The Genomics of Auditory Function and Disease.

Current estimates suggest that nearly half a billion people worldwide are affected by hearing loss. Because of the major psychological, social, economic, and health ramifications, considerable efforts have been invested in identifying the genes and molecular pathways involved in hearing loss, whether genetic or environmental, to promote prevention, improve rehabilitation, and develop therapeutics. Genomic sequencing technologies have led to the discovery of genes associated with hearing loss. Studies of the transcriptome and epigenome of the inner ear have characterized key regulators and pathways involved in the development of the inner ear and have paved the way for their use in regenerative medicine. In parallel, the immense preclinical success of using viral vectors for gene delivery in animal models of hearing loss has motivated the industry to work on translating such approaches into the clinic. Here, we review the recent advances in the genomics of auditory function and dysfunction, from patient diagnostics to epigenetics and gene therapy.

SMARCA4 mutation causes human otosclerosis and a similar phenotype in mice.

BACKGROUND: Otosclerosis is a common cause of adult-onset progressive hearing loss, affecting 0.3%-0.4% of the population. It results from dysregulation of bone homeostasis in the otic capsule, most commonly leading to fixation of the stapes bone, impairing sound conduction through the middle ear. Otosclerosis has a well-known genetic predisposition including familial cases with apparent autosomal dominant mode of inheritance. While linkage analysis and genome-wide association studies suggested an association with several genomic loci and with genes encoding structural proteins involved in bone formation or metabolism, the molecular genetic pathophysiology of human otosclerosis is yet mostly unknown. METHODS: Whole-exome sequencing, linkage analysis, generation of CRISPR mutant mice, hearing tests and micro-CT. RESULTS: Through genetic studies of kindred with seven individuals affected by apparent autosomal dominant otosclerosis, we identified a disease-causing variant in SMARCA4, encoding a key component of the PBAF chromatin remodelling complex. We generated CRISPR-Cas9 transgenic mice carrying the human mutation in the mouse SMARCA4 orthologue. Mutant Smarca4+/E1548K mice exhibited marked hearing impairment demonstrated through acoustic startle response and auditory brainstem response tests. Isolated ossicles of the auditory bullae of mutant mice exhibited a highly irregular structure of the incus bone, and their in situ micro-CT studies demonstrated the anomalous structure of the incus bone, causing disruption in the ossicular chain. CONCLUSION: We demonstrate that otosclerosis can be caused by a variant in SMARCA4, with a similar phenotype of hearing impairment and abnormal bone formation in the auditory bullae in transgenic mice carrying the human mutation in the mouse SMARCA4 orthologue.

A homozygous AP3D1 missense variant in patients with sensorineural hearing loss as the leading manifestation.

Loss-of-function variants in AP3D1 have been linked to Hermansky-Pudlak syndrome (HPS) 10, a severe multisystem disorder characterized by oculocutaneous albinism, immunodeficiency, neurodevelopmental delay, hearing loss (HL), and neurological abnormalities, fatal in early childhood. Here, we report a consanguineous family who presented with presumably isolated autosomal recessive (AR) HL. Whole-exome sequencing was performed on all core family members, and selected patients were screened using array-based copy-number analysis and karyotyping. Candidate variants were validated by Sanger sequencing and assessed in silico. A homozygous, likely pathogenic p.V711I missense variant in AP3D1 segregated with the HL. The family was characterized by thorough medical and laboratory examination. The HL was consistent across patients and accompanied by neurological manifestations in two brothers. The sole female patient was diagnosed with premature ovarian failure. Further findings, including mild neutropenia and reduced NK-cell cytotoxicity in some as well as brain alterations in all homozygous patients, were reminiscent of HPS10, though milder and lacking the characteristic albinism. Previously unrecognized, milder, isolated HL was identified in all heterozygous carriers. A protein model indicates that the variant interferes with protein-protein interactions. These results suggest that a missense variant alters inner-ear-specific functions leading to HL with mild HPS10-like symptoms of variable penetrance. Milder HL in heterozygous carriers may point towards semi-dominant inheritance of this trait. Since all previously reported HPS10 cases were pediatric, it is unknown whether the observed primary ovarian insufficiency recapitulates the subfertility in Ap3d1-deficient mice.

Shared and organ-specific gene-expression programs during the development of the cochlea and the superior olivary complex.

