Phenotypic and genetic characterization of a family carrying two Xq21.1-21.3 interstitial deletions associated with syndromic hearing loss
- Sandra Iossa1, 2,
- Valerio Costa3,
- Virginia Corvino4,
- Gennaro Auletta4,
- Luigi Barruffo4,
- Stefania Cappellani5,
- Carlo Ceglia2,
- Giovanni Cennamo6,
- Adamo Pio D’Adamo5, 7,
- Alessandra D’Amico8,
- Nilde Di Paolo8,
- Raimondo Forte9,
- Paolo Gasparini5, 7,
- Carla Laria4,
- Barbara Lombardo1, 2,
- Rita Malesci4,
- Andrea Vitale10,
- Elio Marciano4 and
- Annamaria Franzè2, 4Email author
© Iossa et al.; licensee BioMed Central. 2015
Received: 29 September 2014
Accepted: 17 February 2015
Published: 20 March 2015
Sensorineural hearing impairment is a common pathological manifestation in patients affected by X-linked intellectual disability. A few cases of interstitial deletions at Xq21 with several different phenotypic characteristics have been described, but to date, a complete molecular characterization of the deletions harboring disease-causing genes is still missing. Thus, the aim of this study is to realize a detailed clinical and molecular analysis of a family affected by syndromic X-linked hearing loss with intellectual disability.
Clinical analyses revealed a very complex phenotype that included inner ear malformations, vestibular problems, choroideremia and hypotonia with a peculiar pattern of phenotypic variability. Genomic analysis revealed, for the first time, the presence of two close interstitial deletions in the Xq21.1-21.3, harboring 11 protein coding, 9 non-coding genes and 19 pseudogenes. Among these, 3 protein coding genes have already been associated with X-linked hearing loss, intellectual disability and choroideremia.
In this study we highlighted the presence of peculiar genotypic and phenotypic details in a family affected by syndromic X-linked hearing loss with intellectual disability. We identified two, previously unreported, Xq21.1-21.3 interstitial deletions. The two rearrangements, containing several genes, segregate with the clinical features, suggesting their role in the pathogenicity. However, not all the observed phenotypic features can be clearly associated with the known genes thus, further study is necessary to determine regions involved.
Sensorineural hearing impairment (SNHI) can be considered the most frequent form of children’s hearing loss (1/1000 born) and about 60% of the cases can be ascribed to genetic causes . Most of SNHI forms are non-syndromic, representing 70% of all cases, even though at least 400 known syndromes have been described. Intellectual disability (ID) is a disorder that sometimes co-occurs with sensorineural hearing impairment. It represents one of the major handicaps affecting 1% to 3% of the general population, with onset before age 18. It is more common in males because of the high incidence of mutations in genes located on the X chromosome . To date, more than 80 chromosome X-linked ID genes (XLID) have been identified . In most of them mutations give rise to syndromic ID forms, often accompanied by behavioral, somatic, metabolic or neurological disorders [4,5]. Syndromic forms can also result from contiguous gene deletions. In particular, deletions and rearrangements in the Xq21 have been linked to other complex syndromes including ID, deafness, choroideremia (CHM), seizures, and multiple congenital anomalies [6-11]. Smaller deletions lead to a milder clinical phenotype [9,10,12-16]. ID has been reported - in few cases - in association with deafness and CHM [17,18]. In these cases, the patients carried interstitial deletions within the Xq21. Schwartz et al.  mapped the deletion identified by Rosenberg and collaborators  at the Xq21.2-Xq21.31. More recently, a new 16 Mb deletion in the same region has been described . Nonetheless, a complete molecular characterization of such deletions at the Xq21, harboring disease-causing genes, is still missing. In this study, we describe for the first time - to the best of our knowledge - a systematic clinical, genomic and molecular analysis of a family with a complex phenotype including syndromic congenital sensorineural X-linked hearing loss. Our analysis revealed the presence of two close interstitial deletions in the Xq21.1-21.3, harboring several protein-coding genes. We also discuss the potential role of the deletions in the described phenotype.
This paper is dedicated to the memory of the Pediatrician Dr. Alfredo Pisacane (1951–2012).
