引導式測驗 · 中等 等級

AI 在遺傳變異判讀中的應用測驗

測試變異分類類別、證據類型、AI 的支持作用、ClinGen 規範、臨床解釋的限制。

問題 1 的 8

Which categories are included in the ACMG/AMP sequence-variant framework?

預覽每個問題8 問題

當您參加測驗時,就會出現答案和解釋。

  1. Which categories are included in the ACMG/AMP sequence-variant framework?
    • Pathogenic, likely pathogenic, uncertain significance, likely benign, and benign
    • High, medium, and low risk only
    • Diagnostic, prognostic, and therapeutic
    • Positive, negative, and inconclusive test result only
  2. What does a computational prediction contribute to variant interpretation?
    • Proof that a variant causes disease
    • A substitute for phenotype and inheritance review
    • One evidence type to assess alongside other evidence and criteria
    • A standalone clinical classification
  3. How should a variant of uncertain significance be described?
    • It is equivalent to a pathogenic variant
    • It proves the person has the associated disorder
    • It can always guide treatment as a positive result
    • Its clinical significance is unresolved; it is not a confirmed disease-causing finding
  4. Why check the original paper after an AI tool retrieves it?
    • A model can determine whether the evidence is clinically valid from its title
    • The paper may involve a different variant, assay, disease, or inheritance context
    • The retrieval score already confirms every claim
    • The abstract guarantees the clinical classification
  5. How does a ClinGen expert-panel specification guide variant curation?
    • It converts a model ranking directly into a diagnosis
    • It removes the need to review the patient phenotype
    • It refines how variant criteria are applied for a defined gene or disorder
    • It replaces ACMG/AMP evidence categories across all diseases
  6. Which evidence types may contribute to a classification?
    • Population, computational, functional, segregation, and clinical evidence
    • Only a language model’s confidence score
    • Only the variant’s position in the genome
    • Only a single family member’s reported symptom
  7. What should be preserved for a reproducible curation decision?
    • A screenshot of the classification with no provenance
    • Source evidence, applied criteria, date, reviewer, and conflicts
    • Only the model’s final ranking
    • The patient’s name in a public prompt
  8. What can an AI variant-prioritization score establish by itself?
    • That a variant is pathogenic
    • That a family has a specific disease
    • That treatment should begin
    • It can rank candidates for review but does not establish a clinical classification