The peripheral and central auditory subsystems together form a complex sensory network that allows an organism to hear. The genetic programs of the two subsystems must therefore be tightly coordinated during development. Yet, their interactions and common expression pathways have never been systematically explored. MicroRNAs (miRNAs) are short non-coding RNAs that regulate gene expression and are essential for normal development of the auditory system. We performed mRNA and small-RNA sequencing of organs from both auditory subsystems at three critical developmental timepoints (E16, P0, P16) to obtain a comprehensive and unbiased insight of their expression profiles. Our analysis reveals common and organ-specific expression patterns for differentially regulated mRNAs and miRNAs, which could be clustered with a particular selection of functions such as inner ear development, Wnt signalling, K+ transport, and axon guidance, based on gene ontology. Bioinformatics detected enrichment of predicted targets of specific miRNAs in the clusters and predicted regulatory interactions by monitoring opposite trends of expression of miRNAs and their targets. This approach identified six miRNAs as strong regulatory candidates for both subsystems. Among them was miR-96, an established critical factor for proper development in both subsystems, demonstrating the strength of our approach. We suggest that other miRNAs identified by this analysis are also common effectors of proper hearing acquirement. This first combined comprehensive analysis of the developmental program of the peripheral and central auditory systems provides important data and bioinformatics insights into the shared genetic program of the two sensory subsystems and their regulation by miRNAs.

Genomic analysis of inherited hearing loss in the Palestinian population.

The genetic characterization of a common phenotype for an entire population reveals both the causes of that phenotype for that place and the power of family-based, population-wide genomic analysis for gene and mutation discovery. We characterized the genetics of hearing loss throughout the Palestinian population, enrolling 2,198 participants from 491 families from all parts of the West Bank and Gaza. In Palestinian families with no prior history of hearing loss, we estimate that 56% of hearing loss is genetic and 44% is not genetic. For the great majority (87%) of families with inherited hearing loss, panel-based genomic DNA sequencing, followed by segregation analysis of large kindreds and transcriptional analysis of participant RNA, enabled identification of the causal genes and mutations, including at distant noncoding sites. Genetic heterogeneity of hearing loss was striking with respect to both genes and alleles: The 337 solved families harbored 143 different mutations in 48 different genes. For one in four solved families, a transcription-altering mutation was the responsible allele. Many of these mutations were cryptic, either exonic alterations of splice enhancers or silencers or deeply intronic events. Experimentally calibrated in silico analysis of transcriptional effects yielded inferences of high confidence for effects on splicing even of mutations in genes not expressed in accessible tissue. Most (58%) of all hearing loss in the population was attributable to consanguinity. Given the ongoing decline in consanguineous marriage, inherited hearing loss will likely be much rarer in the next generation.

PNPT1, MYO15A, PTPRQ, and SLC12A2-associated genetic and phenotypic heterogeneity among hearing impaired assortative mating families in Southern India.

The study was conducted between 2018 and 2020. From a cohort of 113 hearing impaired (HI), five non-DFNB12 probands identified with heterozygous CDH23 variants were subjected to exome analysis. This resolved the etiology of hearing loss (HL) in four South Indian assortative mating families. Six variants, including three novel ones, were identified in four genes: PNPT1 p.(Ala46Gly) and p.(Asn540Ser), MYO15A p.(Leu1485Pro) and p.(Tyr1891Ter), PTPRQ p.(Gln1336Ter), and SLC12A2 p.(Pro988Ser). Compound heterozygous PNPT1 variants were associated with DFNB70 causing prelingual profound sensorineural hearing loss (SNHL), vestibular dysfunction, and unilateral progressive vision loss in one family. In the second family, MYO15A variants in the myosin motor domain, including a novel variant, causing DFNB3, were found to be associated with prelingual profound SNHL. A novel PTPRQ variant was associated with postlingual progressive sensorineural/mixed HL and vestibular dysfunction in the third family with DFNB84A. In the fourth family, the SLC12A2 novel variant was found to segregate with severe-to-profound HL causing DFNA78, across three generations. Our results suggest a high level of allelic, genotypic, and phenotypic heterogeneity of HL in these families. This study is the first to report the association of PNPT1, PTPRQ, and SLC12A2 variants with HL in the Indian population.

The noncoding genome and hearing loss.

