- Genetic diseases arise from changes in DNA that affect gene structure, gene expression, protein function, or biological pathways. These changes may be inherited from parents or occur de novo during the formation of reproductive cells or early development. Genetic variants can contribute to single-gene disorders, chromosomal conditions, complex diseases, cancer predisposition, and differences in susceptibility to disease. Understanding these relationships requires access to reliable information about genes, mutations, inherited disorders, clinical features, inheritance patterns, and supporting scientific evidence. Bioinformatics databases provide this information by collecting, organizing, annotating, and connecting genetic variants with human diseases and phenotypes.
- The Human Gene Mutation Database (HGMD) is a major resource for disease-associated genetic variation in humans. It was established in 1996 and is maintained by researchers at the Institute of Medical Genetics in Cardiff. HGMD collects published germline mutations in human nuclear genes that are reported to underlie or be associated with inherited diseases, together with selected disease-associated and functional polymorphisms. It was originally developed to support the study of mutation mechanisms but has become widely used in molecular genetics, genomic research, diagnostic investigation, and bioinformatics. Its records are based substantially on published scientific literature and provide a searchable catalogue of reported gene lesions. <Cite refs={[“turn307423search0″,”turn307423search6”]}/>
- HGMD contains information about different classes of genetic alterations. These include missense and nonsense substitutions, splice-site mutations, regulatory variants, small insertions and deletions, repeat variations, gross deletions, insertions and duplications, and complex genomic rearrangements. Missense variants change an amino acid in a protein, while nonsense variants introduce a premature termination codon. Splice-site variants may disrupt normal RNA processing, and regulatory variants can affect gene expression. Small insertions and deletions may cause frameshifts, while larger deletions and duplications can remove or increase the copy number of genes or genomic regions. Repeat variations may alter the number of repeated DNA units and contribute to disorders such as Huntington disease and several inherited neurological conditions.
- HGMD records generally connect a reported mutation with its associated gene, disease or phenotype, mutation type, literature reference, and available sequence information. Depending on the record and database version, additional information may include nucleotide and protein descriptions, genomic coordinates, external identifiers, and annotations relevant to variant interpretation. These connections allow researchers to investigate the known mutation spectrum of a gene, identify previously reported disease-associated variants, and explore the genetic basis of inherited conditions. HGMD should nevertheless be understood as a literature-based mutation resource rather than a comprehensive catalogue of every human variant or a definitive clinical classification system.
- HGMD distinguishes different categories of reported mutation evidence. Its traditional mutation classes include disease-causing mutations (DM), likely disease-causing mutations (DM?), disease-associated polymorphisms (DP), disease-associated polymorphisms with functional evidence (DFP), and functional polymorphisms (FP). These labels reflect the evidence and classification framework used by HGMD and should not be treated as identical to the five-tier pathogenicity classification commonly used in clinical variant interpretation. In particular, a record labelled as disease-causing in HGMD does not automatically establish that the same variant is pathogenic in every individual, phenotype, inheritance context, or clinical setting. Researchers should examine the original publication, evidence quality, phenotype, segregation data, and current clinical interpretations.
- HGMD provides a public version and a subscription-based HGMD Professional version. The public resource is available to registered academic and non-profit users, but its data are delayed relative to the Professional release. HGMD Professional provides more current mutation records and additional search, annotation, and data-access features, subject to licensing. The Professional version is updated quarterly according to the database’s published description, while the public version is intentionally less current. Users should therefore record which version they accessed and avoid assuming that a search of the public database represents the latest published mutation literature. <Cite refs={[“turn307423search0″,”turn307423search14”]}/>
- HGMD can be used in gene-oriented and mutation-oriented investigations. A researcher studying a disease-associated gene can search for reported variants and review their mutation types, associated phenotypes, and supporting publications. A researcher analysing sequencing data can compare candidate variants with previously reported entries to identify alterations that have been described in inherited disease. Such comparisons can support candidate prioritization, literature review, and the investigation of genotype–phenotype relationships. However, a previously reported variant may have been described in a different phenotype or family, and the original evidence may be limited or subsequently revised. Database matching should therefore be treated as one component of variant assessment rather than a final diagnostic conclusion.
- An important distinction in disease genetics is the difference between a gene–disease relationship and a variant–disease relationship. A gene–disease relationship describes whether changes in a particular gene can cause or contribute to a specific disorder. A variant–disease relationship concerns the interpretation of an individual genetic alteration in that gene. A gene may be associated with several disorders, and different variants in the same gene may produce different phenotypes through distinct molecular mechanisms. Conversely, variants in different genes may cause similar clinical presentations. Understanding these relationships requires integration of mutation databases with disease catalogues, phenotype resources, clinical archives, population-frequency databases, and functional evidence.
