Biothings API

Overview

BioThings.io is a collection of APIs that provide comprehensive data annotation services for various types of biomedical information. It is designed to facilitate data access and integration for researchers working in genomics, drug discovery, and related fields. BioThings APIs are optimized for high-speed queries and data annotation, making it an essential tool for accessing gene, variant, chemical, and disease-related information.

Key BioThings APIs

BioThings offers four main services, each specializing in a specific type of annotation:

1. MyGene.info: Provides gene-centric annotation, including gene descriptions, biological functions, interactions, and pathway information.
2. MyVariant.info: Focuses on variant annotation, offering detailed insights into genetic variations, their impacts, and clinical significance.
3. MyChem.info: A chemical and drug annotation service that provides data on chemical structures, drug interactions, and pharmacological information.
4. MyDisease.info: Specializes in disease annotation, linking diseases to associated genes, phenotypes, clinical trials, and more.

Each of these services is designed to be scalable, fast, and accessible via a simple RESTful API interface.

Services Overview

1. MyGene.info

• Description: MyGene.info offers a powerful gene annotation service, providing researchers access to comprehensive gene-centric data. It aggregates information from various sources such as NCBI, Ensembl, UniProt, and more.
• Key Features:
  • Gene descriptions and summaries
  • Biological functions and interactions
  • Pathway information
  • Support for human and model organism genes

2. MyVariant.info

• Description: MyVariant.info is a variant-centric service that delivers high-quality annotations for genetic variations. It aggregates variant data from multiple sources, including dbSNP, ClinVar, and ExAC.
• Key Features:
  • Population frequencies
  • Clinical significance and pathogenicity
  • Functional predictions and annotations
  • Support for SNPs, indels, and structural variants

3. MyChem.info

• Description: MyChem.info provides an efficient way to retrieve chemical and drug annotations. It integrates data from ChEMBL, DrugBank, PubChem, and other chemical databases.
• Key Features:
  • Chemical structure and formula
  • Drug interactions and adverse effects
  • Pharmacological properties and indications
  • Toxicology and safety information

4. MyDisease.info

• Description: MyDisease.info is a disease-centric annotation service designed to provide detailed information about human diseases, including gene-disease associations, clinical trials, and ontology information.
• Key Features:
  • Disease descriptions and ontology terms
  • Gene and variant associations
  • Links to clinical trials and disease models
  • Disease-gene interaction networks

How to Access

Each of the BioThings APIs provides a well-documented RESTful interface, allowing users to send simple HTTP queries to retrieve data in JSON format. The APIs support a wide range of query options, including gene/variant/disease names, chemical identifiers, and specific properties.

Biothing's API (https://biothings.io/)

MyGene.info (https://mygene.info/)

MyVariant.info (https://myvariant.info/)

MyChem.info (https://mychem.info/)

MyDisease.info (https://mydisease.info/)

Querying Genes

You can query genes from BioThings' MyGene.info API using a variety of identifiers and attributes. Here are the main types of queries you can run:

1. Retrieve Gene Information by Specific ID

• You can look up detailed gene information using various gene identifiers:
  • Entrez Gene ID: https://mygene.info/v3/gene/1017
  • Ensembl Gene ID: Similar format to Entrez ID queries.
  • UniProt ID: Example query: https://mygene.info/v3/query?q=uniprot:P24941

2. Search by Gene Symbol

• Retrieve genes by their symbol (e.g., CDK2):
  • https://mygene.info/v3/query?q=symbol:cdk2

3. Search by Gene Summary Text

• Search for genes based on their descriptions or related summaries. For example, to get genes related to "Insulin":
  • https://mygene.info/v3/query?q=summary:insulin

4. Retrieve Data Source Metadata

• Retrieve metadata about the sources of gene data:
  • https://mygene.info/metadata

5. Get a List of All Available Fields

• Get a list of all fields available in the MyGene.info database:
  • https://mygene.info/metadata/fields

These queries provide flexible options for retrieving gene-centric data using different identifiers, symbols, or biological descriptions.

