_id=562 AND assembly_type="primary metagenome"

# Sequences by study accession
result=sequence&query=study_accession="PRJEB40665"

# Samples from a country with collection date
result=sample&query=country="Germany" AND collection_date>=2024-01-01

# Coding sequences for a gene keyword
result=coding&query=description="*BRCA1*" AND tax_id=9606

# Count available datasets
/count?result=read_run&query=tax_id=9606

# Get metadata fields available for a result type
/returnFields?result=read_run
/searchFields?result=read_run

Rate Limits

  • No authentication required
  • No formal published rate limit, but be courteous: avoid more than ~5 concurrent requests
  • Portal /search has no documented offset. A limit can truncate results; use /count to check size and retrieve a complete bounded query or partition by supported indexed fields. Do not repeatedly request an ignored offset.
  • For bulk downloads of sequence data (FASTQ, etc.), use ENA's FTP/Aspera services rather than the REST API

Tips

  • ENA vs SRA: ENA and NCBI SRA mirror each other's data (both are INSDC members). ENA accessions (ERR/ERX/ERS/PRJEB) and NCBI accessions (SRR/SRX/SRS/PRJNA) are cross-referenced. Use whichever API has the query features you need.
  • Portal API for search, Browser API for retrieval: Use the Portal API when you need to search across many records with filters. Use the Browser API when you have a specific accession and want the full record.
  • JSON vs XML: The Portal API returns JSON or TSV. The Browser API primarily returns XML, EMBL, or FASTA.
  • Field discovery: Always check /returnFields?result={type} and /searchFields?result={type} to see what's available for each result type -- fields differ between result types.
  • Cross-references: Use the xref service to find links from ENA records to UniProt, PDB, PubMed, and other databases.