Folklore Clinical Variant Interpretation

Overview: Folklore Clinical Variant Interpretation

Folklore helps geneticists interpret variants from VCF files, bringing ACMG/AMP classification and supporting evidence together for review by a qualified specialist.

Developed by Helena Bioinformatics, it addresses a central genomics bottleneck: interpreting the clinical and scientific significance of genetic variation requires teams to reconcile large numbers of variant records, classification labels, and supporting evidence, and variants of uncertain significance can be especially time-consuming to investigate. The public interface presents genomic classifications spanning pathogenic, likely pathogenic, VUS, likely benign, and benign records, enabling structured exploration of variant-interpretation evidence.

At a high level, Folklore Clinical Variant Interpretation organises genomic information into a searchable interface that lets users move from a question- or variant-oriented query to relevant evidence and classification context. This can reduce the time required to locate and compare information across dispersed genomic resources, support more consistent preliminary evidence review, and make complex variant data more accessible for research and interpretation workflows. By consolidating evidence discovery into one environment, the platform may help accelerate investigation timelines and improve the traceability of early-stage genomic assessment, while expert review remains necessary for any clinical or diagnostic conclusion.

Last checked on 8 September 2026: Folklore remained active at the time of review. Its latest publicly indexed performance material was the Cohort 4: Deterministic Classification Performance Study, published in August 2026. The study evaluated a locked deployment of Folklore v3.39.1, reported results from a defined real-world cohort, and described a correction-and-full-cohort-regression programme for confirmed classifier defects and reference-data findings. Those study findings are specific to the stated cohort, methodology, reference conditions, and software version; they should not be interpreted as a general performance guarantee for future releases or all clinical settings.

What is Folklore Clinical Variant Interpretation?

Folklore Clinical Variant Interpretation is Helena Bioinformatics’ platform for annotation and classification of supported GRCh38 germline variants. It provides structured evidence for pathogenic, likely pathogenic, VUS, likely benign, and benign assessments, intended for clinical geneticists, laboratory directors, accreditation and quality-review personnel, and genomics researchers.

Its public documentation describes a workflow that combines VCF annotation, versioned reference sources, ACMG/AMP-oriented classification logic, and manual professional review. The documentation for the platform’s VCF workflow states that it applies rule-based ACMG/AMP classification across the 28 evidence criteria, with recorded data sources and thresholds. This is a description of the documented workflow rather than an independent validation result, and it should be read separately from Folklore’s other methodological descriptions and specialised modules.

Folklore also documents its use of a Bayesian point-based interpretation framework. Its methodology page describes 20 automated ACMG criteria within that framework; this figure relates to the specified methodology and should not be presented as a universal count across every Folklore workflow, data source, configuration, or future software release.

Performance evidence and study scope

Folklore publishes technical and performance materials that describe study design, cohort composition, methodological divergences, limitations, and conflict-of-interest disclosures. Unless stated otherwise, the findings described below are vendor-published results, rather than independently conducted clinical-validation studies.

All reported performance figures should be interpreted only within the scope of the underlying study: the specified cohort, case-selection method, ground-truth approach, reference-data snapshot, analytical method, and locked Folklore software version. They are not general performance guarantees for future releases, other laboratories, other patient populations, or all clinical settings.

• In Helena Bioinformatics’ 29 August 2026 Cohort 4: Deterministic Classification Performance Study, using locked Folklore v3.39.1, the company reported that 100 of 100 real-world cases were completed. Its primary analysis covered 367 evaluable outcomes from 395 reported outcomes. The report also describes forensic review of priority variants and a correction-and-full-cohort-regression process for confirmed classifier defects and reference-data findings. These are vendor-published findings for the specified study design and software version; they are not independent estimates of clinical performance and do not substitute for laboratory-specific validation.

• In Folklore’s public Cohort 3 material, the company reported no 'opposite-direction' partial discordance in the documented analysis, while also reporting adjacent-category divergences, such as likely pathogenic versus VUS or likely benign versus VUS. And outcomes it categorised as both-defensible. Read this observation only in relation to the Cohort 3 dataset and its reported methodology, not as a platform-wide claim that opposite-direction discordance cannot occur.

• Folklore’s documentation describes a VCF classification workflow that applies rule-based ACMG/AMP classification across 28 evidence criteria, with recorded data sources and thresholds. Separately, its Bayesian point-based methodology page describes 20 automated ACMG criteria. These are documentation-based descriptions of particular workflows and methodologies, not comparative performance outcomes; their applicability may differ by module, reference-data release, configuration, and future software version.