Biological databases for human research.

The completion of the Human Genome Project lays a foundation for systematically studying the human genome from evolutionary history to precision medicine against diseases. With the explosive growth of biological data, there is an increasing number of biological databases that have been developed in aid of human-related research. Here we present a collection of human-related biological databases and provide a mini-review by classifying them into different categories according to their data types. As human-related databases continue to grow not only in count but also in volume, challenges are ahead in big data storage, processing, exchange and curation.

Introduction

As biological data accumulate at larger scales and increase at exponential paces, thanks principally to higher-throughput and lower-cost DNA sequencing technologies, the number of biological databases that have been developed to manage such data deluge is growing at ever-faster rates. The major objectives of biological databases are not only to store, organize and share data in a structured and searchable manner with the aim to facilitate data retrieval and visualization for humans, but also to provide web application programming interfaces (APIs) for computers to exchange and integrate data from various database resources in an automated manner. Therefore, developing databases to deal with gigantic volumes of biological data is a fundamentally essential task in bioinformatics. To be short, biological databases integrate enormous amounts of omics data, serving as crucially important resources and becoming increasingly indispensable for scientists from wet-lab biologists to in silico bioinformaticians.

According to a report of 2014 Molecular Biology Database Collection in the journal Nucleic Acids Research, there are a sum of 1552 databases that are publicly accessible online [1]. It should be noted, however, that such count of publicly accessible databases is conservative. In fact, there are some databases providing online services without publication in peer-reviewed journal (e.g., The RNA Modification Database at http://mods.rna.albany.edu) or being developed by commercial companies (e.g., Ingenuity Pathway Analysis at http://www.ingenuity.com/products/ipa), making them underrepresented in the scientific community. Considering the continuously proliferating number of biological databases, it becomes increasingly daunting and time-consuming to navigate in the huge volume of databases of interest. The completion of the Human Genome Project in 2003 holds significant benefits for many fields from human evolution to personalized healthcare and precision medicine. In this report, we present a collection of biological databases relevant to human research and provide a mini-review by classifying them into different categories.

Database classification

Biological databases are developed for diverse purposes, encompass various types of data at heterogeneous coverage and are curated at different levels with different methods, so that there are accordingly several different criteria applicable to database classification.

Scope of data coverage

According to the scope of data coverage, biological databases can be classified as comprehensive and specialized databases. Comprehensive databases cover different types of data from numerous species and typical examples are GenBank [2], European Molecular Biology Laboratory (EMBL) [3], and DNA Data Bank of Japan (DDBJ) [4]. These three databases were established as the International Nucleotide Sequence Database Collaboration in 1988 to collect and disseminate DNA and RNA sequences. On the other hand, specialized databases contain specific types of data or data from specific organisms. For example, WormBase [5] is for nematode biology and genomics and RiceWiki [6] is for community curation of rice genes.

Level of biocuration

According to level of data curation, biological databases can roughly fall into primary and secondary or derivative databases. Primary databases contain raw data as archival repository such as the NCBI Sequence Read Archive (SRA) [7], whereas secondary or derivative databases contain curated information as added value, e.g., NCBI RefSeq [8].

Method of biocuration

As a consequence of the explosive growth of data, curation increasingly requires collective intelligence for collaborative data integration and annotation. Therefore, biological databases can also be classified as (1) expert-curated databases, e.g., RefSeq [8] and TAIR, [9] and (2) community-curated databases, which are curated in a collective and collaborative manner by a number of researchers, e.g., LncRNAWiki [10] and GeneWiki [11].

Type of data managed

According to the types of data managed in different databases, biological databases can roughly fall into the following categories: (1) DNA, (2) RNA, (3) protein, (4) expression, (5) pathway, (6) disease, (7) nomenclature, (8) literature, and (9) standard and ontology.

Human-related databases

Decoding the human genome bears great significance in, from a theoretical view, unveiling human evolutionary history, and from an application view, exploring personalized medicine against diverse diseases. Considering the heterogeneity in data type, scope and curation, biological databases can be classified into multiple categories under different criteria as presented above, making it easier for people to effectively characterize databases and identify the database(s) of interest. However, some databases are inaccessible over time or poorly maintained/updated or even never used [12]. In this study, therefore, we assemble a collection of human-related databases that are widely used and currently accessible via the Internet (Table 1). As database classification based on data type is informative and straightforward, we assign one major category to each database, albeit one database may correspond to multiple categories. In what follows, we focus on databases categorized in DNA, RNA, protein, expression, pathway and disease, respectively.