The age of sequencing has provided unprecedented insights into the human genome. The coding region of the genome comprises nearly 20,000 genes, of which approximately 4000 are associated with human disease. Beyond the protein-coding genome, which accounts for only 3% of the genome, lies a vast pool of regulatory elements in the form of promoters, enhancers, RNA species, and other intricate elements. These features undoubtably influence human health and disease, and as a result, a great deal of effort is currently being invested in deciphering their identity and mechanism. While a paucity of material has caused a lag in identifying these elements in the inner ear, the emergence of technologies for dealing with a minimal number of cells now has the field working overtime to catch up. Studies on microRNAs (miRNAs), long non-coding RNAs (lncRNAs), methylation, histone modifications, and more are ongoing. A number of microRNAs and other noncoding elements are known to be associated with hearing impairment and there is promise that regulatory elements will serve as future tools and targets of therapeutics and diagnostics. This review covers the current state of the field and considers future directions for the noncoding genome and implications for hearing loss.

Autosomal dominant non-syndromic hearing loss maps to DFNA33 (13q34) and co-segregates with splice and frameshift variants in ATP11A, a phospholipid flippase gene.

Sequencing exomes/genomes have been successful for identifying recessive genes; however, discovery of dominant genes including deafness genes (DFNA) remains challenging. We report a new DFNA gene, ATP11A, in a Newfoundland family with a variable form of bilateral sensorineural hearing loss (SNHL). Genome-wide SNP genotyping linked SNHL to DFNA33 (LOD = 4.77), a locus on 13q34 previously mapped in a German family with variable SNHL. Whole-genome sequencing identified 51 unremarkable positional variants on 13q34. Continuous clinical ascertainment identified several key recombination events and reduced the disease interval to 769 kb, excluding all but one variant. ATP11A (NC_000013.11: chr13:113534963G>A) is a novel variant predicted to be a cryptic donor splice site. RNA studies verified in silico predictions, revealing the retention of 153 bp of intron in the 3' UTR of several ATP11A isoforms. Two unresolved families from Israel were subsequently identified with a similar, variable form of SNHL and a novel duplication (NM_032189.3:c.3322_3327+2dupGTCCAGGT) in exon 28 of ATP11A extended exon 28 by 8 bp, leading to a frameshift and premature stop codon (p.Asn1110Valfs43Ter). ATP11A is a type of P4-ATPase that transports (flip) phospholipids from the outer to inner leaflet of cell membranes to maintain asymmetry. Haploinsufficiency of ATP11A, the phospholipid flippase that specially transports phosphatidylserine (PS) and phosphatidylethanolamine (PE), could leave cells with PS/PE at the extracellular side vulnerable to phagocytic degradation. Given that surface PS can be pharmaceutically targeted, hearing loss due to ATP11A could potentially be treated. It is also likely that ATP11A is the gene underlying DFNA33.

Disease-specific ACMG/AMP guidelines improve sequence variant interpretation for hearing loss.

PURPOSE: The ClinGen Variant Curation Expert Panels (VCEPs) provide disease-specific rules for accurate variant interpretation. Using the hearing loss-specific American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) guidelines, the Hearing Loss VCEP (HL VCEP) illustrates the utility of expert specifications in variant interpretation. METHODS: A total of 157 variants across nine HL genes, previously submitted to ClinVar, were curated by the HL VCEP. The curation process involved collecting published and unpublished data for each variant by biocurators, followed by bimonthly meetings of an expert curation subgroup that reviewed all evidence and applied the HL-specific ACMG/AMP guidelines to reach a final classification. RESULTS: Before expert curation, 75% (117/157) of variants had single or multiple variants of uncertain significance (VUS) submissions (17/157) or had conflicting interpretations in ClinVar (100/157). After applying the HL-specific ACMG/AMP guidelines, 24% (4/17) of VUS and 69% (69/100) of discordant variants were resolved into benign (B), likely benign (LB), likely pathogenic (LP), or pathogenic (P). Overall, 70% (109/157) variants had unambiguous classifications (B, LB, LP, P). We quantify the contribution of the HL-specified ACMG/AMP codes to variant classification. CONCLUSION: Expert specification and application of the HL-specific ACMG/AMP guidelines effectively resolved discordant interpretations in ClinVar. This study highlights the utility of ClinGen VCEPs in supporting more consistent clinical variant interpretation.

Expression pattern of cochlear microRNAs in the mammalian auditory hindbrain.

