microRNA annotation

Since January 2015 we have been using the Gene Ontology (GO) to describe the biological roles of microRNAs (miRNAs). To date we have provided all manual human RNA GO annotations.

The RNA annotations are available in QuickGO, miRBase, NCBIGene, RNAcentral and AmiGO2, or in the regular GOA release files. In addition, our miRNA:target mRNA binding annotations are accessible for inclusion in network analyses from the PSICQUIC web service, in the EBI-GOA-miRNA file. Information about the data included in the PSICQUIC web service is also available.

Example: View the annotations associated with hsa-miR-133a-3p

miRNA curation guidelines

As part of this project we have drawn up miRNA curation guidelines (Huntley et al., 2016) in consultation with the Gene Ontology (GO) Consortium and experts in miRNA research, which cover how a curator should translate into GO annotations the common experimental assays used to investigate miRNA function, including how to annotate experimentally verified targets of miRNAs as well as the physiological effects that silencing has on the cell or organism. Note that GO Consortium guidelines have changed since the publication of the miRNA curation guidelines (Huntley et al., 2016). For more specific details see the Molecular interaction data set for miRNAs and their targets section below

Any queries about the miRNA project should be directed to r.lovering@ucl.ac.uk.

Molecular interaction dataset for miRNAs and their targets

The miRNA:mRNA interaction file is available: https://www.ebi.ac.uk/QuickGO/psicquic-rna/webservices/current/search/interactor/* and via the PSCIQUIC web service. The data set is in PSI-MITAB 2.7 format, which is described at https://psicquic.github.io/MITAB27Format.html.

 

This file contains the miRNA (RNAcentral ID) and the mRNA (UniProt ID) that it has been experimentally shown to bind and/or regulate.

In this file the interactions are indicated as either:

1. 'physical association', meaning the miRNA has been shown experimentally to bind and regulate the mRNA (usually a luciferase assay)
2. 'association', meaning the miRNA has been shown experimentally to regulate expression of the mRNA (e.g. by western blot or RT-PCR), but there was no experiment to demonstrate binding

EBI-GOA-miRNA curation details

Huntley, et al (2018) provides a description of how the PSI-MITAB format interactions are derived from the GO annotations:

A molecular interaction dataset (“EBI-GOA-miRNA”) was created in PSI-MI format and made available on the PSICQUIC web service, to enable computational access to miRNA interactions with their experimentally validated targets. The source of the information in this data set is GO annotations that we have created containing experimentally verified miRNA:target interaction data. GO annotations used for this purpose conform to the following criteria; the “Database Object ID” field of the GO annotation file must be a miRNA, specified by an RNAcentral ID, AND the “Annotation Extension” field must contain an mRNA target, specified by a UniProt ID (del-Toro et al. 2013; Huntley et al. 2014). For the PSICQUIC specification, those interactions described with the Molecular Function GO term mRNA binding involved in post-transcriptional gene silencing (GO:1903231) are assigned interaction type “physical association,” indicating direct binding of the miRNA to the mRNA target. Interactions described only with one of the following GO Biological Process terms: gene silencing by miRNA (GO:0035195); miRNA mediated inhibition of translation (GO:0035278); mRNA cleavage involved in gene silencing by miRNA (GO:0035279); deadenylation involved in gene silencing by miRNA (GO:0098806), but without the Molecular Function term above are assigned the interaction type “association,” indicating that the evidence demonstrated miRNA regulation of the target only (see “Data set availability” section below for the file format information and access to this data set).

miRNA annotation projects undertaken

Alzheimer's Research UK grant ARUK-NSG2018A-003

The ARUK funded the creation of over 1,000 GO annotations which have facilitated the detailed and high-quality annotation of 160 Alzheimer's-relevant microRNAs. A variety of focused projects were undertaken including the curation of:

• microRNAs that regulate microglial proteins
  • The goal of this project was to annotate the microRNAs (miRNAs) that regulated the expression of the proteins with a role in microglial processes. Over 100 human miRNAs were annotated by Rachael Huntley and Barbara Kramarz, including hsa-miR-520b-3p, which we have annotated as regulating the expression of C3 and as positive regulation of complement activation, alternative pathway. In addition, 19 papers describing the experimentally verified mRNA targets of IL6 were curated, creating 90 annotations, and annotations describing 'neuron projection development' and 'gliogenesis' were contributed, yielding 41 annotations from 19 papers and 57 annotations from 13 papers, respectively.
• Alzheimer's Disease microRNA biomarkers
  • Swarbrick S, et al., 2019, decribed a panel of 10 microRNAs suggested to be disregulated early in Alzheimer's disease. Barbara Kramarz curated 14 papers describing the microRNA target interactions for these microRNAs, when available, creating 54 annotations for 10 human microRNAs, including the potential biomarkers hsa-miR-146a-5p, hsa-miR-107, hsa-miR-26b-5p and hsa-miR-210-3p, and 9 proteins.

