ExcellmiRDB for Translational Genomics: A Curated Online Resource for Extracellular MicroRNAs

Abstract

A large number of studies have suggested extracellular microRNAs (microRNAs in biofluids) as potential noninvasive biomarkers for pathophysiological conditions such as cancer. However, reported differentially expressed signatures of extracellular miRNAs in diseases are not uniformly consistent among studies. Here, we present “ExcellmiRDB”, a curated online database that provides integrated information about miRNAs levels in biofluids in a user-friendly way. Although many miRNA databases, including disease-oriented databases, have been launched before, the ExcellmiRDB is so far the only one specialized for storing curated data on miRNA levels in biofluid samples. At present, ExcellmiRDB has 2773 disease-extracellular miRNAs and 1108 biofluid-extracellular miRNAs relationships curated from 108 articles selected from more than 600 surveyed PubMed abstracts. Information about 992 miRNAs, 82 diseases, 21 biofluids, 8 species, 63 normalization reference genes, 5 techniques, 14 GEO profiles accession numbers, 7 human ethnic groups, and 18 compared clinical biomarkers have been provided in the database. A user can query ExcellmiRDB by selecting a disease or a miRNA or a biofluid. Additionally, the database provides two online network graphs to visualize and interact with the content of the database. The first network shows disease-extracellular miRNAs relationships, along with expression patterns and number of articles for a relationship. The second network visualizes biofluid-extracellular miRNAs relationships showing miRNAs spectrum across different types of biofluids. In conclusion, ExcellmiRDB is a new innovative resource for both academic and industrial researchers in translational omics who are developing miRNA biomarkers for noninvasive diagnostic or prognostic technologies. ExcellmiRDB is publicly available on www.excellmirdb.brfjaisalmer.com/.

Introduction

MicroRNAs (miRNAs) are small (18–22 nucleotides) evolutionarily conserved noncoding RNAs. They regulate fundamental cellular and biological processes, such as cell cycle, apoptosis, development, immune response, and also epigenetic processes, by post-transcriptionally suppressed expression of the target genes involved in these processes. miRNAs degrade their targeted messenger RNAs in cytoplasm by binding to 3′-untranslated regions of mRNAs (Bartel, 2004; Benetti et al., 2008; Lewis et al., 2005). Alteration in cellular miRNAs expression profiles were found in many diseases compared to controls. Several miRNAs in tissue biopsy samples were suggested as biomarkers for diagnosis, prognosis, and classification of diseases (Calin and Croce, 2006; Jazbutyte and Thum, 2010).

miRNAs are stable and detectable in biofluids such as blood, urine, saliva, feces, and breast milk. The miRNAs in biofluids are named as extracellular or circulatory or cell-free miRNAs. The mechanism of transportation of the miRNAs to circulation (or cell medium), functions of miRNAs in biofluids, and excretion of miRNAs from biofluids are largely unknown (Link et al., 2010; Michael et al., 2010; Rabinowits et al., 2009; Russo et al., 2012; Turchinovich et al., 2012; Yang et al., 2013). However, significant changes in the expression of extracellular miRNAs were reported in many diseases including cancer, cardiovascular, and infectious diseases, suggesting extracellular miRNAs as potential noninvasive biomarkers for diagnosis and prognosis of pathophysiological conditions (Corsten et al., 2010; Link et al., 2010; Liu et al., 2012; Mitchell et al., 2008; Weber et al., 2010). Also, alterations in the expression of several tissue-specific miRNAs were reported in blood and other biofluids, suggesting extracellular miRNAs can be noninvasive biomarkers for internal tissue injuries (Corcoran et al., 2011; Corsten et al., 2010). For example, liver-enriched miR-122 and miR-192 have significantly higher levels in serum of acetaminophen-induced acute liver injury patients (n=53) compared to healthy controls (n=25) using qRT-PCR (Starkey Lewis et al., 2011).