Table 1

Human-related biological databases∗

Name	Link	Brief description	Refs.	Category#
1000 Genomes	http://www.1000genomes.org	A deep catalog of human genetic variation	[17]	DNA
AFND	http://www.allelefrequencies.net	Allele Frequency Net Database	[37]
dbSNP	http://www.ncbi.nlm.nih.gov/snp	Database of single nucleotide polymorphisms	[13]
DEG	http://www.essentialgene.org	Database of Essential Genes	[38]
EGA	http://www.ebi.ac.uk/ega	European Genome–phenome Archive	[14]
Ensembl	http://www.ensembl.org	Ensembl genome browser	[39]
euGenes	http://eugenes.org	Genomic information for eukaryotic organisms	[40]
GeneCards	http://www.genecards.org	Integrated database of human genes	[41]
IMG/HMP	https://img.jgi.doe.gov/imgm_hmp	Human Microbiome MetaGenomes	[15]
JASPAR	http://jaspar.genereg.net	Transcription factor binding profile database	[42]
JGA	http://trace.ddbj.nig.ac.jp/jga	Japanese Genotype–phenotype Archive	[43]
KEGG	http://www.genome.jp/kegg	Kyoto Encyclopedia of Genes and Genomes	[44]
MITOMAP	http://www.mitomap.org	Human mitochondrial genome database	[45]
NCBI RefSeq	http://www.ncbi.nlm.nih.gov/refseq	NCBI Reference Sequence Database	[8]
PolymiRTS	http://compbio.uthsc.edu/miRSNP	Polymorphism in miRNAs and their Target Sites	[46]
UCSC Genome Browser	http://genome.ucsc.edu	UCSC Genome Browser database	[47]
ChIPBase	http://deepbase.sysu.edu.cn/chipbase	Database of transcriptional regulation of lncRNA and miRNA genes	[48]	RNA
DARNED	http://darned.ucc.ie	DAtabase of RNa EDiting in humans	[49]
DIANA-LncBase	http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=lncBase/index	miRNA targets on lncRNAs	[50]
GENCODE	http://www.gencodegenes.org	Encyclopedia of genes and gene variants	[17]
H-DBAS	http://www.h-invitational.jp/h-dbas	Human-transcriptome DataBase for Alternative Splicing	[51]
HEXEvent	http://hexevent.mmg.uci.edu	Database of Human EXon splicing Events	[52]
LNCipedia	http://www.lncipedia.org	Annotated human lncRNA sequences	[53]
LncRNA2Target	http://www.lncrna2target.org	Database of differentially-expressed genes after lncRNA knockdown or overexpression	[54]
lncRNAdb	http://www.lncrnadb.org	lncRNA Database	[20]
lncRNASNP	http://bioinfo.life.hust.edu.cn/lncRNASNP	Database of SNPs in lncRNAs	[55]
LncRNAWiki	http://lncrna.big.ac.cn	Human lncRNA Wiki	[10]
miRBase	http://www.mirbase.org	miRNA Database	[21]
miRTarBase	http://mirtarbase.mbc.nctu.edu.tw	Experimentally-validated miRNA–target interactions	[56]
miRWalk	http://mirwalk.uni-hd.de	Database of miRNA–target interactions	[57]
NONCODE	http://www.noncode.org	Database of ncRNA genes	[58]
NPInter	http://www.bioinfo.org/NPInter	Database of ncRNA interactions	[59]
RADAR	http://RNAedit.com	Rigorously Annotated Database of A-to-I RNA editing	[60]
piRNABank	http://pirnabank.ibab.ac.in	Database of piwi-interacting RNAs	[61]
RBPDB	http://rbpdb.ccbr.utoronto.ca	Database of RNA-binding specificities	[62]
RDB	http://ndbserver.rutgers.edu	The nucleic acid database	[63]
Rfam	http://rfam.xfam.org	Database of ncRNA families	[19]
RNAcentral	http://rnacentral.org	International database of ncRNA sequences	[18]
snoRNABase	https://www-snorna.biotoul.fr	Database of human H/ACA and C/D box snoRNAs	[64]
starBase	http://starbase.sysu.edu.cn	Database of ncRNA interaction networks	[65]
TarBase	http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=tarbase/index	Experimentally-validated miRNA:gene interactions	[66]
TargetScan	http://www.targetscan.org	Predicted miRNA targets in mammals	[67]
CATH	http://cath.biochem.ucl.ac.uk	Protein structure classification	[68]	Protein
CPLM	http://cplm.biocuckoo.org	Compendium of Protein Lysine Modifications	[69]
DIP	http://dip.doe-mbi.ucla.edu	Database of Interacting Proteins	[70]
EKPD	http://ekpd.biocuckoo.org	Eukaryotic Kinase and Phosphatase Database	[71]
HPRD	http://www.hprd.org	Human Protein Reference Database	[72]
hUbiquitome	http://bioinfo.bjmu.edu.cn/hubi/	Ubiquitination sites and cascades	[73]
InterPro	http://www.ebi.ac.uk/interpro	Protein sequence analysis and classification	[74]
MEROPS	http://merops.sanger.ac.uk	Database of proteolytic enzymes, their substrates, and inhibitors	[75]
MINT	http://mint.bio.uniroma2.it/mint	Molecular INTeraction Database	[76]
ModBase	http://salilab.org/modbase	Database of comparative protein structure models	[77]
mUbiSiDa	http://reprod.njmu.edu.cn/mUbiSiDa	Mammalian Ubiquitination Site Database	[78]
PANTHER	http://www.pantherdb.org	Protein ANalysis THrough Evolutionary Relationships	[79]
PDB	http://www.rcsb.org/pdb	Protein Data Bank for 3D structures of biological macromolecules	[25]
PDBe	http://www.ebi.ac.uk/pdbe	Protein Data Bank in Europe	[80]
Pfam	http://pfam.xfam.org	Database of conserved protein families and domains	[23]
PhosSNP	http://phossnp.biocuckoo.org	Genetic polymorphisms that influence protein phosphorylation	[81]
