Linkanno:Integrated Bio-molecule Mapping Networks And Their Access Interfaces For Data Annotation

A list of genes,transcripts or proteins is usually a preliminary result of omics data analyses.The interpretation,however,of the properties/features(such as structure,physical and chemical properties,biological function and their variations in disease,as well as the relationship between the different biological molecules,etc.)of these bio-molecules relies on biomedical knowledge accumulated and stored in distributed public databases.Data annotation is a process to associate existing knowledge to the molecules in the list by data mapping,and it has become an indispensable step in omics data analysis pipeline.Because of the distributed and heterogenous nature of data sources,efforts are required to integrate the knowledge/information by resolving the data problems such as data redundancy,semantic ambiguity,conflict or incomplete content,as well as incompatible identifiers(IDs)among the target databases.In this study,we built a bioinformatic toolkit Linkanno that is consisted of backend knowledgebase,frontend web interface,and Restful API for client applications.Briefly,a bio-molecule mapping network was integrated from representative authoritative public databases by ID cross-referencing.We have integrated the gene annotation information in the HUGO Gene Nomenclature(HGNC),Entrez Gene and Ensembl databases,the transcript annotation information in the Ensembl and RefSeq databases,and the protein annotation information in the Ensembl,RefSeq and Swiss-Prot databases.The resulting integrated knowledgebase was optimized for retrieval and storage by using graphic database and HBase technologies.Two main functions realized by Linkanno(http://106.38.63.155:8013/)are information integration by identifier mappings and annotation retrieval,in helping biological researchers to search,browse and download their interested molecular annotation,as well as the relationships through molecule networks.