It has been reported that diagnosis potential of extracellular miRNAs was comparable with commonly used clinical markers (Bryant et al., 2012; Chen et al., 2012; Liu et al., 2012). For instance, Liu et al. (2012), using RT-qPCR, showed serum miR-15b and miR-130b can better classify hepatocellular carcinoma (HCC) (n=30 healthy, 57 patients) with 98.2% sensitivity and 91.5% specificity, compared to serum α-fetoprotein (AFP), the commonly used biomarker for screening HCC (sensitivity: 39%–65% and specificity: 76%–94%).

Despite highly significant results for alteration in the expression of extracellular miRNAs in a disease, results from individual studies do not match or overlap with other studies (Farina et al., 2014; Rao et al., 2013). For example, Fu et al. (2011) reported that miR-518d-5p and miR-93* were most significant differentially expressed miRNAs in serum of tuberculosis patients (n=75) compared to healthy subjects (n=55). However, different miRNAs, miR-19b-2* and miR-3179, using sputum by Yi et al. (2012), were reported as most significant differentially expressed miRNAs in tuberculosis patients (n=58; healthy controls n=38). Both studies used microarray and validated their results by RT-qPCR (Fu et al., 2011; Yi et al., 2012).

In addition to non-overlapping results between individual studies for extracellular miRNAs level in a disease, results could even be opposite. For example, Alexandrov et al. (2012) reported hsa-miR-146a was upregulated in human cerebrospinal fluid (CSF) and tissue-derived extracellular fluid (ECF) of Alzheimer's disease (n=12) using fluorescent miRNA-array-based analysis, and Muller et al. (2014) reported the same miRNA (hsa-miR-146a) was downregulated in cerebrospinal fluid of Alzheimer's disease (n=20) compared to matched controls (n=20) using qRT-PCR.

Moreover, reported differentially expressed extracellular miRNAs are not disease specific. An extracellular miRNA can be associated with several cancers and unrelated diseases. For example, miR-21 is reported upregulated in urine and plasma of patients having acute kidney injury 2 (n=80) compared to controls (n=40) using qRT-PCR (Du et al., 2013), and the same miRNA (miR-21) is also reported significantly differentially expressed in colorectal cancer patients compared to controls (90% specificity and sensitivity) using microfluid array technology (Kanaan et al., 2012). These nonconsistent and scattered information about disease–extracellular miRNAs relationships are available in an increasing number of research articles.

To manage this large amount of information, a database system could be an effective approach. Several miRNA-related databases have been developed. These databases are on-line repositories of different information about miRNAs. For example, miRBase (Griffiths-Jones et al., 2008), miRGator V3.0 (Cho et al., 2013), miRDeep V2 (Friedländer et al., 2012), DeepBase (Yang et al., 2010), and miRNEST (Szczesniak and Makalowska, 2014; Szcześniak et al., 2012) manage annotation and nomenclature of miRNAs; TarBase (Sethupathy et al., 2006), miRNAMap 2.0 (Hsu et al., 2008), microRNA.org (Betel et al., 2008), and mirTarBase (Hsu et al., 2011) provide miRNA- gene target relationships; miR2disase (Jiang et al., 2009), HMDD V2.0 (Li et al., 2014), and PhenomiR (Ruepp et al., 2010) collect and manage miRNA-disease relationships in human and other species. So far, three databases have cataloged information about extracellular miRNAs: Exocarta (Mathivanan and Simpson, 2009), miRandola (Russo et al., 2012), and recently, HMDD V2.0 (Li et al., 2014). miRandola is the only database specific for extracellular miRNAs. These databases mainly provide names of miRNA, biofluid type, and associated diseases. miRandola also provides classification of extracellular miRNAs.

These databases have less or no information about the level of miRNAs in biofluids and factors that can affect disease-extracellular miRNAs relationships. Other than a pathophysiological condition, several biological and technical factors may affect extracellular miRNAs signatures such as sample type, species, ethnic group, sex, isolation kit, technique, normalization methods, and sample size (McDonald et al., 2011; Meyer et al., 2010; Turchinovich et al., 2012). The information about miRNA level in biofluids in disease conditions and factors that can effect disease-extracellular miRNAs relationships are scattered and unstructured in a large number of published articles on extracellular miRNAs. To catalog the above specific information, we developed a manually curated database named “ExcellmiRDB” and provide it in a user-friendly way.