PIR	http://pir.georgetown.edu	Protein Information Resource	[82]
PROSITE	http://www.expasy.org/prosite	Database of protein domains, families and functional sites	[83]
SysPTM	http://lifecenter.sgst.cn/SysPTM	Post-translational modifications	[84]
TreeFam	http://www.treefam.org	Database of phylogenetic trees of animal species	[24]
UniPROBE	http://thebrain.bwh.harvard.edu/uniprobe	Universal PBM Resource for Oligonucleotide Binding Evaluation	[85]
UniProt	http://www.uniprot.org	Universal protein resource	[22]
UUCD	http://uucd.biocuckoo.org	Ubiquitin and Ubiquitin-like Conjugation Database	[86]
ArrayExpress	http://www.ebi.ac.uk/arrayexpress	Database of functional genomics experiments	[87]	Expression
BioGPS	http://biogps.org	Portal for querying and organizing gene annotation resources	[88]
Expression Atlas	http://www.ebi.ac.uk/gxa	Differential and baseline expression	[27]
Human Protein Atlas	http://www.proteinatlas.org	Tissue-based map of the human proteome	[29]
MOPED	https://www.proteinspire.org	Multi-Omics Profiling Expression Database	[89]
NCBI GEO	http://www.ncbi.nlm.nih.gov/geo	Gene Expression Omnibus	[26]
NRED	http://nred.matticklab.com	Database of lncRNA expression	[90]
ONCOMINE	https://www.oncomine.org	Cancer microarray database	[91]
PrimerBank	http://pga.mgh.harvard.edu/primerbank	Public resource for PCR primers	[92]
PRIDE	http://www.ebi.ac.uk/pride	PRoteomics IDEntifications	[93]
TiGER	http://bioinfo.wilmer.jhu.edu/tiger	Tissue-specific Gene Expression and Regulation	[28]
WikiCell	http://www.wikicell.org	Unified resource for Human transcriptomics research	[94]
CPDB	http://consensuspathdb.org	Database of human interaction networks	[95]	Pathway
HMDB	http://www.hmdb.ca	Human Metabolome Database	[96]
KEGG PATHWAY	http://www.genome.jp/kegg/pathway.html	KEGG pathway maps	[30]
MetaCyc	http://metacyc.org	Metabolic pathway database	[97]
Pathway Commons	http://www.pathwaycommons.org	Pathway commons	[98]
PID	http://pid.nci.nih.gov	Pathway Interaction Database	[99]
Reactome	http://www.reactome.org	Curated and peer-reviewed pathway database	[100]
UniPathway	http://www.grenoble.prabi.fr/obiwarehouse/unipathway	Universal Pathway	[101]
AlzBase	http://alz.big.ac.cn/alzBase	Database for gene dysregulation in Alzheimer’s disease	[102]	Disease
CADgene	http://www.bioguo.org/CADgene	Coronary Artery Disease gene database	[103]
COSMIC	http://cancer.sanger.ac.uk	Catalog Of Somatic Mutations In Cancer	[104]
DiseaseMeth	http://bioinfo.hrbmu.edu.cn/diseasemeth	Human disease methylation database	[105]
DisGeNET	http://www.disgenet.org/web/DisGeNET/v2.1	Gene–disease associations	[106]
GOBO	http://co.bmc.lu.se/gobo	Gene expression-based Outcome for Breast cancer Online	[107]
GWAS Central	http://www.gwascentral.org	A comprehensive resource for the comparison and interrogation of genome-wide association studies	[108]
GWASdb	http://jjwanglab.org/gwasdb	Human genetic variants identified by genome-wide association studies	[109]
HbVar	http://globin.cse.psu.edu/hbvar	Hemoglobin variants and thalassemias	[110]
HGMD	http://www.hgmd.org	Human Gene Mutation Database	[111]
ICGC	http://icgc.org	International Cancer Genome Consortium	[33]
IDbases	http://structure.bmc.lu.se/idbase	Immunodeficiency-causing variations	[112]
LncRNADisease	http://cmbi.bjmu.edu.cn/lncrnadisease	lncRNA and disease database	[113]
LOVD	http://www.lovd.nl	Leiden open (source) Variation Database	[114]
MalaCards	http://www.malacards.org	Human maladies and their annotations	[115]
MethHC	http://methhc.mbc.nctu.edu.tw	Database of DNA methylation and gene expression in human cancer	[116]
MethyCancer	http://methycancer.psych.ac.cn	Database of human DNA Methylation and cancer	[117]
miR2Disease	http://www.miR2Disease.org	Database for miRNA deregulation in human disease	[118]
MITOMAP	http://www.mitomap.org/MITOMAP	Polymorphisms and mutations in human mitochondrial DNA	[119]
NHGRI GWAS Catalog	http://www.genome.gov/gwastudies	Curated resource of SNP-trait associations	[120]
OMIM	http://omim.org	Online Mendelian Inheritance in Man	[121]
T2D@ZJU	http://tcm.zju.edu.cn/t2d	Connections associated with type 2 diabetes	[122]
TCGA	http://cancergenome.nih.gov	The Cancer Genome Atlas	[32]
Universal Mutation Database	http://www.umd.be/	Locus-specific database	[123]
ViRBase	http://www.rna-society.org/virbase	Virus–host ncRNA associated interactions	[124]
GO	http://geneontology.org	Gene ontology	[125]	Standard and ontology
HGNC	http://www.genenames.org	Database of human gene names	[126]
Europe PMC	http://europepmc.org	Literature database in Europe	[127]	Literature
PubMed	http://www.ncbi.nlm.nih.gov/pubmed	Database of biomedical literature from MEDLINE	[128]
PubMed Central	http://www.ncbi.nlm.nih.gov/pmc	Free full-text literature archive	[129]

