PTM-SD: a database of structurally resolved and annotated posttranslational modifications in proteins

Abstract

Posttranslational modifications (PTMs) define covalent and chemical modifications of protein residues. They play important roles in modulating various biological functions. Current PTM databases contain important sequence annotations but do not provide informative 3D structural resource about these modifications. Posttranslational modification structural database (PTM-SD) provides access to structurally solved modified residues, which are experimentally annotated as PTMs. It combines different PTM information and annotation gathered from other databases, e.g. Protein DataBank for the protein structures and dbPTM and PTMCuration for fine sequence annotation. PTM-SD gives an accurate detection of PTMs in structural data. PTM-SD can be browsed by PDB id, UniProt accession number, organism and classic PTM annotation. Advanced queries can also be performed, i.e. detailed PTM annotations, amino acid type, secondary structure, SCOP class classification, PDB chain length and number of PTMs by chain. Statistics and analyses can be computed on a selected dataset of PTMs. Each PTM entry is detailed in a dedicated page with information on the protein sequence, local conformation with secondary structure and Protein Blocks. PTM-SD gives valuable information on observed PTMs in protein 3D structure, which is of great interest for studying sequence-structure- function relationships at the light of PTMs, and could provide insights for comparative modeling and PTM predictions protocols. Database URL: PTM-SD can be accessed at http://www.dsimb.inserm.fr/dsimb_tools/PTM-SD/.

Figure 1.

PTM-SD flowchart. Four different databanks are used to generate the data. The protein structures are taken from PDB (17), while PTMs annotations are extracted from dbPTM (14) and PTMCuration (20). UniProt sequences (21) are aligned against the extracted PDB sequences. Thus, we obtained protein structures with PTM annotations and modified residues at exact same positions. At last a semantic mining was made to accept or not the correspondence between the modifications and the annotations.

Figure 2.

Example of PTM-SD usage. (A) A simple mode is available to use PTM-SD. It is possible to look for a list of PDB and/or UniProt ids, combined with a specific organism and (a) particular PTM(s). (B) An advanced mode allows more complex requests, as specific amino acid(s), secondary structure(s), SCOP fold(s), number of PTM by PDB chain, length of PDB chain and detailed annotation(s) found in PDB records and dbPTM. By clicking on the search button, (C) a results table appears. It gives for each entry the information on organism, cross-linking with the PDB and UniProt id, the precise position of the PTM in sequence and structural data, PTM annotations and its structural environment in terms of PBs and secondary structures. (D) By clicking on the ‘details’ link, the visitor is redirected to a new page containing extra information on the selected PTM site. Related PTM-SD entries found in same PDB chain and same UniProt AC are accessible through cross-link. Image gallery was done, thanks to PyMOL software, and below alignment section, done thanks to Clustal W, allows direct observation of the sequence/structure relationship surrounding PTM sites. On this page, scripts for PyMOL software and template sequence for Modeller comparative software are provided. (E) From the complete set of selected entries, it is possible to look at the distribution of organisms, proteins, PDB ids/chains and PTM types. Neq entropy index quantifies the local structural divergences between the PTMs. From this step, the visitor can reduce the protein redundancy into his selected entries. (F) At last he can download the list of PDB chain and the PTM-SD data related to his selection.