Material and Methods

Data curation

We retrieved abstracts about extracellular miRNAs, published until December, 2013, from PubMed, using the following keywords: [“Circulatory microRNAs ” OR “Extracellular microRNAs”], [“miRNA and biofluid name” e.g., plasma, blood, saliva, urine, semen, vaginal fluid, aqueous humor, vitreous humor, tears, mucus, cerebrospinal fluid, breast milk, follicular fluid]. We surveyed the retrieved abstracts and selected articles to include varieties of diseases, biofluids, techniques, and types of extracellular miRNAs (e.g., cell-free, exosomal, protein bound) as defined by Russo et al. (2012). For manual information annotation, all selected articles were deposited in a library created on Mendeley Software (Lo Russo et al., 2013). First, we defined an experiment in an article as statistical comparison between two biological conditions (e.g., healthy vs. breast cancer) or biofluids or time points. Then, for each defined experiment in an article, we manually curated information shown in Table 1. For each article, the curated information was extracted as a PDF file. Names of deregulated miRNAs and expression pattern information (up/down) were saved in an Excel file. These files were used for final data submission in ExcellmiRDB for each selected article.

Table 1.

Curated Information and Statistics

Information	Statistics	Details/example
Citation	108	Full citation
Year of publication	6 years (2008–2014)	From 2008
PubMed ID	108	PMID:23559272
No. of Experiment	321	Comparing two diseases
For each defined experiment in a selected research article
Condition	82	Breast cancer
Body fluid/sample types	21	Urine
No. of Differential expressed miRNA	992	miR-21
Experiment types	25	Healthy vs. disease
miRNA Name	992	miR-21
Expression pattern	1681 up and 1094 down	Up/Down
Species	8	Human
Technique	8	RT-PCr
Compared clinical biomarkers	37	IL-6
Sample size	Max. 800	20
Human ethnic group	7	White American
Location (country)	16	China
miRNA types	5	Exosomal
miRNAs for normalization	63	U6
GEO/Arrayexpress ID	14	GSE38716
Isolation KIT	16	miVarna
Sensitivity and specificity	32 miRNAs and 28 studies	98%

ExcellmiRDB is a MySQL database running on Apache server with a web interface constructed in HTML, JAVAscript and CSS. PHP was used for server side programming. Cytoscapeweb was used for interactive networks visualization of disease-extracellular miRNAs and biofluid-extracellular miRNAs relationships in the database (Lopes et al., 2010). Node degree (number of the connecting nodes to a node) was calculated for the networks, using NetworkAnalyzer plugin (Assenov et al., 2008). The user can use node degree (e.g., node degree=one) as a filter to select extracellular miRNAs specific to a disease or a biofluid. For each edge (a line connecting a disease—an extracellular miRNA or a biofluid extracellular miRNA) we provided a p-value as the significance of the relationship using the Fisher t-test. We used Fisher's exact test to determine p-values, computed from a 2×2 contingency table composed of: 1) Number of articles having the disease/biofluid and the miRNA; 2) Number of articles that do not have the disease/biofluid but have the miRNA; 3) Number of articles that do not have the miRNA but have the disease/biofluid; and 4) Number of articles that neither have the miRNA nor the disease/biofluid.

Results

ExcellmiRDB cataloged 2773 disease-extracellular miRNAs and 1108 biofluid-extracellular miRNA relationships for 992 miRNAs in 21 biofluids for 82 associated biological conditions curated from 108 articles. These articles were selected by surveying more than 600 abstracts. Table 1 shows the statistics of various curated information. The database is implemented with standard web-technologies as MySQL and PHP and Apache web-server. A novel aspect in accessing the database is added by providing a web-based network graph visualization.