British Heart Foundation (BHF) grant RG/13/5/30112

The BHF funded the creation of over 3,600 GO annotations which have facilitated the detailed and high-quality annotation of almost 400 cardiovascular-relevant microRNAs. A variety of focused projects were undertaken including the curation of:

• BHF-funded research papers
  • From 2014 to 2018, Rachael Huntley curated all BHF funded research papers describing the role of microRNAs in cardiovascular-relevant processes.
• Regulation of cardiac physiology
  • Rachael Huntley curated 7 microRNAs, as these had been identified as modulating cardiac excitability; literature available for these miRNAs was curated up to March 2016.
• Angiogenesis
  • Over a period of 3 months Vanessa Acquaah curated the role of microRNAs in angiogenesis. The two major cell types are involved in this process: endothelial and smooth muscle cells (SMC). During this project Vanessa annotated 114 papers that demonstrate the roles that miRNAs play in regulating endothelial and SMC cell proliferation, migration and endothelial tube formation. These papers were identified by searches in pubmed using the key terms “mirnas in angiogenesis”. GO annotations associated with a selection of these curated papers are available in QuickGO.
• Pulmonary artery remodelling
  • Pulmonary hypertension is caused by impaired pulmonary arterial growth and remodelling. Vanessa Acquaah annotated 24 papers that describe the role of miRNAs in vascular smooth muscle cell pulmonary artery remodelling, creating 167 annotations.  These papers were identified by searches in pubmed using the key words “miRNAs and pulmonary smooth muscle cells”.

Familial hypercholesterolemia, MSc project

Hao Chen, MSc project 2017. Familial hypercholesterolemia (FH) is a common inherited lipid disorder that can cause cardiovascular diseases (CVD) if untreated. Treatment on FH depend on management of plasma low-density-lipoprotein (LDL) cholesterol. High level of plasma LDL cholesterol can induce atherosclerosis and plaque buildup in the walls of the arteries. There is evidence that suggest miRNAs’ regulation of LDL-C is due to their regulation of FH-associated genes, such as LDLR, PCSK9, APOB, LDLRAP1, ABCA1 and ABCG1. This project created 290 GO annotations, of which 22 were transferred from mouse miRNAs to human miRNAs using the ISS evidence code. The prioritised mRNA targets are listed below:

#	HGNC Symbol	HGNC Name	UniProt ID
1	ABCA1	ATP binding cassette subfamily A member 1	O95477
2	ABCG1	ATP binding cassette subfamily G member 1	P45844
3	APOB	apolipoprotein B	P04114
4	LDLR	low density lipoprotein receptor	P01130
5	LDLRAP1	low density lipoprotein receptor adaptor protein 1	Q5SW96
6	PCSK9	proprotein convertase subtilisin/kexin type 9	Q8NBP7

Angiogenesis, IBSc project

Marios Makris, IBSc project 2019. Angiogenesis is the formation of blood vessels from pre-existing vasculature, involving endothelial cell migration, proliferation, survival, vascular tube formation and regulation of angiogenic growth factors. A total of 237 GO annotations were created to associate 40 distinct microRNAs to 11 pro-angiogenic growth factors. The downstream angiogenesis-relevant effects of regulating these factors were also captured. The prioritised mRNA targets are listed below:

No.	Approved gene name	Approved Gene Symbol
1	Vascular endothelial growth factor A	VEGFA
2	Kinase insert domain receptor*	KDR*
3	Fibroblast growth factor 2	FGF2
4	Fibroblast growth factor receptor 1	FGFR1
5	Fibroblast growth factor receptor 2	FGFR2

Aortic aneurysm, MSc project

Zara Umrao, MSc project 2015. Aortic aneurysm (AA) is a cardiovascular disease characterized by swelling of the aorta, the main vessel that runs from the heart to the abdomen. Recently, research has revealed the role of microRNAs (miRNA) in some diseases, including AA, which may enable a better understanding of the disease. The role of one such family of miRNAs, miRNA29, was investigated by creating annotations from existing literature. Forty papers were curated, from which 497 GO annotations were made. The annotations spanned a range of species, including human, mouse, rat, zebrafish and dog. The role of miRNA29 in regulating essential extracellular matrix proteins, such as elastin, collagen, fibrillin and matrix metalloproteinases was captured. In addition, the GO annotations also described the role of miRNA29s in activities, such as aorta morphogenesis, cell migration, cell proliferation, DNA methylation and epithelium regeneration.