The auditory system comprises the auditory periphery, engaged in sound transduction and the central auditory system, implicated in auditory information processing and perception. Recently, evidence mounted that the mammalian peripheral and central auditory systems share a number of genes critical for proper development and function. This bears implication for auditory rehabilitation and evolution of the auditory system. To analyze to which extent microRNAs (miRNAs) belong to genes shared between both systems, we characterize the expression pattern of 12 cochlea-abundant miRNAs in the central auditory system. Quantitative real-time PCR (qRT-PCR) demonstrated expression of all 12 genes in the cochlea, the auditory hindbrain and the non-auditory prefrontal cortex (PFC) at embryonic stage (E)16 and postnatal stages (P)0 and P30. Eleven of them showed differences in expression between tissues and nine between the developmental time points. Hierarchical cluster analysis revealed that the temporal expression pattern in the auditory hindbrain was more similar to the PFC than to the cochlea. Spatiotemporal expression analysis by RNA in situ hybridization demonstrated widespread expression throughout the cochlear nucleus complex (CNC) and the superior olivary complex (SOC) during postnatal development. Altogether, our data indicate that miRNAs represent a relevant class of genetic factors functioning across the auditory system. Given the importance of gene regulatory network (GRN) components for development, physiology and evolution, the 12 miRNAs provide promising entry points to gain insights into their molecular underpinnings in the auditory system.

Identification and characterization of key long non-coding RNAs in the mouse cochlea.

The auditory system is a complex sensory network with an orchestrated multilayer regulatory programme governing its development and maintenance. Accumulating evidence has implicated long non-coding RNAs (lncRNAs) as important regulators in numerous systems, as well as in pathological pathways. However, their function in the auditory system has yet to be explored. Using a set of specific criteria, we selected four lncRNAs expressed in the mouse cochlea, which are conserved in the human transcriptome and are relevant for inner ear function. Bioinformatic characterization demonstrated a lack of coding potential and an absence of evolutionary conservation that represent properties commonly shared by their class members. RNAscope®  analysis of the spatial and temporal expression profiles revealed specific localization to inner ear cells. Sub-cellular localization analysis presented a distinct pattern for each lncRNA and mouse tissue expression evaluation displayed a large variability in terms of level and location. Our findings establish the expression of specific lncRNAs in different cell types of the auditory system and present a potential pathway by which the lncRNA Gas5 acts in the inner ear. Studying lncRNAs and deciphering their functions may deepen our knowledge of inner ear physiology and morphology and may reveal the basis of as yet unresolved genetic hearing loss-related pathologies. Moreover, our experimental design may be employed as a reference for studying other inner ear-related lncRNAs, as well as lncRNAs expressed in other sensory systems.

miR-96 is required for normal development of the auditory hindbrain.

The peripheral deafness gene Mir96 is expressed in both the cochlea and central auditory circuits. To investigate whether it plays a role in the auditory system beyond the cochlea, we characterized homozygous Dmdo/Dmdo mice with a point mutation in miR-96. Anatomical analysis demonstrated a significant decrease in volume of auditory nuclei in Dmdo/Dmdo mice. This decrease resulted from decreased cell size. Non-auditory structures in the brainstem of Dmdo/Dmdo mice or auditory nuclei of the congenital deaf Cldn14-/- mice revealed no such differences. Electrophysiological analysis in the medial nucleus of the trapezoid body (MNTB) showed that principal neurons fired preferentially multiple action potentials upon depolarization, in contrast to the single firing pattern prevalent in controls and Cldn14-/- mice. Immunohistochemistry identified significantly reduced expression of two predicted targets of the mutated miR-96, Kv1.6 and BK channel proteins, possibly contributing to the electrophysiological phenotype. Microscopic analysis of the Dmdo/Dmdo calyx of Held revealed a largely absent compartmentalized morphology, as judged by SV2-labeling. Furthermore, MNTB neurons from Dmdo/Dmdo mice displayed larger synaptic short-term depression, slower AMPA-receptor decay kinetics and a larger NMDA-receptor component, reflecting a less matured stage. Again, these synaptic differences were not present between controls and Cldn14-/- mice. Thus, deafness genes differentially affect the auditory brainstem. Furthermore, our study identifies miR-96 as an essential gene regulatory network element of the auditory system which is required for functional maturation in the peripheral and central auditory system alike.

Spectrum of genes for inherited hearing loss in the Israeli Jewish population, including the novel human deafness gene ATOH1.