Data access

The database is provided publicly at http://www.excellmirdb.brfjaisalmer.com. It can be queried by selecting a disease (e.g., breast cancer) or an miRNA (e.g., miR-21) or a biofluid (e.g., urine). When a user queries the database for a disease or an miRNA or a biofluid name, the database system searches all experiments containing the queried disease or miRNA or body fluid name. Then, from the selected experiments, the system will retrieve the information such as PubMed Id, biofluid/disease/miRNA, species, technique, GEO dataset, sensitivity, or specificity.

The results will be displayed on a webpage and can be downloaded as a .txt file. The results will have the name of miRNA, link to the corresponding entry of the miRNA in the miRBase database, disease name, expression pattern (up/down), potential biomarker (Yes/No), body-fluid, sample type, sensitivity value, specificity value, and link to the PubMed abstract.

The query can be further filtered by sample size (e.g., <20), technique (e.g., next generation sequencing or RT-PCR), species (e.g., human or pig), availability of GEO data set (YES/NO), allowing combination of intelligent and complex queries for refined relationships of diseases-extracellular miRNAs or biofluid-extracellular miRNAs.

Interactive network visualization for disease-extracellular miRNAs and biofluid-extracellular miRNAs relationships provide a user-friendly novel way to access and visualize information in a global view. To reduce the network complexity, we have provided an option to filter the network by number of articles, expression level (up or down) and node degree. Additionally, we provide a search option on network menu to select a miRNA and a disease/biofluid by name. Also, the user can select a region of the network and the information in that region (i.e., disease and miRNA name, expression pattern, number of articles) will be shown as a table in the bottom of a network.

miRNA–disease network

Consistent deregulated extracellular miRNAs in a specific biological condition (more than one article) may be considered as potential biomarkers for associated diseases. Network visualization of disease-extracellular miRNAs relationships (Fig. 1) along with expression pattern and number of articles, showed that certain relationships were consistent between studies. For example, miR-21 in Lung Cancer, miR-122 in Chronic Hepatitis B, and miR-29a in Pulmonary Tuberculosis were upregulated in at least three research articles.

FIG. 1.

Interactive network of the disease–extracellular miRNA relationships. A node represents a disease or an miRNA. A line connecting two nodes shows a physical relationship based on manually curated data. Color of line represents expression pattern of an miRNA linked with the disease. Thickness of line represents number of articles for that relationship. miRNAs with node degree=1 are shown in the figure. Table at the bottom has information for selected region of the network [i.e., miRNA name, disease names, and p value (Fisher test)].

We found most extracellular miRNAs are overlapped between diseases. For instance, miR-21, miR-122, AND miR-197 were upregulated in 14, 10, and 7 biological conditions, respectively. Similarly, miR-16, miR-223, and miR-122 were downregulated in 10, 9, and 5 biological conditions, respectively. Node degree of an miRNA gives information about number of diseases associated with the miRNA. For instance, with node degree of one, the network will show miRNAs associated with only one disease. In our database, several miRNAs are disease specific, for example, miR-1937 is associated with primary biliary cirrhosis, mir-23b is associated with gastric cancer, and miR-30b is associated with pulmonary tuberculosis, specifically.

miRNA–biofluid network

We visualized the biofluid–miRNA relationships as a network (Fig. 2). The network showed biofluid specific and overlapped extracellular miRNAs. For example, breast milk have certain extracellular miRNAs that are not associated with any other biofluid such as miR-182-5p, miR-191-5p and miR-200c-3p. Additionally, the network showed that highest information were curated for blood samples (30 publications, 491 dysregulated extracellular miRNAs, and 29 diseases for serum).

FIG. 2.

Interactive network of the biofluid–miRNA relationships. A node represents biofluid or an miRNA. A line connecting two nodes shows a physical relationship based on manually curated data. Table at the bottom has information for selected region of the network [i.e., miRNA name, biofluid names, and p value (Fisher test)].

Normalization methods

ExcellmiRDB have extracted technical information such as normalization methods (Fig. 3). The current database showed widely used normalization methods (e.g., U6 in 36 studies and miR-16 in 27 studies).