Early heart development, MSc project

Vanessa Acquaah, MSc project 2016. Recent evidence has shown that miRNAs, in particular microRNA-1 (miR-1) are found in abundance within cardiac tissues and may be particularly important in early heart development. A total of 313 annotations were made through the use of experimental evidence from 49 papers using GOC curation guidelines. Annotations covered the role of miR-1 as well as other miRNAs in cardiomyocyte proliferation and cardiomyocyte differentiation. Additionally, development of cardiac neural crest cells through processes such as endothelial to mesenchymal transition (EMT) and differentiation of epithelial cells are processes regulated by miRNAs; these roles have been captured as GO annotations.

Regulation of amyloid-beta 'good' receptors, MSc project

Shirin Saverimuttu, MSc project 2018. Alzheimer’s disease (AD), the most common form of dementia, is characterised by amyloid beta (Aβ) accumulation in extracellular plaques. Moreover, Aβ accumulation has been hypothesised as a primary cause of the disease, with a dysregulation to the proteins involved in Aβ clearance thought to be critical. Interestingly, hundreds of microRNAs have been found to be altered in brain tissue affected by AD and several have been experimentally proven to regulate the expression of Aβ clearance proteins. This MSc project led to a total of 258 GO annotations being created. This allowed for knowledge describing how microRNAs regulate Aβ clearance proteins to be captured, as well as the capturing the downstream effects of these microRNAs. The prioritised mRNA targets are listed below:

#	HGNC Symbol	HGNC Name	UniProt ID
1	APOE	Apolipoprotein E	P02649
2	CHRNA7*	Cholinergic receptor nicotinic alpha 7 subunit	P36544
3	CLU	Clusterin	P10909
4	FPR2	Formyl peptide receptor 2	P25090
5	HSPG2	Heparan sulfate proteoglycan 2	P98160
6	ITGA2*	Integrin subunit alpha 2	P17301
7	ITGAM*	Integrin subunit alpha M	P11215
8	LRP1	LDL receptor related protein 1	Q07954
9	LRP2	LDL receptor related protein 2	P98164
10	LDLR	Low density lipoprotein receptor	Q07954
11	MARCO*	Macrophage receptor with collagenous structure	Q9UEW3
12	MSR1*	Macrophage scavenger receptor 1	P21757
13	PICALM*	Phosphatidylinositol binding clatherin assembly protein	Q13492
14	PRNP	Prion protein	P04156
15	TLR2	Toll like receptor 2	O60603
16	TLR4	Toll like receptor 4	O00206

* there was no experimental evidence to support annotation of microRNAs regulating these proteins in July 2019

Regulation of interleukins

Shirin Saverimuttu is responsible for this project, the goal of which is to annotate microRNAs (miRNAs) that regulate the expression of interleukins with a role in Alzheimer’s-relevant processes. In order to achieve this goal, it was necessary to generating a priority list of interleukins (below) and then to identify their regulatory miRNAs. The miRNAs are identified on a target-by-target basis, firstly using mirTarBase and then with specific searches in PubMed. This project is funded from Functional Gene Annotation group’s discretionary funds. The prioritised mRNA targets are listed below:

#	HGNC Symbol	HGNC Name	UniProt ID
1	IL1A	Interleukin 1 alpha	P01583
2	IL1B	Interleukin 1 beta	P01584
3	IL1F10	Interleukin 1 family member 10	Q8WWZ1
4	IL2	Interleukin 2	P60568
5	IL3	Interleukin 3	P08700
6	IL4	Interleukin 4	P05112
7	IL5	Interleukin 5	P05113
8	IL6	Interleukin 6	P05231
9	IL7	Interleukin 7	P13232
10	CXCL8	C-X-C motif chemokine ligand 8 (alias: interleukin 8)	P10145
11	IL9	Interleukin 9	P15248
12	IL10	Interleukin 10	P22301
13	IL11	Interleukin 11	P20809
14	IL12A	Interleukin 12A	P29459
15	IL12B	Interleukin 12B	P29460
16	IL13	Interleukin 13	P35225
17	IL15	Interleukin 15	P40933
18	IL16	Interleukin 16	Q14005
19	IL17A	Interleukin 17A	Q16552
20	IL17B	Interleukin 17B	Q9UHF5
21	IL17C	Interleukin 17C	Q9P0M4
22	IL17D	Interleukin 17D	Q8TAD2
23	IL17F	Interleukin 17F	Q96PD4
24	IL18	Interleukin 18	Q14116
25	IL19	Interleukin 19	Q9UHD0
26	IL20	Interleukin 20	Q9NYY1
27	IL21	Interleukin 21	Q9HBE4
28	IL22	Interleukin 22	Q9GZX6
29	IL23A	Interleukin 23 alpha	Q9NPF7
30	IL24	Interleukin 24	Q13007
31	IL25	Interleukin 25	P01589
32	IL26	Interleukin 26	Q9NPH9
33	IL27	Interleukin 27	Q8NEV9
34	IL31	Interleukin 31	Q6EBC2
35	IL32	Interleukin 32	P24001
36	IL33	Interleukin 33	O95760
37	IL34	Interleukin 34	Q6ZMJ4
38	IL36A	Interleukin 36 alpha	Q9UHA7
39	IL36B	Interleukin 36 beta	Q9NZH7
40	IL36G	Interleukin 36 gamma	Q9NZH8
41	IL37	Interleukin 37	Q9NZH6

Endothelial connections at the Blood Brain Barrier

Katherine Thurlow, MSc project 2019. Evidence for cerebrovascular mechanisms, particularly increased breakdown and dysfunction of the blood-brain barrier (BBB), as key contributors to Alzheimer’s Disease (AD), the most common form of dementia, has grown in recent years,. The BBB is key in regulating influx and efflux of substances to the brain and as such maintains normal neuronal and cognitive function. Alterations in BBB constituent proteins could alter its ability to selectively transport molecules and result in a change in permeability to potentially toxic substances. The list of priority mRNA targets for this project was identified from Figure 2 of Sweeney et al., 2019. This MSc project led to a total of 253 GO annotations being created. This allowed for knowledge describing the role of microRNAs at the BBB to be captured. The prioritised mRNA targets are listed below:

#	UniProt ID	HGNC Symbol	HGNC Name
1	P60709	ACTB	actin beta
2	P63261	ACTG1	actin gamma 1
3	P33151	CDH5	cadherin 5
4	O95832	CLDN1	claudin 1
5	P56749	CLDN12	claudin 12
6	O15551	CLDN3	claudin 3
7	O00501	CLDN5	claudin 5
8	P11532	DMD	dystrophin
9	Q96AP7	ESAM	endothelial cell adhesion molecule
10	Q9Y624	F11R	F11 receptor
11	P17302	GJA1	gap junction protein alpha 1
12	O95452	GJB6	gap junction protein beta 6
13	P05556	ITGB1	Integrin subunit beta 1
14	P57087	JAM2	junctional adhesion molecule 2
15	Q9BX67	JAM3	junctional adhesion molecule 3
16	P24043	LAMA2	Laminin subunit alpha-2
17	P11047	LAMC1	Laminin subunit gamma-1
18	Q86X29	LSR	lipolysis stimulated lipoprotein receptor
19	Q16625	OCLN	occludin
20	P16284	PECAM1	platelet and endothelial cell adhesion molecule 1
21	Q07157	TJP1	tight junction protein 1
22	Q9UDY2	TJP2	tight junction protein 2
23	O95049	TJP3	tight junction protein 3
24	P18206	VCL	vinculin

Regulation of key proteins involved in APP processing and Aβ formation

Sandra De Miranda Pinheiro, MSc project 2019. Accumulation of amyloid-beta (Aβ) and tau proteins in the brain is the hallmark of Alzheimer’s disease and lead to neurodegeneration. An imbalance between the production and the clearance of Aβ from the brain is thought to drive the development of the disease. Aβ derives from a larger protein, the amyloid precursor protein (APP), which is processed by different secretases in a sequential way. In the amyloidogenic pathway, β-secretase cleaves APP at the β-site and cleavage of the remaining APP fragment by γ-secretase results in the production of Aβ peptides. Increased APP expression has been associated with an increased risk of developing Alzheimer’s disease. This MSc project led to a total of 254 GO annotations being created. This allowed for knowledge describing how microRNAs regulate the processing of APP and the formation of Aβ. The prioritised mRNA targets are listed below:

#	HGNC Gene Symbol	HGNC Approved Name	Protein Group
1	APP	amyloid beta precursor protein
2	ADAM9	ADAM metallopeptidase domain 9	Alpha-secretase
3	ADAM10	ADAM metallopeptidase domain 10	Alpha-secretase
4	ADAM17	ADAM metallopeptidase domain 17	Alpha-secretase
5	ADAM19	ADAM metallopeptidase domain 19	Alpha-secretase
6	APH1A	aph-1 homolog A	Gamma-secretase subunit
7	APH1B	aph-1 homolog B	Gamma-secretase subunit
8	BACE1	beta-secretase 1	Beta-Secretase 1
9	NCSTN	nicastrin	Gamma-secretase
10	PSEN1	presenilin 1	Gamma-secretase
11	PSEN2	presenilin 2	Gamma-secretase
12	PSENEN	presenilin enhancer	Gamma-secretase subunit

 

Regulation of Aβ-binding 'bad' receptors

Diana Luna Buitrago, MSc project 2020. Amyloid-beta (Aβ) accumulation within the brain and abnormal processing of Aβ by various Aβ-binding receptors are known to be characteristics of Alzheimer's disease pathology. Through previous studies, Aβ-binding receptors have been identified to be “good” or “bad”. Here, “bad” receptors are those that bind to Aβ to promote downstream neurotoxic processes associated with AD. MicroRNAs may contribute to AD by acting as regulators of Aβ-binding receptor expression, as microRNAs that alter “bad” Aβ-binding receptor levels and may influence the neurotoxicity of Aβ. The table below lists the microRNAs curated to regulate the expression of known “bad” Aβ-binding receptors during this project:

MicroRNAs identified to regulate key proteins of interest
AMPA receptor subunits				Ephrin receptors		INSR
GRIA1*	GRIA2*	GRIA3*	GRIA4*	EPHA4	EPHB2
hsa-miR-124-3p	hsa-miR-124-3p hsa-miR-181b-5p hsa-miR-181a-5p	hsa-miR-124-3p	hsa-miR-28-5p	hsa-miR-10a-5p hsa-miR-519d-3p hsa-miR-335-5p hsa-miR-145-5p	hsa-miR-185-5p hsa-miR-520d-3p hsa-miR-204-5p	hsa-miR-195-5p hsa-miR-7-5p hsa-miR-1271-5p hsa-miR-15b-5p
Frizzled receptors		AGER	SCARB2	PGRMC1	NGFR
FZD5	FZD4
hsa-miR-29c-3p hsa-miR-515-5p hsa-miR-224-5p	hsa-miR-493-3p hsa-miR-199a-5p hsa-miR-29c-3p	hsa-miR-185-5p	hsa-miR-127-5p	hsa-miR-98-5p	hsa-miR-296-5p

Regulation of tau-associated processes

Miao Long, MSc project 2020. Alzheimer's disease (AD) is characterised neuropathologically as the presence of intraneuronal neurofibrillary lesions expressing the Tau protein. One of the characteristic pathologies of sporadic AD patients, is hyperphosphorylation of Tau and neurofibrillary tangles. Remarkably, hundreds of microRNAs have been shown to be changed in AD-affected brain regions, with some of them being experimentally demonstrated to regulate tau-related activities. This project focused on capturing the role of microRNAs in regulation of proteins involved in Tau hyperphosphorylation. The mRNA targets identified in this project are listed below:

#	UniProt ID	HGNC Symbol	HGNC Name
1	P35222	CTNNB1	catenin beta 1
2	O94907	DKK1	dickkopf WNT signaling pathway inhibitor 1
3	P49841	GSK3B	glycogen synthase kinase 3 beta
4	P10636	MAPT	microtubule associated protein tau
5	Q9P0L2	MARK1	microtubule affinity regulating kinase 1
6	P27448	MARK3	microtubule affinity regulating kinase 3
7	Q96L34	MARK4	microtubule affinity regulating kinase 4
8	Q969G9	NKD1	NKD inhibitor of WNT signaling pathway 1
9	P67775	PPP2CA	protein phosphatase 2 catalytic subunit alpha
10	P62714	PPP2CB	protein phosphatase 2 catalytic subunit beta
11	Q13362	PPP2R5C	protein phosphatase 2 regulatory subunit B'gamma
12	P10644	PRKAR1A	protein kinase cAMP-dependent type I regulatory subunit alpha
13	Q04206	RELA	RELA proto-oncogene, NF-kB subunit
14	Q13464	ROCK1	Rho associated coiled-coil containing protein kinase 1
15	P01375	TNF	tumor necrosis factor
16	Q5TCY1	TTBK1	tau tubulin kinase 1