- Online Mendelian Inheritance in Man (OMIM) is one of the most important complementary resources for human genetic disease research. It is a curated catalogue of human genes and genetic phenotypes, with particular emphasis on relationships between genetic variation and clinical characteristics. OMIM provides detailed descriptions of inherited disorders, associated genes, allelic variants, inheritance patterns, molecular mechanisms, and relevant scientific literature. It assigns stable identifiers to gene and phenotype records, commonly known as MIM numbers. OMIM is especially useful for understanding the genetic basis of Mendelian disorders and exploring the relationship between different variants in a gene and the clinical features they produce. It is updated regularly and is distinct from HGMD: OMIM is primarily a gene-and-phenotype knowledgebase, whereas HGMD focuses on reported disease-associated mutations. <Cite refs={[“turn307423search4″,”turn307423search5”]}/>
- The ClinVar database, maintained by the National Center for Biotechnology Information (NCBI), is a public archive of submitted interpretations of human genetic variants in relation to inherited disease, cancer, and drug response. It collects variant classifications and supporting evidence from clinical laboratories, research groups, expert panels, and other submitters. ClinVar presents information about the variant, the condition being evaluated, the submitting organization, the interpretation, and the evidence provided. Multiple submitters may interpret the same variant differently, so users should examine the individual submissions, review status, evidence criteria, and any conflicts rather than relying only on a summary label. ClinVar supports both interactive searching and programmatic access through downloadable files and APIs. <Cite refs={[“turn307423search3″,”turn307423search12”]}/>
- ClinGen (Clinical Genome Resource) provides expert-curated resources and frameworks for evaluating the clinical relevance of genes and genetic variants. Its activities include gene–disease validity assessment, dosage sensitivity evaluation, variant interpretation, and the development of expert specifications for applying clinical classification guidelines. ClinGen’s gene–disease validity framework evaluates the strength of evidence linking a gene to a disease, with classifications such as definitive, strong, moderate, limited, disputed, and refuted, as applicable. This is different from classifying a specific variant as pathogenic or benign. ClinGen also supports expert curation groups that develop disease- or gene-specific guidance, helping improve consistency in clinical genomic interpretation.
- Orphanet is a major resource dedicated to rare diseases and orphan drugs. It provides information about rare disorders, their clinical features, associated genes, inheritance patterns, diagnostic resources, and available expert centres. Each disease entity is assigned an ORPHAcode, a stable identifier that helps standardize disease names and support interoperability between clinical and research systems. Orphanet also provides gene–disease information and links to other genetic resources. Its rare-disease classifications are useful for organizing disorders that may have multiple names, overlapping phenotypes, or different levels of clinical and genetic definition. <Cite refs={[“turn307423search8″,”turn307423search13”]}/>
- GeneReviews is an expert-authored clinical resource that provides detailed reviews of inherited disorders. Its chapters commonly describe disease characteristics, molecular genetics, inheritance, diagnosis, management, genetic counselling, and family testing. GeneReviews is particularly useful for connecting molecular findings with clinical practice and understanding how genetic disorders are evaluated and managed. Unlike a mutation archive, it presents disease-focused clinical knowledge in a structured narrative format. It can therefore complement HGMD and ClinVar when researchers need to understand the clinical context of a reported variant or genetic condition.
- NCBI MedGen is a resource for human diseases, disorders, and phenotypes with a genetic component. It integrates standardized disease and phenotype terminology with links to other NCBI and external resources, including ClinVar, OMIM, GeneReviews, Orphanet, PubMed, and genetic testing information. MedGen supports searches by disease name, causative gene, or clinical feature and provides links between related concepts. It is especially useful when the same disorder is described using different names or when a researcher wants to move from a clinical phenotype to associated genes and genetic evidence. <Cite refs={[“turn307423search2”]}/>
- DECIPHER is a genomic and phenotypic resource used particularly in the investigation of rare and undiagnosed genetic disorders. It supports the interpretation of sequence variants and copy-number changes in relation to clinical features. Its data and tools can help researchers examine genes, genomic regions, inheritance patterns, and phenotype similarities. DECIPHER also supports the analysis of developmental disorders and complex genomic findings through connections between genomic alterations and standardized phenotype descriptions. Because some information is derived from clinical submissions and collaborative research, access and data visibility may depend on the type of record and applicable consent or sharing arrangements.