FIG. 3.

Frequency of used normalization methods (total 63) by number of publication. Top five widely used normalization methods highlighted by blue color.

Data upload

Any user can submit data to ExcellmiRDB using the “Upload” menu on the webpage. A sample .txt file is available on the database webpage for a large number of extracellular miRNAs and expression pattern information. A JavaScript will automatically upload data from the submitted file. We will examine the uploaded information carefully before final submission.

Discussion

As the number of published articles on extracellular miRNAs increases, ExcellmiRDB could serve as an entry point to access useful information on extracellular miRNAs implicated in human health. For instance, in 2008, there were only 23 articles published, but by October, 2014, more than 1130 articles have been published (PubMed search with term “circulatory microRNA or extracellular microRNA”).

Although many miRNA databases including diseas-oriented databases have been published before, so far ExcellmiRDB is the only one specialized to store information about the miRNA levels in body-fluid samples along with other information that have not been catalogued by other databases.

We have compared our curated information with other related databases (Table 2). The first database that cataloged information about extracellular miRNA was Exocarta (Mathivanan and Simpson, 2009). It manages information about the contents of exosomes, for example, protein, DNA, RNA, and miRNA. miRandola is the first database specifically designed for extracellular miRNAs (Russo et al., 2012). It gives information about extracellular miRNAs, associated diseases, body fluids, classification of extracellular miRNAs, and techniques used. miRandola has information curated from more than 89 articles (at the time of publication). Recently, the Human miRNA Disease Database (HMDD) V 2.0 listed extracellular miRNAs information including expression pattern information (Li et al., 2014). Certain other databases, for example, PhenomiR, miR2disease, provide miRNA-disease relationships without specifying extracellular or cellular miRNAs (Jiang et al., 2009; Ruepp et al., 2010). Our database accommodates information that previously has not been cataloged altogether in a single miRNA-diseases database, such as biofluids, species, ethnic group, normalization methods, sample size, and expression pattern.

Table 2.

Comparison of ExcellmiRDB with Other Databases ^*

Contents	Exocarta	Mirandola	ExcellmiRDB	HMDD V.2
Year of publication	2009	2012	NA	2014
No. of miRNAs	444	602	590	240
No. of Pubmed articles	NA	139	80	167
Type of miRNAs defined	1 type (Exosomes)	4 types	4 types and more	NA
No. of conditions	NA	99	53	76
No. of body fluids/sample	cell studies	19 (including cells cultures)	14	NA
No. of species	4	NA	5	NA
Ethnic groups	NA	NA	7	NA
Sample size	NA	NA	YES	NA
Normalization methods	NA	NA	37	NA
Expression pattern	NA	NA	YES	YES
GEO accession number	NA	NA	YES	NA

information extracted on July, 14, 2014).

We have cataloged such information for several reasons. First, extracellular miRNAs were studied in organisms other than human and mouse, including dog (Mizuno et al., 2011), cow (Li et al., 2012), pig (Gidlöf et al., 2011), and horse (da Silveira et al., 2012). Their intercellular and inter-organ expression pattern and functions may be evolutionarily conserved. Second, there were differences found in expression profiles of miRNAs between human ethnic groups (Zhao et al., 2010). Third, identifying suitable normalization reference genes/miRNAs for analysis of the extracellular miRNAs using RT-PCR or microarray is a current problem and may cause inconsistency in the results (Thomas et al., 2009). Our database could be used to get the most widely used or body-fluid specific normalization methods. Fourth, small sample size has less confident results (Allison et al., 2006). We curated sample size information so a user can get most confident disease-miRNA relationships based on sample size. Additionally, we provided GEO dataset accession numbers so a user may access the raw/processed data for meta-analysis/re-analysis (Edgar et al., 2002; Wirapati et al., 2008).