Querying Variants

You can query variants from BioThings' MyVariant.info API using various types of identifiers, genomic coordinates, and annotation criteria. Below is a summary of the main types of queries:

1. Retrieve Variant by Specific HGVS ID

• Look up variant information using a specific HGVS (Human Genome Variation Society) ID:
  • https://myvariant.info/v1/variant/chr7:g.55241707G>T

2. Retrieve Variants Based on Specific Annotations

• Query for variants that have specific annotations, such as those annotated in ClinVar:
  • https://myvariant.info/v1/query/?q=_exists_:clinvar

3. Find Variants for a Specific Gene

• Retrieve all non-synonymous variants for a given gene using its symbol, e.g., BTK:
  • https://myvariant.info/v1/query/?q=dbnsfp.genename:BTK

4. Search Variants by Genomic Range

• Search for variants within a specified genomic range:
  • https://myvariant.info/v1/query/?q=chr1:69000-70000

5. Complex Queries Combining Multiple Annotations

• Create advanced queries that combine multiple annotations (e.g., variants with wellderly annotation and possibly damaging CADD PolyPhen predictions):
  • https://myvariant.info/v1/query/?q=_exists_:wellderly AND cadd.polyphen.cat:possibly_damaging&fields=wellderly,cadd.polyphen

6. Retrieve Data Source Metadata

• Retrieve metadata about the sources of variant data:
  • https://mygene.info/metadata/

7. Get a List of All Available Fields

• Get a complete list of all fields available in MyVariant.info:
  • https://mygene.info/metadata/fields

These queries enable a comprehensive exploration of variant data using various identifiers, genomic ranges, annotations, and gene symbols.

Example Gene querying results

Here's a categorized table summarizing the useful information from the given JSON file:

Category	Field Name	Description/Content
Identifiers	AllianceGenome	1771
	HGNC	1771
	MIM	116953
	_id	1017
	entrezgene	1017
	ensembl	Gene: ENSG00000123374, Protein: multiple ENSP, Transcript: multiple ENST
	symbol	CDK2 (Cyclin Dependent Kinase 2)
	other_names	cdc2-related protein kinase, cell division protein kinase 2, p33 protein kinase
Accession Numbers	accession	Genomic, Protein, RNA, Translation
	protein	Various accession numbers: e.g., AAA35667.1, NP_001277159.1
	rna	Various accession numbers: e.g., NM_001290230.2, XM_011537732.2
Aliases	alias	CDKN2, p33(CDK2)
Chemical Data	chembl	Multiple entries linking to ChEMBL data, e.g., CHEMBL4106153
Genomic Position	genomic_pos	chr 12: 55966781-55972789, Strand: 1
	genomic_pos_hg19	chr 12: 56360553-56366568, Strand: 1
Ensembl Information	ensembl	Gene: ENSG00000123374, multiple protein and transcript identifiers
	type_of_gene	protein_coding
Functional Annotations	go	Gene Ontology (GO) annotations for BP, MF, and CC categories
	BP	Biological processes like G1/S transition, protein phosphorylation, signal transduction
	CC	Cellular components: nucleus, centrosome, cyclin-CDK complex
	MF	Molecular functions: ATP binding, cyclin binding, protein kinase activity
Orthologs	orthologs	Homologous genes across multiple species
Pathway Information	pathway	Various pathway identifiers: BioCarta, KEGG, NetPath, PID, Reactome, WikiPathways
Protein Domains	interpro	Protein domain information, e.g., Serine/threonine-protein kinase, ATP binding site
Protein Structure	pdb	Multiple PDB structure identifiers for CDK2
Pharmaceutical Data	pharmgkb	PA101
Proteomics	refseq	Reference sequence identifiers for genomic, protein, and RNA
	translation	Protein to RNA translation mappings
Gene Information	summary	Brief gene description
	map_location	12q13.2
Protein Interactions	ipi	IPI identifiers
Transcript Information	exons	Genomic exon positions for various transcripts
	exons_hg19	Exon positions in HG19 genome version
	transcript	Associated transcripts, e.g., ENST00000266970.4
Gene Function Information	generif	PubMed-based gene references for functional studies
	pubmed	PubMed IDs
	text	Description of study findings
Evolutionary Data	homologene	Homologene gene entries
	id	74409
	genes	Multiple homologous genes
Protein Family	pantherdb	PANTHER family information
Experimental Data	reagent	Experimental reagents (siRNA, shRNA, etc.) used for the gene
Wikipedia Information	wikipedia	URL: Cyclin-dependent kinase 2
Protein Interactions	reporter	Gene expression array probe identifiers
UniProt Data	uniprot	Swiss-Prot: P24941, TrEMBL: multiple entries
	unii	IEQ8K30M47