- The Human Phenotype Ontology (HPO) provides standardized terms for describing the clinical features of human diseases. It organizes phenotypic abnormalities into a structured ontology, allowing symptoms and clinical findings to be represented consistently across databases and research studies. Terms may describe features such as developmental delay, seizures, skeletal abnormalities, cardiac defects, or metabolic disturbances. HPO is particularly important in rare-disease diagnosis because patients with the same disorder may show different combinations of features, while unrelated disorders may share overlapping clinical characteristics. Standardized phenotype annotation supports computational comparison between patients and known disease profiles.
- The Gene Curation Coalition (GenCC) brings together organizations that evaluate and report gene–disease validity. Its purpose is to improve consistency and transparency in gene–disease assertions across different resources. GenCC collects classifications from participating curation groups and provides a way to compare their assessments. It is useful when a researcher needs to determine whether a gene–disease association has been evaluated by multiple expert organizations and whether the available evidence is consistent. Different groups may use different curation frameworks or evaluate evidence at different times, so the source and date of each assertion remain important.
- PanelApp provides expert-curated gene panels for different diseases and clinical testing indications. It is used to identify genes considered relevant to particular diagnostic areas, such as inherited cardiac conditions, neuromuscular disorders, metabolic diseases, and developmental disorders. Gene entries may be assigned evidence categories, commonly green, amber, or red, reflecting the level of support for their relevance to a panel. PanelApp is useful in clinical sequencing and diagnostic pipeline design because it helps define which genes should be considered for a particular phenotype or testing indication. Panel versions and review dates should be recorded because gene inclusion and evidence classifications can change.
- The Leiden Open Variation Database (LOVD) is a platform for collecting and sharing genetic variant information, often through gene-specific or disease-specific databases. It supports the storage of variants, associated phenotypes, references, and other relevant information. LOVD installations may be maintained by individual laboratories, clinical groups, or research communities, which means that the scope, curation procedures, and update schedules can vary. It is particularly useful for examining variants in specific genes and accessing specialist community knowledge. Users should evaluate the provenance and quality of each submission and distinguish between different LOVD installations.
- Other resources provide complementary information for particular disease categories. DECIPHER, ClinVar, and ClinGen are especially relevant to clinical variant interpretation and rare-disease genomics. OMIM, Orphanet, and GeneReviews provide detailed disease and phenotype knowledge. gnomAD provides population allele-frequency information that can help assess whether a variant is too common to be compatible with a particular rare disorder, depending on disease prevalence, inheritance, penetrance, and allelic heterogeneity. dbSNP and the European Variation Archive (EVA) provide archival records of genetic variation, while dbVar stores structural-variation observations. These resources answer different questions and should be integrated according to their specific purposes.
- Disease and mutation databases support several important applications in bioinformatics. One major application is the identification and prioritization of candidate disease-causing variants from whole-exome sequencing and whole-genome sequencing data. A typical analysis begins with sequencing reads, followed by alignment to a reference genome, variant calling, quality filtering, and functional annotation. Candidate variants are then evaluated using gene–disease associations, inheritance patterns, phenotype matching, population frequencies, previous reports, and available functional evidence. HGMD and ClinVar can identify previously reported disease-associated variants, while OMIM and Orphanet provide disease context and HPO supports phenotype comparison. The combined evidence helps narrow a large set of detected variants to a smaller group for detailed investigation.
- Variant interpretation commonly considers several evidence categories. These include population frequency, computational predictions, functional studies, segregation within families, de novo occurrence, phenotype consistency, allelic data, and previously established clinical classifications. For example, a rare variant in a gene associated with an autosomal recessive disorder may be more relevant when two disease-associated alleles are identified in the appropriate configuration. A de novo variant in a gene associated with a dominant developmental disorder may also be a strong candidate when the patient’s clinical features match the known disease spectrum. Nevertheless, rarity or phenotype similarity alone is not sufficient to establish pathogenicity.
- Phenotype-driven analysis is another important application. Patients with rare diseases may have complex or incomplete clinical presentations, making diagnosis difficult. HPO terms allow clinical features to be represented in a standardized format and compared with known disease profiles. Tools such as Exomiser and Phen2Gene can combine phenotype similarity with genomic information to prioritize candidate genes and variants. Such methods can help identify disorders that might not be obvious from a single clinical feature. Their results are computational rankings rather than definitive diagnoses and require review of the underlying genetic and clinical evidence.