Altogether, the information from ExcellmiRDB along with other databases will be useful for extracellular miRNAs research. For example, miRNAs functional annotation databases such as miRBase (Griffiths-Jones et al., 2008), miRNAMap 2.0 (Hsu et al., 2008), miRNApath (Chiromatzo et al., 2007) can provide great resources to investigate the functions of extracellular miRNAs in recipient cells. Databases such as microRNA.org (Betel et al., 2008) that provide expression level of miRNAs in tissues could be useful to get information about the tissue specific extracellular miRNAs.

Conclusions

ExcellmiRDB is an innovative and comprehensive online resource about extracellular miRNAs. It offers information about the disease-extracellular miRNAs and extracellular miRNAs-biofluid relationships and curated information. Future directions will be automatized for the extraction of cataloged information using machine learning and test mining algorithms for which our current content of the database can serve as a validation set.

Footnotes

Acknowledgment

Financial support from Council of Scientific and Industrial Research as Junior Research Fellowship to JB and computational facility provided by Bioinformatics Infrastructure Facility at University of Rajasthan is gratefully acknowledged.

Author Disclosure Statement

The authors declare no conflict of interest. No competing financial interests exist.

References

Alexandrov

, Dua

, Hill

, Bhattacharjee

, Zhao

, and Lukiw

. (2012). microRNA (miRNA) speciation in Alzheimer's disease (AD) cerebrospinal fluid (CSF) and extracellular fluid (ECF). Int J Biochem Mol Biol, 3, 365–373.

Allison

, Cui

, Page

, and Sabripour

. (2006). Microarray data analysis: From disarray to consolidation and consensus. Nat Rev Genet, 7, 55–65.

Assenov

, Ramírez

, Schelhorn

, Lengauer

, and Albrecht

. (2008). Computing topological parameters of biological networks. Bioinformatics, 24, 282–284.

Bartel

. (2004). MicroRNAs genomics, biogenesis, mechanism, and function. Cell, 116, 281–297.

Benetti

, Gonzalo

, Jaco

, et al. (2008). A mammalian microRNA cluster controls DNA methylation and telomere recombination via Rbl2-dependent regulation of DNA methyltransferases.Nature Publishing Group. Nat Struct Mol Biol, 15, 268–279.

Betel

, Wilson

, Gabow

, Marks

, and Sander

. (2008). The microRNA.org resource: Targets and expression. Nucleic Acids Res, 36, D149–153.

Bryant

, Pawlowski

, Catto

JWF

, et al. (2012). Changes in circulating microRNA levels associated with prostate cancer. Br J Cancer, 106, 768–774.

Calin

, and Croce

. (2006). MicroRNA signatures in human cancers. Nat Rev Cancer, 6, 857–866.

Chen

, Zhang

, Li

, et al. (2012). A panel of five circulating microRNAs as potential biomarkers for prostate cancer. Prostate, 72, 1443–1452.

10.

Chiromatzo

, Oliveira

TYK

, Pereira

, et al. (2007). miRNApath: A database of miRNAs, target genes and metabolic pathways. Genet Mol Res, 6, 859–865.

11.

Cho

, Jang

, Jun

, et al. (2013). MiRGator v3.0: A microRNA portal for deep sequencing, expression profiling and mRNA targeting. Nucleic Acids Res, 41, D252–257.

12.

Corcoran

, Friel

, Duffy

, Crown

, and O'Driscoll

. (2011). Intracellular and extracellular microRNAs in breast cancer. Clin Chem, 57, 18–32.

13.

Corsten

, Dennert

, Jochems

, et al. (2010). Circulating MicroRNA-208b and MicroRNA-499 reflect myocardial damage in cardiovascular disease. Circ Cardiovasc Genet, 3, 499–506.

14.

Da Silveira

, Veeramachaneni

DNR

, Winger

, Carnevale

, and Bouma

. (2012). Cell-secreted vesicles in equine ovarian follicular fluid contain miRNAs and proteins: A possible new form of cell communication within the ovarian follicle. Biol Reprod, 86, 71.

15.

, Cao

, Zou

, et al. (2013). MicroRNA-21 and risk of severe acute kidney injury and poor outcomes after adult cardiac surgery. PLoS One, 8, e63390.