Clinical details for affected subjects
- Vestibular problems
Mixed sufferance: central and peripheral
- Age of hearing aids fitting
- Constant use of hearing aids
Very poor: recognition only of the use and function of simple and familiar objects. Psychotic behavioral and relational disorder
Poor: recognition only of the use and function of simple and familiar objects
- Verbal language
Absent: use of a mimic sign language
Concrete and fluent
- Trunk and limbs hypotonicity
Present, but milder than IV:1
- Motor stereotypies
- CHM evaluation
Alteration typical of choroideremia
- Brain MRI/CT
Mega cisterna ampia. Cerebral cysticercosis
- Ear MRI/CT
Inner ear strongly altered. Small oval window
Inner ear strongly altered. Small oval window
Inner ear strongly altered. Dysplastic fine irregularity of the profile of the semicircular canals
- Large forehead, down turned rimae palpebrarum and large ears
- Flat feet
Subject III-1: Born in 1969; he was diagnosed in infancy with bilateral profound SNHL and psycho-neuro-motor disability. The intellectual disability is of medium severity: he is sufficiently able to perform activities of daily living on his own. Neuropsychological tests determined a deficit that primarily occurs in short term memory for spatial material. He had been fitted with hearing aids for language development at nine years of age, and presents a simple concrete verbal language with a good verbal comprehension. Also this subject presents balance problems and flat feet but motor stereotypies are absent. Subject II-2: Born in 1947; he was diagnosed with profound bilateral hearing loss and very severe developmental delay and intellectual disability. He was not able to perform his activities of daily living unaided. Detailed clinical examinations had never been performed.
No unaffected individuals or carrier females had mental, auditory or other significant problems related to those described in the affected individuals.
Prior to this study, no significant visual disturbance had been noticed in any of the family subjects. Therefore, as also the CHM gene - involved in choroideremia onset - was deleted, we carried out a detailed ophthalmological study on individuals III-1 and III-2. Clinical alterations typical of a CHM affected individual were observed for patient III-1 (Figure 2, panel 3), whereas a mild phenotype, typical of carrier individuals was reported in patient III-2 (Figure 2, panel 4).
Sensorineural congenital hearing loss, vestibular problems, intellectual disability, choroideremia, hypotonia and some peculiar facial dysmorphisms have been observed in the family members (Table 1). Few studies have reported some of the phenotypic characteristics (intellectual disability, hearing loss and choroideremia) here described [17,20]. However, a comprehensive clinical, genomic and molecular characterization has not yet been performed. It is particularly relevant to observe the phenotypic variability in the family subjects reported in this study. In particular, the large difference in the degree of intellectual disability, which is more pronounced for proband (IV-1) compared to his brother (IV-2) and above all to their uncle (III-1). Indeed, even though individual III-1 did not undergo any treatment during his infancy he had a very milder phenotype compared to his relatives. Some differences among affected subjects have also been observed in the conformation of the inner ear and brain (Table 1).
From a molecular point of view this is the first report of two close deletions in the Xq21.1-21.3 regions: Del I and Del II (Figure 3D). This genomic region encompasses several genes, reported in detail in Figure 3C and Additional file 1: Table S1_1 (see Supplementary data online S1.doc, section 1). Three of them - POU3F4, ZNF711 and CHM - are particularly significant for the clinical symptoms described in the affected individuals. POU3F4 has a proven role in hearing loss and inner ear malformations , ZNF711 has been associated with simple ID  and CHM with the choroideremia . In the analyzed family, choroideremia symptoms are not particularly significant. Indeed, CHM has not been previously diagnosed in family members. CHM alterations have been determined only by instrumental ophthalmic exams (Table 1; Figure 2, panel 3 and 4). Our study also documented other peculiar phenotypic features, such as hypotonia and facial dysmorphism, in the family members. Interestingly, none of the genes present in the deletion described is clearly associated with these symptoms, as shown in Additional file 1: Table S1_1 (see Supplementary data online S1.doc, section 1).