Mutations in more than 150 genes are responsible for inherited hearing loss, with thousands of different, severe causal alleles that vary among populations. The Israeli Jewish population includes communities of diverse geographic origins, revealing a wide range of deafness-associated variants and enabling clinical characterization of the associated phenotypes. Our goal was to identify the genetic causes of inherited hearing loss in this population, and to determine relationships among genotype, phenotype, and ethnicity. Genomic DNA samples from informative relatives of 88 multiplex families, all of self-identified Jewish ancestry, with either non-syndromic or syndromic hearing loss, were sequenced for known and candidate deafness genes using the HEar-Seq gene panel. The genetic causes of hearing loss were identified for 60% of the families. One gene was encountered for the first time in human hearing loss: ATOH1 (Atonal), a basic helix-loop-helix transcription factor responsible for autosomal dominant progressive hearing loss in a five-generation family. Our results show that genomic sequencing with a gene panel dedicated to hearing loss is effective for genetic diagnoses in a diverse population. Comprehensive sequencing enables well-informed genetic counseling and clinical management by medical geneticists, otolaryngologists, audiologists, and speech therapists and can be integrated into newborn screening for deafness.

Striatin is a novel modulator of cell adhesion.

The adherens junctions (AJs) and tight junctions (TJs) provide critical adhesive contacts between neighboring epithelial cells and are crucial for epithelial adhesion, integrity, and barrier functions in a wide variety of tissues and organisms. The striatin protein family, which are part of the striatin interaction phosphatases and kinases complex, are multidomain scaffolding proteins that play important biologic roles. We have previously shown that striatin colocalizes with the tumor suppressor protein adenomatous polyposis coli in the TJs of epithelial cells. Here we show that striatin affects junction integrity and cell migration, probably through a mechanism that involves the adhesion molecule E-cadherin. Cells engaged in cell-cell adhesion expressed a high MW-modified form of striatin that forms stable associations with detergent-insoluble, membrane-bound cellular fractions. In addition, striatin has recently been suggested to be a target of the poly (ADP-ribose) polymerases Tankyrase 1, and we have found that striatin interacts with Tankyrase 1 and is subsequently poly-ADP-ribosylated. Taken together, our results suggest that striatin is a novel cell-cell junctional protein that functions to maintain correct cell adhesion and may have a role in establishing the relationship between AJs and TJs that is fundamental for epithelial cell-cell adhesion.-Lahav-Ariel, L., Caspi, M., Nadar-Ponniah, P. T., Zelikson, N., Hofmann, I., Hanson, K. K., Franke, W. W., Sklan, E. H., Avraham, K. B., Rosin-Arbesfeld, R. Striatin is a novel modulator of cell adhesion.

[INNOVATIONS IN RESEARCH OF HEREDITARY DEAFNESS].

INTRODUCTION: Deafness is the most common sensory disability in humans affecting all aspects of life. Approximately 50% of congenital deafness is hereditary and about half of genetic deafness is still unsolved. To date, more than 150 genes are known to cause hearing loss worldwide, with specific genes contributing to deafness in distinct populations. Of these, more than 20 genes are involved in deafness among the Jewish Israeli hearing-impaired population. The most common gene in many worldwide populations, including Israel, is GJB2, which encodes the connexin 26 protein. The second most common gene among Jews is TMC1, with most pathogenic variants found only among Jews of Moroccan origin. Most other pathogenic variants found in the Jewish population are origin-specific and not found in other Jewish ethnic groups or in other worldwide populations. In patients where hereditary deafness is suspected, known variants in the specific ethnicity are routinely examined. In Israel, the GJB2 gene is screened in all cases of hereditary deafness and the TMC1 gene is screened in deaf persons of Jewish Moroccan origin. In cases where no variant is found in a known gene, more comprehensive diagnostic tests should be used. Since the beginning of the deep sequencing era, less than a decade ago, the number of deafness-related genes in the Jewish population has increased by threefold. Identifying the pathogenic variant makes it possible to study molecular pathogenesis, to anticipate and understand the prognosis, to calculate probability of concomitant morbidity, to offer prenatal diagnosis, prevent recurrence of deafness in the family and early rehabilitation. Currently, cochlear implant offers the greatest chance for rehabilitation. The hope is that understanding the molecular pathogenesis will in the future lead to personalized medical treatment. We review the genetics of deafness, with an emphasis on the Jewish population in Israel, new diagnostic methods and suggest a diagnostic algorithm and future treatment methods.