16.

Edgar

, Domrachev

, and Lash

. (2002). Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res, 30, 207–210.

17.

Farina

, Wood

, Perrapato

, et al. (2014). Standardizing analysis of circulating microRNA: Clinical and biological relevance. J Cell Biochem, 115, 805–811.

18.

Friedländer

, Mackowiak

, Li

, Chen

, and Rajewsky

. (2012). miRDeep2 accurately identifies known and hundreds of novel microRNA genes in seven animal clades. Nucleic Acids Res, 40, 37–52.

19.

, Yi

, Wu

, Li

, and Xu

. (2011). Circulating microRNAs in patients with active pulmonary tuberculosis. J Clin Microbiol, 49, 4246–4251.

20.

Gidlöf

, Andersson

, van der Pals

, Götberg

, and Erlinge

. (2011). Cardiospecific microRNA plasma levels correlate with troponin and cardiac function in patients with ST elevation myocardial infarction, are selectively dependent on renal elimination, and can be detected in urine samples. Cardiology, 118, 217–226.

21.

Griffiths-Jones

, Saini

, van Dongen

, and Enright

. (2008). miRBase: Tools for microRNA genomics. Nucleic Acids Res, 36, D154–158.

22.

Hsu

, Chu

, Tsou

, et al. (2008). miRNAMap 2.0: Genomic maps of microRNAs in metazoan genomes. Nucleic Acids Res, 36, D165–169.

23.

Hsu

, Lin

, Wu

, et al. (2011). miRTarBase: A database curates experimentally validated microRNA-target interactions. Nucleic Acids Res, 39, D163–169.

24.

Jazbutyte

, and Thum

. (2010). MicroRNA-21: From cancer to cardiovascular disease. Curr Drug Targets, 11, 926–935.

25.

Jiang

, Wang

, Hao

, et al. (2009). miR2Disease: A manually curated database for microRNA deregulation in human disease. Nucleic Acids Res, 37, D98–104.

26.

Kanaan

, Rai

, Eichenberger

, et al. (2012). Plasma miR-21: A potential diagnostic marker of colorectal cancer. Ann Surg, 256, 544–551.

27.

Lewis

, Burge

, and Bartel

. (2005). Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell, 120, 15–20.

28.

, Qiu

, Tu

, et al. (2014). HMDD v2.0: A database for experimentally supported human microRNA and disease associations. Nucleic Acids Res, 42, D1070–1074.

29.

, Liu

, Jin

, Lo

, and Liu

. (2012). Expression profiles of microRNAs from lactating and non-lactating bovine mammary glands and identification of miRNA related to lactation. BMC Genomics, 13, 731.

30.

Link

, Balaguer

, Shen

, et al. (2010). Fecal MicroRNAs as novel biomarkers for colon cancer screening. Cancer Epidemiol Biomarkers Prev, 19, 1766–1774.

31.

Liu

, Yao

, Wang

, et al. (2012). Circulating miR-15b and miR-130b in serum as potential markers for detecting hepatocellular carcinoma: A retrospective cohort study. BMJ Open, 2, e000825.

32.

Lopes

, Franz

, Kazi

, Donaldson

, Morris

, and Bader

. (2010). Cytoscape Web: An interactive web-based network browser. Bioinformatics, 26, 2347–2348.

33.

Lo Russo

, Spolveri

, Ciancio

, and Mori

. (2013). Mendeley: An easy way to manage, share, and synchronize papers and citations. Plast Reconstr Surg, 131, 946e–947e.

34.

Mathivanan

, and Simpson

. (2009). ExoCarta: A compendium of exosomal proteins and RNA. Proteomics, 9, 4997–5000.

35.

McDonald

, Milosevic

, Reddi H

, Grebe

, and Algeciras-Schimnich

. (2011). Analysis of circulating microRNA: Preanalytical and analytical challenges. Clin Chem, 57, 833–840.

36.