However, although the other genes have not been associated with any of the observed pathological traits, we cannot exclude that the encoded proteins may interact with known disease-causing genes. Interestingly, browsing BioGRID - an online interaction repository - we observed that the protein encoded by RPS6KA6 gene has more than 35 interactors, with a significant involvement in “central nervous system development” and “negative regulation of mesoderm development”. This observation suggests that some deleted genes may potentially encode proteins that interact, both directly and indirectly, with other proteins whose function is strictly correlated with the observed phenotypes. Other protein coding genes, not mapping to the deleted X chromosome region, may account for the unexpected pathological features. It indicates that further genome-wide analyses are needed to identify them. Moreover, our results highlight the presence of pseudogenes and of other non-coding genes in the deleted region. Notably, both pseudogenes and ncRNAs have been implicated in a novel regulatory networking (acting as ceRNA) and have been recently been indicated as relevant in neural cell plasticity and in neurodegenerative processes [25-28]. Therefore, we cannot exclude that the deletion of some processed pseudogenes, long ncRNAs (lncRNAs) and microRNAs (miRNAs) (Figure 3, panel F), may affect the expression of other genes, located elsewhere in the genome. Large-scale gene expression studies in these patients (by RNA-Sequencing) will clarify this crucial point. Del II deletion does not contain regulatory or coding sequences. This region may harbor a non-pathogenic copy number variation (CNV), which may have existed before the larger deletion arose. However, browsing of public CNV databases  excluded the presence of rearrangements encompassing the small deletions. However, the presence of still uncharacterized regions - with relevant coding or regulatory functions - in Del II cannot be excluded.
In this study we identified two, previously unreported, Xq21.1-21.3 interstitial deletions in a family with a syndromic form of hearing loss with variable phenotype. The two rearrangements, segregate with the clinical features in affected subjects, suggesting their role in the pathogenicity. Not all the observed phenotypic features can be clearly associated to known genes. The presence of several pseudogenes and non-coding genes within the deleted region may correlate to the variable phenotype observed, and potentially affect unexplored molecular pathways or involve yet underestimated epigenetic mechanisms. So, further studies are necessary to obtain a more detailed genotype-phenotype correlation.
Data herein described confirm that this region of the X chromosome is characterized by increased proneness to chromosome breakage.
Affected individuals IV-1, IV-2 and III-1 were analyzed by CT scan of ears and brain without contrast medium and by MRI of brain and ears (Philips 1.5 T). Gadolinium injection was subsequently required because of a suspected associated brain cysticercosis in individual III-1. Vestibular examination was performed with videonystagmoscopy (VNS) and included testing for spontaneous and positional nystagmus (head-hanging, left and right lateral supine positions) with and without visual fixation; testing for VOR generated by passive head rotation in the horizontal plane; VOR suppression test. Ophthalmic examination included best-corrected visual acuity (BCVA) by Snellen chart, intraocular pressure tonometry, anterior segment examination and ophthalmoscopy. Combined spectral domain optical coherence tomography/microperimetry system (SD-OCT, OCT/SLO, OPKO-OTI, Miami, FL) was used to obtain fundus-related microperimetry [30,31]. For fundus-related microperimetry, we used the Polar 3 testing pattern, a stimulus size Goldmann III test spot (Ø = 26 arc/min), stimulus intensity between 16 and 0 dB (10–400 asb). The RTVue-100 SD-OCT device (Optovue, Inc., Fremont, CA, USA) was used to obtain cross sectional B-scans of the posterior pole, retinal thickness in the 3 mm diameter central ring of the ETDRS map (central macular thickness, CMT), thickness of the retinal ganglion cell complex (GCC) and retinal nerve fiber layer (RNFL). RTVue-100 incorporates a GCC scan mode to measure inner macular retinal layer thickness, consisting of retinal nerve fiber layer, ganglion cell layer and inner plexiform layer, centered on the fovea and covering the central macula. For the peripapillary RNFL measurements, the