Note: ∗This collection, however, by no means pictures the whole range of human-related databases that are currently available. Primary databases (DDBJ/EMBL/GenBank) are not shown, as they contain raw data.

#A database may correspond to multiple categories and only one major category is shown here.

DNA databases

A DNA database centers on managing DNA data from many or some specific species. The primary function of human DNA databases includes establishment of the reference genome (e.g., NCBI RefSeq [8]), profiling of human genetic variation (e.g., dbSNP [13]), association of genotype with phenotype (e.g., EGA [14]), and identification of human microbiome metagenomes (e.g., IMG/HMP [15]). A representative example of DNA database is GenBank [2], a collection of all publicly-available DNA sequences (http://www.ncbi.nlm.nih.gov/genbank). Since its inception in 1982, GenBank grows at an extraordinary pace and as of December 2014, contains over 184 billion nucleotide bases in more than 179 million sequences (http://www.ncbi.nlm.nih.gov/genbank/statistics).

RNA databases

It is well acknowledged that only a tiny proportion of the human genome is transcribed into mRNAs, whereas the vast majority of the genome is transcribed into “dark matter”—non-coding RNAs (ncRNAs) that do not encode proteins [16], including microRNAs (miRNAs), small nucleolar RNAs (snoRNAs), piwiRNAs (piRNAs), and long non-coding RNA (lncRNA). Therefore, an increasing number of human RNA databases have been built for deciphering ncRNAs (e.g., GENCODE [17]), in particular lncRNAs that attract the rising interest (e.g., LncRNAWiki [10]), and characterizing their functions and interactions (e.g., RNAcentral [18]). A representative example of RNA database is RNAcentral [18]. It provides unified access to the ncRNA sequence data supplied by multiple databases including Rfam [19], lncRNAdb [20], and miRBase [21] (http://rnacentral.org).

