G2PDB REST

Latest revision as of 17:53, 10 June 2016

The purpose of this document is to define a set of straightforward REST (REpresentational State Transfer) services to commonly required information based on the data in the novel database g2pDB. g2pDB is a collection of auto-curated post-translational modification acceptor sites mapped to their associated genomic codons (in chromosome coordinates) stored in a MongoDB database. The current version has mappings for the most reproducible human STY-phosphorylation, K-ubiquitinylation and K-acetylation acceptor sites found using the data available in the GPMDB system. The details of the construction of the database and REST interface have been published (doi:10.1021/acs.jproteome.5b01018).

These services will provide simple, non-SQL methods to extract this information and make it available over the Internet to anyone wishing to use it as part of a network-aware application. The services are available to anyone involved in computational, biological or biomedical research. We, the providers, retain the right to block access by any user or system that we feel is using the information or infrastructure inappropriately.

The methods described here use the following base URLs:

1. http://openslice.fenyolab.org/g2pdb; or
2. http://g2pdb.thegpm.org/g2pdb.

Anyone interested in direct access to the production database of experimental results that was used to construct g2pDB should look at the GPMDB REST API.

Web site

A preliminary web site that shows the contents of g2pDB for a particular gene, transcript or protein is available at http://g2pdb.thegpm.org/search.

Conventions

The following statements refer to all methods and description given below:

1. all GETs are made to the base URL, which is either:
   1. "http://openslice.fenyolab.org/g2pdb";
   2. or "http://g2pdb.thegpm.org/g2pdb".;
2. all bases are expressed in single letter code (A, C, T, G);
3. modifications allowed are "Phospho", "Acetyl" or "GlyGly" (ubiquitinyl);
4. genome coordinates refer to Genome Reference Consortium Human genome builds (GRCh37 and GRCh38);
5. protein accession numbers use ENSEMBL numbers for the appropriate build;
6. all return values are in JSON notation (RFC 4627) as either an ARRAY or OBJECT; and
7. protein post-translational modifications currently covered in g2pDB are S,T & Y phosphorylation (MOD=Phospho), lysine ubiquitinylation (MOD=GlyGly) and lysine acetylation (MOD=Acetyl).

Available sequence builds

The protein and genome sequence dependent information requires the specification of particular ENSEMBL and GRCh sequence assemblies. The assemblies currently available are as follows:

1. /grch37/ensembl_70; and
2. /grch38/ensembl_76.

/interface/ services

Interface services give the programmer access to information about the current state of the REST interface.

GET / => [ARRAY]

 input: none
return: [ARRAY - "string"] a text description of the current REST interface

Example: find help information through the interface

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/

/dna/ services

Dna services give access to PTM-linked information about particular genomic locations.

GET /grch37/ensembl_70/dna/CHR/POS => [{OBJECT1},{OBJECT2},...]

 input: CHR = human chromosome 1-22, MT, X, or Y), POS = base position on chromosome 
return: ARRAY of objects containing text description of all modification linked to the specified base

Example: list PTM information linked to base 8925354 on Chromosome 1

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/dna/1/8925354

GET /grch37/ensembl_70/dna/CHR/POS/mod=MOD => [{OBJECT1},{OBJECT2},...]

 input: CHR = human chromosome 1-22, MT, X, or Y), POS = base position on chromosome, MOD = modification
return: ARRAY of objects containing text description of the specified modification linked to the specified base

Example: list acetylation modification information linked to base 8925354 on Chromosome 1

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/dna/1/8925354/mod=Acetyl

GET /grch37/ensembl_70/dna/CHR/POS/snp=BASE => [[{OBJECT1},{OBJECT2},...],[{OBJECT1},{OBJECT2},...]]

 input: CHR = human chromosome 1-22, MT, X, or Y), POS = base position on chromosome, BASE = variant base 
return: ARRAY of objects containing text description of modifications linked to the canonical BASE
 at the given position, followed by ARRAY of objects containing text description of modifications
 changed by the variant BASE

Example: list modification changes caused by the variant base A at position 8925354 on Chromosome 1 (no change)
Note: First and second ARRAYs are identical due to no change as a result of variant base

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/dna/1/8030998/snp=A

Example: list modification changes caused by the variant base C at position 8925354 on Chromosome 1 (removes modification)
Note: Second ARRAY is empty due to removal of modification as a result of the variant base

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/dna/1/8030998/snp=C

/protein/ services

Peptide services access to all of the genomic mapping information associated with PTMs for a particular protein.

GET /grch37/ensembl_70/protein/ACC => [{OBJECT1},{OBJECT2},...]

 input: ACC = ENSEMBL protein accession number
return: ARRAY of PTM mapping information objects

Example: retrieve the PTM mappings for ENSP00000234590

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/protein/ENSP00000234590.

GET /grch37/ensembl_70/protein/ACC/mod=MOD => [{OBJECT1},{OBJECT2},...]]

 input: ACC = ENSEMBL protein accession number, MOD = modification
return: ARRAY of PTM mapping information objects for MOD only

Example: retrieve the ubiquitinylation mappings for ENSP00000234590

http://openslice.fenyolab.org/g2pdb/grch37/ensembl_70/g2pdb/protein/ENSP00000234590/mod=GlyGly.

Source code

These services and for creating g2pDB were implemented in Python. The code is available at ftp://ftp.thegpm.org/projects/g2pDB/human/code.

Data sources

The information used to construct g2pDB can be found at ftp://ftp.thegpm.org/projects/g2pDB. The data directories are as follows:

1. /projects/g2pDB/human/ptm - contains the files mapping the three PTM types in proteome and genome coordinates;
2. /projects/g2pDB/human/gff - contains the corresponding GFF3 genome feature annotation files; and
3. /projects/g2pDB/human/mongodb - the database files used by MongoDB.

Comments and suggestions

Any one interested in making suggestions or commenting on the ideas in this document should send them by email to Ron Beavis, rbeavis@thegpm.org.

Revision date and status

1. 2014.11.26 - first draft available
2. 2014.11.27 - revised URLs to include proteome and genome version information
3. 2015.03.04 - updated text with expanded discussion of g2pDB
4. 2015.10.26 - updated text to better explain some methods