Example Variant querying results

Here is a table summarizing the key categories and fields from the given JSON file:

Category	Field	Value/Description
General Information	_id	chr7:g.55241707G>T
	_version	2
Variant Type	type	SNV (Single Nucleotide Variant)
	ref	G
	alt	T
	chrom	7
	pos	55241707
	rsid	rs28929495
	observed	true
CADD Scores	cadd.phred	26.2
	cadd.annotype	CodingTranscript
	cadd.bstatistic	822
	cadd.consdetail	missense
	cadd.consequence	NON_SYNONYMOUS
	cadd.gc	0.57
Gene Information	cadd.gene.genename	EGFR
	cadd.gene.ccds_id	CCDS5514.1
	cadd.gene.feature_id	ENST00000275493
	cadd.gene.gene_id	ENSG00000146648
	cadd.gene.cds.cds_pos	2155
	cadd.gene.prot.protpos	719
Protein Change	clinvar.hgvs.protein	Various entries, e.g., NP_001333827.1:p.Gly719Cys
	cadd.oaa	G (Original Amino Acid)
	cadd.naa	C (New Amino Acid)
	cgi.protein_change	EGFR:G719C
Clinical Information	clinvar.rcv[].clinical_significance	Various entries: drug response, Pathogenic, Likely pathogenic
	clinvar.rcv[].conditions.name	Various entries: Nonsmall cell lung cancer, response to tyrosine kinase inhibitor in, somatic
	clinvar.rcv[].origin	somatic
Drug Response	cgi[].association	Responsive
	cgi[].drug	Various entries: HSP90 inhibitors, EGFR TK inhibitors, Afatinib, Gefitinib
	cgi[].primary_tumor_type	Various entries: Lung, Non-small cell lung
	cgi[].evidence_level	Various entries: FDA guidelines, Early trials, NCCN guidelines
Functional Annotations	dbnsfp.aa.alt	C
	dbnsfp.appris	principal1
	dbnsfp.eigen.raw_coding	0.953480363944939
Pathogenicity Predictions	polyphen.cat	probably_damaging
	polyphen.val	0.998
	sift.cat	deleterious
	sift.val	0
Conservation Scores	gerp.rs	387.1
	phylop.mammalian	2.742
	phast_cons.mammalian	1.0
	fitcons	0.604399
Regulatory Annotations	cadd.encode.h3k27ac	2.48
	cadd.encode.exp	341.62
	cadd.encode.nucleo	1.8
	cadd.mapability.20bp	1
Pharmacogenomics	clinvar.rcv[].conditions.identifiers	Various entries, e.g., medgen: C4016032, mondo: MONDO:0005233
Publications	dbsnp.citations	Various PubMed IDs: 15118073, 15329413
	docm.pubmed_id	15118073, 19922469
	mutdb.cosmic_id	20881, 6253
	clinvar.variant_id	16611

Documentation and Community

For detailed documentation and usage examples, visit BioThings.io Documentation. The community is active and provides support through forums, tutorials, and GitHub repositories.

Conclusion

BioThings.io offers a comprehensive suite of annotation services for researchers and bioinformaticians, enabling easy access to high-quality data for genes, variants, chemicals, and diseases. Its APIs are built to be fast, scalable, and easy to integrate, making it a valuable resource for a wide range of biomedical research applications.