Insights into inner ear-specific gene regulation: Epigenetics and non-coding RNAs in inner ear development and regeneration.

The vertebrate inner ear houses highly specialized sensory organs, tuned to detect and encode sound, head motion and gravity. Gene expression programs under the control of transcription factors orchestrate the formation and specialization of the non-sensory inner ear labyrinth and its sensory constituents. More recently, epigenetic factors and non-coding RNAs emerged as an additional layer of gene regulation, both in inner ear development and disease. In this review, we provide an overview on how epigenetic modifications and non-coding RNAs, in particular microRNAs (miRNAs), influence gene expression and summarize recent discoveries that highlight their critical role in the proper formation of the inner ear labyrinth and its sensory organs. Finally, we discuss recent insights into how epigenetic factors and miRNAs may facilitate, or in the case of mammals, restrict inner ear sensory hair cell regeneration.

Consensus interpretation of the p.Met34Thr and p.Val37Ile variants in GJB2 by the ClinGen Hearing Loss Expert Panel.

PURPOSE: Pathogenic variants in GJB2 are the most common cause of autosomal recessive sensorineural hearing loss. The classification of c.101T>C/p.Met34Thr and c.109G>A/p.Val37Ile in GJB2 are controversial. Therefore, an expert consensus is required for the interpretation of these two variants. METHODS: The ClinGen Hearing Loss Expert Panel collected published data and shared unpublished information from contributing laboratories and clinics regarding the two variants. Functional, computational, allelic, and segregation data were also obtained. Case-control statistical analyses were performed. RESULTS: The panel reviewed the synthesized information, and classified the p.Met34Thr and p.Val37Ile variants utilizing professional variant interpretation guidelines and professional judgment. We found that p.Met34Thr and p.Val37Ile are significantly overrepresented in hearing loss patients, compared with population controls. Individuals homozygous or compound heterozygous for p.Met34Thr or p.Val37Ile typically manifest mild to moderate hearing loss. Several other types of evidence also support pathogenic roles for these two variants. CONCLUSION: Resolving controversies in variant classification requires coordinated effort among a panel of international multi-institutional experts to share data, standardize classification guidelines, review evidence, and reach a consensus. We concluded that p.Met34Thr and p.Val37Ile variants in GJB2 are pathogenic for autosomal recessive nonsyndromic hearing loss with variable expressivity and incomplete penetrance.

Computational analysis of mRNA expression profiling in the inner ear reveals candidate transcription factors associated with proliferation, differentiation, and deafness.

BACKGROUND: Hearing loss is a major cause of disability worldwide, impairing communication, health, and quality of life. Emerging methods of gene therapy aim to address this morbidity, which can be employed to fix a genetic problem causing hair cell dysfunction and to promote the proliferation of supporting cells in the cochlea and their transdifferentiation into hair cells. In order to extend the applicability of gene therapy, the scientific community is focusing on discovery of additional deafness genes, identifying new genetic variants associated with hearing loss, and revealing new factors that can be manipulated in a coordinated manner to improve hair cell regeneration. Here, we addressed these challenges via genome-wide measurement and computational analysis of transcriptional profiles of mouse cochlea and vestibule sensory epithelium at embryonic day (E)16.5 and postnatal day (P)0. These time points correspond to developmental stages before and during the acquisition of mechanosensitivity, a major turning point in the ability to hear. RESULTS: We hypothesized that tissue-specific transcription factors are primarily involved in differentiation, while those associated with development are more concerned with proliferation. Therefore, we searched for enrichment of transcription factor binding motifs in genes differentially expressed between the tissues and between developmental ages of mouse sensory epithelium. By comparison with transcription factors known to alter their expression during avian hair cell regeneration, we identified 37 candidates likely to be important for regeneration. Furthermore, according to our estimates, only half of the deafness genes in human have been discovered. To help remedy the situation, we developed a machine learning classifier that utilizes the expression patterns of genes to predict how likely they are to be undiscovered deafness genes. CONCLUSIONS: We used a novel approach to highlight novel additional factors that can serve as points of intervention for enhancing hair cell regeneration. Given the similarities between mouse and human deafness, our predictions may be of value in prioritizing future research on novel human deafness genes.