RNFL exam protocol was used for scan acquisition, which was measured automatically at a diameter of 3.45 mm around the center of the optic nerve. For mutational screening genomic DNA was extracted by conventional salt precipitation protocols from peripheral blood samples collected in EDTA-containing tubes or from buccal swabs (Gentra System, DNA Isolation Kit). Amplified PCR fragments of the exon 1 and 2 of the GJB2 gene were directly sequenced as reported elsewhere . The two hearing loss-related deletions of the GJB6 gene, del (GJB6-D13S1830) and del (GJB6-D13S1854), were screened using a combined detection procedure . For Charcout-Marie-Tooth genes (PRPS1, GJB1) see Supplementary data online S1.doc, section 3). SNPs analysis was performed on the genomic DNA isolated from the patient and his relatives. All samples were genotyped using the Illumina HumanCytoSNP-12 BeadChip using the standard protocol. All samples were then checked for mendelian errors using Pedcheck and Pedstat . Linkage analysis was performed using an X-linked parametric model with minx and complete penetrance . High resolution a-CGH analysis was performed on the genomic DNA of the patient and his relatives. DNA digestion, labeling and hybridization were performed according to the manufacturer’s protocols. DNA specimens were analyzed with the Human Genome CGH Microarray kit 4x180K (Agilent Technologies, Santa Clara, CA), with an average space of 13 Kb and allowing an average resolution of 25 Kb. The proband’s genomic DNA was further analyzed with the Human Genome CGH Microarray kit 1xM (Agilent Technologies, Santa Clara, CA), with an average space of 2,1 Kb. Microarrays were scanned on the Agilent G2565CA scanner and image files were quantified using Agilent’s Feature Extraction software (V10.10.1.1); data were visualized with Agilent’s Genomic Work Bench Standard Edition (V220.127.116.11). Primer pairs for PCR analyses were designed using the Primer3 software . The primer pairs obtained were checked for non-specific hybridizations in other genome regions using the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST) . The different amplified fragments have a variable length due to the need to not amplify regions of the X chromosome frequently repeated in the entire genome, in particular, given the high presence of repeated sequences dispersed in the genomic region analyzed, we took advantage of the RepeatMask track of the USCS Genome Browser  to design oligonucleotide pairs to validate the deletions. Primer sequences and PCR conditions used in this study are reported in supplementary data online S1.doc, section 2). Total RNA was isolated from PBMCs. PBMCs were separated by Ficoll gradient centrifugation, placed in Trizol (Invitrogen), and stored at −80°C according to the manufacturer’s. RNA integrity and concentration were evaluated by using the Nanodrop (Thermo Scientific). The quality of RNA from each sample was considered good if the 260/280 ratio was in between 1.8 and 2.0.
RT-PCR reaction was carried out by reverse transcription of total RNA using High Capacity cDNA Reverse Transcription Kit (Life Technologies). Primer sequences and PCR conditions on cDNA are reported in supplementary data online S1.doc, section 4). An extensive search for the presence of transcripts mapping within the genomic region encompassing the deletions has been performed retrieving the main genomic databases. Gene annotation tracks of RefSeq (release 62), Ensembl (version 74) and GENCODE (v19) have been loaded and visualized in the USCS Genome Browser . All the transcripts, generated by alternative splicing, of protein coding and non-coding genes have been considered for the analysis. Only for genes proximal to the breakpoint, peaks of ChIP-Seq data for transcription factors (TFs) binding (GENCODE) were analyzed. The presence of binding sites for specific transcription factors was evaluated by visual inspection of ChIP-Seq tracks on Genome Browser.
The expression patterns, for both the protein and - where available - the transcript, were evaluated by using The Human Protein Atlas database .
This research was approved by the Ethics Committee of University of Naples “Federico II” and written informed consent for participation in the study was obtained from all participants and from the parents of patients under 18 years of age before blood/saliva sampling.
The authors particularly thank the family who participated in this study. S. I. was supported by MO. DO- Model Organism grant and A.V. was supported by CREME grant.