- Disease databases also support the investigation of genotype–phenotype relationships. Different variants in the same gene can produce different clinical conditions, while similar phenotypes may arise from mutations in different genes. Researchers can examine whether particular mutation types, protein domains, or molecular mechanisms are associated with distinct clinical presentations. For example, variants affecting different regions of a protein may cause loss of function, gain of function, altered protein stability, or abnormal interactions. These mechanisms can produce different disease phenotypes even when the same gene is involved.
- Population genetics is essential for disease-variant interpretation. A variant observed in many apparently healthy individuals may be inconsistent with a highly penetrant, severe, rare Mendelian disorder, although exceptions exist. Population-frequency resources such as gnomAD help estimate allele frequency and examine ancestry-specific distributions. Researchers must consider whether the relevant population is adequately represented and whether the frequency data are suitable for the disease under investigation. Absence from a population database does not prove pathogenicity, and a low-frequency variant is not automatically disease-causing.
- Structural and copy-number variants require additional resources and methods. Large deletions, duplications, inversions, and complex rearrangements may affect multiple genes or regulatory regions. dbVar, DECIPHER, ClinVar, and specialized genomic resources can provide complementary information about these alterations. Interpretation may require evaluation of gene dosage sensitivity, genomic context, breakpoint location, inheritance, and phenotype. A copy-number gain or loss should not be interpreted solely by its size; the genes and regulatory elements affected are often more important than the number of base pairs involved.
- Disease databases are also valuable in clinical genomics, genetic counselling research, and the development of diagnostic tests. They support the identification of known disease genes, the design of targeted gene panels, the assessment of diagnostic yield, and the interpretation of variants detected in families. They can help researchers investigate carrier status, inheritance patterns, recurrence risks, and potential molecular mechanisms. However, database annotations are not substitutes for clinical assessment, laboratory validation, or professional genetic counselling. Variant interpretation must follow current standards and be considered in the context of the individual patient and family.
- Despite their importance, human mutation and disease databases have several limitations. The literature is incomplete, and not all disease-associated variants have been reported or experimentally investigated. Published studies may contain small sample sizes, limited segregation evidence, incomplete phenotype descriptions, or uncertain functional conclusions. Some variants may have been incorrectly classified in older studies or may have been interpreted using standards that have since changed. Different databases may use different curation methods, terminology, evidence thresholds, and update schedules, resulting in inconsistent classifications.
- A further limitation is that disease-associated variants are not always equally represented across populations. Many genetic studies have historically focused on individuals of European ancestry, which can reduce the reliability of variant interpretation in underrepresented populations. Differences in allele frequency between populations can affect the interpretation of rarity, while limited reference data can lead to more variants being classified as uncertain. Improving population diversity in genetic research and reference databases is therefore essential for equitable and accurate genomic medicine.
- Database access, licensing, and data privacy must also be considered. Some resources are freely accessible, while others provide restricted public versions or require subscriptions for complete and current data. HGMD is an important example because its public and Professional versions differ in update status and available features. Clinical genomic records may contain sensitive personal and family information, so researchers must follow consent requirements, data-sharing agreements, and applicable privacy regulations. Automated workflows should preserve database versions, access dates, and provenance to ensure that results can be reproduced.
- Reproducible disease-variant analysis requires careful documentation of reference genome assemblies, transcript versions, variant nomenclature, database releases, annotation software, filtering criteria, and evidence sources. Variants should be normalized and described using standardized nomenclature, such as HGVS, wherever appropriate. Gene symbols and disease identifiers should be mapped to stable identifiers, and phenotype descriptions should use standardized terminology such as HPO. When database records disagree, researchers should preserve the individual assertions and examine their evidence rather than selecting one classification without justification.
- Overall, HGMD and related disease resources provide essential foundations for human genetic research and clinical bioinformatics. HGMD offers a literature-based collection of reported inherited disease-associated mutations, while OMIM, ClinVar, ClinGen, Orphanet, GeneReviews, MedGen, DECIPHER, HPO, GenCC, PanelApp, and LOVD contribute complementary information about genes, variants, diseases, phenotypes, and clinical evidence. Their integration supports variant prioritization, rare-disease diagnosis, genotype–phenotype analysis, clinical genomics, and the development of precision medicine. Effective use requires an understanding of each database’s purpose, evidence standards, coverage, update schedule, licensing conditions, and limitations. Together, these resources help transform genomic data into a more complete understanding of human genetic disease.