Meyer

, Pfaffl

, and Ulbrich

. (2010). Normalization strategies for microRNA profiling experiments: A “normal” way to a hidden layer of complexity?. Biotechnol Lett, 32, 1777–1788.

37.

Michael

, Bajracharya

, Yuen

PST

, et al. (2010). Exosomes from human saliva as a source of microRNA biomarkers. Oral Dis, 16, 34–38.

38.

Mitchell

, Parkin

, Kroh

, et al. (2008). Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci USA, 105, 10513–10518.

39.

Mizuno

, Nakamura

, Aoki

, et al. (2011). Identification of muscle-specific microRNAs in serum of muscular dystrophy animal models: promising novel blood-based markers for muscular dystrophy. PLoS One, 6, e18388.

40.

Müller

, Kuiperij

, Claassen

, Küsters

, and Verbeek

. (2014). MicroRNAs in Alzheimer's disease: Differential expression in hippocampus and cell-free cerebrospinal fluid. Neurobiol Aging, 35, 152–158.

41.

Rabinowits

, Gerçel-Taylor

, Day

, Taylor

, and Kloecker

. (2009). Exosomal microRNA: A diagnostic marker for lung cancer. Clin Lung Cancer, 10, 42–46.

42.

Rao

, Benito

, and Fischer

. (2013). MicroRNAs as biomarkers for CNS disease. Front Mol Neurosci, 6, 39.

43.

Ruepp

, Kowarsch

, Schmidl

, et al. (2010). PhenomiR: A knowledgebase for microRNA expression in diseases and biological processes. Genome Biol, 11, R6.

44.

Russo

, Di Bella

, Nigita

, et al. (2012). miRandola: Extracellular circulating microRNAs database. PLoS One, 7, e47786.

45.

Sethupathy

, Corda

, and Hatzigeorgiou

. (2006). TarBase: A comprehensive database of experimentally supported animal microRNA targets. RNA, 12, 192–197.

46.

Starkey Lewis

, Dear

, Platt

, et al. (2011). Circulating microRNAs as potential markers of human drug-induced liver injury. Hepatology, 54, 1767–1776.

47.

Szcześniak

, Deorowicz

, Gapski

, Kaczyński

, and Makalowska

. (2012). miRNEST database: An integrative approach in microRNA search and annotation. Nucleic Acids Res, 40, D198–204.

48.

Szczesniak

, and Makalowska

. (2014). miRNEST 2.0: A database of plant and animal microRNAs. Nucleic Acids Res, 42, D74–77.

49.

Thomas

, Bueno

, Wirapati

, et al. (2009). Impact of normalization on miRNA microarray expression profiling. RNA, 15, 493–501.

50.

Turchinovich

, Weiz

, and Burwinkel

. (2012). Extracellular miRNAs: The mystery of their origin and function. Trends Biochem Sci, 37, 460–465.

51.

Weber

, Baxter

, Zhang

, et al. (2010). The microRNA spectrum in 12 body fluids. Clin Chem, 56, 1733–1741.

52.

Wirapati

, Sotiriou

, Kunkel

, et al. (2008). Meta-analysis of gene expression profiles in breast cancer: Toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Res, 10, R65.

53.

Yang

, Shao

, Zhou

, Chen

, and Qu

. (2010). deepBase: A database for deeply annotating and mining deep sequencing data. Nucleic Acids Res, 38, D123–130.

54.

Yang

, Xiao

, Li

, Kanwar

, Liu

, and Sun

. (2013). Urine miRNAs: Potential biomarkers for monitoring progression of early stages of diabetic nephropathy. Med Hypotheses, 81, 274–278.

55.

, Fu

, Ji

, Li

, and Guan

. (2012). Altered microRNA signatures in sputum of patients with active pulmonary tuberculosis. PLoS One, 7, e43184.

56.

Zhao

, Shen

, Medico

, Wang

, Ambrosone

, and Liu

. (2010). A pilot study of circulating miRNAs as potential biomarkers of early stage breast cancer. PLoS One, 5, e13735.