- Rehm HL. Genetics and the Genome Project. Ear Hearing. 2003;24:270–4.View ArticlePubMedGoogle Scholar
- Leonard H, Wen X. The epidemiology of mental retardation: challenges and opportunities in the new millennium. Ment Retar Dev Disab Res Rev. 2002;8:117–34.View ArticleGoogle Scholar
- Des Portes V. X-linked mental deficiency. Handb Clin Neurol. 2013;111:297–306.View ArticlePubMedGoogle Scholar
- Kleefstra T, Hamel BCJ. X-linked mental retardation: further lumping, splitting and emerging phenotypes. Clin Genet. 2005;67:451–67.View ArticlePubMedGoogle Scholar
- Ropers HH, Hamel BC. X-linked mental retardation. Nat Rev Genet. 2005;6:46–57.View ArticlePubMedGoogle Scholar
- Ayazi S. Choroideremia, obesity, and congenital deafness. Am J Ophthalmol. 1981;92:63–9.View ArticlePubMedGoogle Scholar
- Tabor A, Andersen O, Lundsteen C, Niebuhr E, Sardemann H. Interstitial deletion in the “critical region” of the long arm of the X chromosome in a mentally retarded boy and his normal mother. Hum Genet. 1983;64:196–9.View ArticlePubMedGoogle Scholar
- Rosenberg T, Schwartz M, Niebuhr E, Yang HM, Sardemann H, Andersen O, et al. Choroideremia in interstitial deletion of the X chromosome. Ophthalmic Paediatr Genet. 1986;7:205–10.View ArticlePubMedGoogle Scholar
- de Kok YJ, Vossenaar ER, Cremers CW, Dahl N, Laporte J, Hu LJ, et al. Identification of a hot spot for microdeletions in patients with X-linked deafness type 3 (DFN3) 900 kb proximal to the DFN3 gene POU3F4. Hum Mol Genet. 1996;5:1229–35.View ArticlePubMedGoogle Scholar
- Hildebrand MS, de Silva MG, Tan TY, Rose E, Nishimura C, Tolmachova T, et al. Molecular characterization of a novel X-linked syndrome involving developmental delay and deafness. Am J Med Genet A. 2007;143A:2564–75.View ArticlePubMedGoogle Scholar
- Rush ET, Schaefer GB. Identification of an X-linked deletion syndrome through comparative genomic hybridization microarray. Semin Pediatr Neurol. 2010;17:51–3.View ArticlePubMedGoogle Scholar
- Cremers FP, van de Pol DJ, Diergaarde PJ, Wieringa B, Nussbaum RL, Schwartz M, et al. Physical fine mapping of the choroideremia locus using Xq21 deletions associated with complex syndromes. Genomics. 1989;4:41–6.View ArticlePubMedGoogle Scholar
- Merry DE, Lesko JG, Sosnoski DM, Lewis RA, Lubinsky M, Trask B, et al. Choroideremia and deafness with stapes fixation: a contiguous gene deletion syndrome in Xq21. Am J Hum Genet. 1989;45:530–40.PubMed CentralPubMedGoogle Scholar
- Naranjo S, Voesenek K, de la Calle-Mustienes E, Robert-Moreno A, Kokotas H, Grigoriadou M, et al. Multiple enhancers located in a 1-Mb region upstream of POU3F4 promote expression during inner ear development and may be required for hearing. Hum Genet. 2010;128:411–9.View ArticlePubMed CentralPubMedGoogle Scholar
- Anger GJ, Crocker S, McKenzie K, Brown KK, Morton CC, Harrison K, et al. X-linked deafness-2 (DFNX2) phenotype associated with a paracentric inversion upstream of POU3F4. Am J Audiol. 2014;23:1–6.View ArticlePubMedGoogle Scholar
- Choi JW, Min B, Kim A, Koo JW, Kim CS, Park WY, et al. De Novo Large Genomic Deletions Involving POU3F4 in Incomplete Partition Type III Inner Ear Anomaly in East Asian Populations and Implications for Genetic Counseling. Otol Neurotol. 2015;36:184–90.PubMedGoogle Scholar
- Nussbaum RL, Lesko JG, Lewis RA, Ledbetter SA, Ledbetter DH. Isolation of anonymous DNA sequences from within a submicroscopic X chromosomal deletion in a patient with choroideremia, deafness, and mental retardation. Proc Natl Acad Sci U S A. 1987;84:6521–5.View ArticlePubMed CentralPubMedGoogle Scholar
- Rosenberg T, Niebuhr E, Yang HM, Parving A, Schwartz M. Choroideremia, congenital deafness and mental retardation in a family with an X chromosomal deletion. Ophthalmic Paediatr Genet. 1987;8:139–43.View ArticlePubMedGoogle Scholar