Protein databases

The purpose of constructing protein databases includes collection of universal proteins (e.g., UniProt [22]), identification of protein families and domains (e.g., Pfam [23]), reconstruction of phylogenetic trees (e.g., TreeFam [24]), and profiling of protein structures (e.g., PDB [25]). A representative example of protein database is PDB, the main primary database for 3D structures of biological macromolecules determined by X-ray crystallography and NMR. Established in 1971, PDB contains 105,465 biological macromolecular structures as of 30 December 2014, in which 27,393 entries belong to human (http://www.rcsb.org/pdb). Another example is the Universal Protein Resource (UniProt). As a collaborative project between EMBL-EBI, Swiss Institute of Bioinformatics (SIB), and Protein Information Resource (PIR), UniProt provides a comprehensive, high-quality, and freely-accessible resource of protein sequence and functional information. Currently, UniProt includes three member databases: UniProt Knowledgebase (UniProtKB), UniProt Reference Clusters (UniRef), and UniProt Archive (UniParc). In addition, UniProtKB consists of two sections: Swiss-Prot (containing a collection of 547,357 manually-annotated and -reviewed proteins as of January 2015) and TrEMBL (containing a collection of 89,451,166 un-reviewed proteins as of January 2015) (http://www.uniprot.org).

Expression databases

Expression databases can be used for various purposes, including archiving expression data (e.g., GEO [26]), detecting differential and baseline expression (e.g., Expression Atlas [27]), exploring tissue-specific gene expression and regulation (e.g., TiGER [28]), and profiling expression information based on both RNA and protein data (e.g., Human Protein Atlas [29]). A representative case of expression database is Human Protein Atlas. As of 30 December 2014, it encompasses expression profiles for a large majority of human protein-coding genes based on both RNA (transcriptome analysis based on 213 tissue and cell line samples) and protein data (proteome analysis based on 24,028 antibodies) (http://www.proteinatlas.org).

Pathway databases

Pathway databases contain biological pathways for metabolic, signaling, and regulatory pathway analysis. A representative example is KEGG PATHWAY [30], a curated biological pathway resource on the molecular interaction and reaction networks. As the core of KEGG, KEGG PATHWAY integrates many entities that are stored in KEGG sibling databases, including genes, proteins, RNAs, chemical compounds, and chemical reactions (http://www.genome.jp/kegg/pathway.html).

Disease databases

There are at least 200 forms of cancer in the world, causing 14.6% of all human deaths (http://en.wikipedia.org/wiki/Cancer). Thus, obtaining complete cancer genomes and identifying molecular mutations and abnormal genes can provide new insights for cancer prevention, detection, and eventually, personalized treatment [31]. Toward this end, there are two well-known cancer projects, viz., The Cancer Genome Atlas (TCGA) [32] and International Cancer Genome Consortium (ICGC) [33]. TCGA, founded in 2006 by the National Cancer Institute and National Human Genome Research Institute at the National Institutes of Health, aims to collect a wide diversity of omics data (including exome, SNP, mRNA, miRNA, and methylation) for more than 20 different types of human cancer (http://cancergenome.nih.gov). Unlike TCGA, ICGC is a voluntary collaborative organization initiated in 2008 and open to all cancer and genomic researchers in the world. It aims to obtain a comprehensive description of genomic, transcriptomic, and epigenomic changes in 50 different tumor types and/or subtypes, which are of clinical and societal importance across the globe (http://icgc.org).

Perspectives

Here we summarize a collection of biological databases relevant to human research. This collection, however, by no means pictures the whole range of human-related databases that are currently available. As primary databases store raw data, databases in this collection are most derivative databases, which are built from primary databases and contain curated information for different data types, and thus would be of great usefulness for studying the human genome. In the era of big data, human-related biological databases continue to grow not only in count but also in volume, posing unprecedented challenges in data storage, processing, exchange, and curation. From this point, it would be necessary to establish a cloud computing platform to store and process such big data and facilitate construction/update of a secondary or derivative database [34]. As biological databases are physically distributed and heterogeneous in data type and format, it is additionally required to build web open APIs to ease data exchange and sharing among different resources [35]. The last but not the least is curation, which becomes an indispensable part in biological databases, principally because curation involves added value by standardization and quality control and accordingly enhances data interoperability and consistency [36]. Taken together, biological databases hold great utilities for human research and can be regarded as an indicator of our potential to translate big data into big discovery. Considering the current situation in China when compared to other countries, it is our hope that this report may raise the general awareness, albeit better improved nowadays, of the significant role of human-related biological databases not only for academic studies but also for clinical applications.

Competing interests

The authors declared that there are no competing interests.