- Schwartz M, Yang HM, Niebuhr E, Rosenberg T, Page DC. Regional localization of polymorphic DNA loci on the proximal long arm of the X chromosome using deletions associated with choroideremia. Hum Genet. 1988;78:156–60.View ArticlePubMedGoogle Scholar
- Song MH, Lee HK, Choi JY, Kim S, Bok J, Kim UK. Clinical evaluation of DFN3 patients with deletions in the POU3F4 locus and detection of carrier female using MLPA. Clin Genet. 2010;78:524–32.View ArticlePubMedGoogle Scholar
- Chinetti V, Iossa S, Auletta G, Corvino V, De Luca M, De Falco F, et al. Mutational analysis for GJB2, GJB6, and GJB3 genes in Campania within a universal neonatal hearing screening programme. Int J Audiol. 2011;50:866–70.View ArticlePubMedGoogle Scholar
- de Kok YJ, van der Maarel SM, Bitner-Glindzicz M, Huber I, Monaco AP, Malcolm S, et al. Association between X-linked mixed deafness and mutations in the POU domain gene POU3F4. Science. 1995;267:685–8.View ArticlePubMedGoogle Scholar
- Tarpey PS, Smith R, Pleasance E, Whibley A, Edkins S, Hardy C, et al. A systematic, large-scale resequencing screen of X-chromosome coding exons in mental retardation. Nat Genet. 2009;41:535–43.View ArticlePubMed CentralPubMedGoogle Scholar
- van Bokhoven H, van den Hurk JA, Bogerd L, Philippe C, Gilgenkrantz S, de Jong P, et al. Cloning and characterization of the human choroideremia gene. Hum Mol Genet. 1994;3:1041–6.View ArticlePubMedGoogle Scholar
- Junn E, Mouradian MM. MicroRNAs in neurodegenerative diseases and their therapeutic potential. Pharmacol Ther. 2012;133:142–50.View ArticlePubMed CentralPubMedGoogle Scholar
- Costa V, Esposito R, Aprile M, Ciccodicola A. Non-coding RNA and pseudogenes in neurodegenerative diseases: “The (un)Usual Suspects”. Front Genet. 2012;3:231.View ArticlePubMed CentralPubMedGoogle Scholar
- Poliseno L, Salmena L, Zhang J, Carver B, Haveman WJ, Pandolfi PP. A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature. 2010;465:1033–8.View ArticlePubMed CentralPubMedGoogle Scholar
- Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP. A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell. 2011;146:353–8.View ArticlePubMed CentralPubMedGoogle Scholar
- Database of Genomic Variants [http://dgv.tcag.ca/dgv/app/home]
- Landa G, Rosen RB, Garcia PM, Seiple WH. Combined three-dimensional spectral OCT/SLO topography and microperimetry: steps toward achieving functional spectral OCT/SLO. Ophthalmic Res. 2009;43:92–8.View ArticlePubMedGoogle Scholar
- Forte R, Cennamo G, Finelli ML, Bonavolontà P, Greco GM, de Crecchio G. Retinal Micropseudocysts in Diabetic Retinopathy: Prospective Functional and Anatomic Evaluation. Ophthalmic Res. 2012;2(48):6–11.View ArticleGoogle Scholar
- Chinetti V, Iossa S, Auletta G, Laria C, De Luca M, Di Leva F, et al. Screening for GJB2 and GJB6 gene mutations in patients from Campania region with sensorineural hearing loss. Int J Audiol. 2010;49:326–31.View ArticlePubMedGoogle Scholar
- del Castillo FJ, Rodríguez-Ballesteros M, Alvarez A, Hutchin T, Leonardi E, de Oliveira CA, et al. A novel deletion involving the connexin-30 gene, del(GJB6-d13s1854), found in trans with mutations in the GJB2 gene (connexin-26) in subjects with DFNB1 non-syndromic hearing impairment. J Med Genet. 2005;42:588–94.View ArticlePubMed CentralPubMedGoogle Scholar
- O’Connell JR, Weeks DE. PedCheck: A program for identifying genotype incompatibilities in linkage analysis. Am J Hum Genet. 1998;63:259–66.View ArticlePubMed CentralPubMedGoogle Scholar
- Abecasis GR, Cherny SS, Cookson WO, Cardon LR. Merlin–rapid analysis of dense genetic maps using sparse gene flow trees. Nat Genet. 2002;30:97–101.View ArticlePubMedGoogle Scholar
- Primer3 software [http://bioinfo.ut.edu/primer3-0.4.0/]
- BLAST [http://blast.ncbi.nlm.nih.gov/Blast.cgi]
- Genome Browser [http://genome.ucsc.edu/]
- Human Protein Atlas database [http://www.proteinatlas.org]
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.