Skip to main content Link Menu Expand (external link) Document Search Copy Copied
Biology Skills

Coding sequence and protein consequences

Summary

A predicted protein consequence is derived from a particular transcript, coding sequence, reading frame, strand, and genetic code. Codon-level annotation describes a molecular consequence of sequence representation; it is not a clinical pathogenicity classification.

Core rules

  • Translation reads mRNA 5-prime to 3-prime in codons of three nucleotides within a defined reading frame.
  • In a simple coding example, the coding DNA strand has the same base order as mRNA except T replaces U; the template strand is complementary and antiparallel.
  • A genomic allele cannot be translated without the relevant transcript, strand, coding boundaries, and reading frame.
  • Human mitochondrial translation differs from the standard nuclear genetic code at several codons.
  • Substitutions, insertions, and deletions can have different consequences across transcripts and coding frames.
  • A synonymous consequence does not prove absence of biological effect.
  • Sequence Ontology terms and Ensembl IMPACT categories describe predicted molecular consequence or heuristic severity, not clinical pathogenicity.
  • Protein notation is reference-sequence dependent. Exact HGVS nomenclature must use the appropriate reference sequence and current HGVS rules.

Required context

For a coding or protein consequence preserve:

  • genomic reference sequence
  • gene and transcript identifier
  • transcript version
  • transcript strand
  • coding sequence boundaries and reading frame
  • protein accession or isoform when reporting protein coordinates
  • genetic code used
  • annotation software and release when consequence terms are generated computationally
  • distinction between predicted consequence and experimentally measured molecular effect

AI behaviour

  • Do not translate a genomic variant into a protein change without establishing transcript and coding-frame context.
  • Do not assume the standard nuclear genetic code for mitochondrial sequence.
  • Do not infer pathogenicity from a consequence term such as missense, frameshift, HIGH, MODERATE, or stop-gained.
  • Do not describe a synonymous variant as biologically neutral solely because the amino acid is unchanged.
  • Do not use a copied consequence-severity table as a substitute for the current annotation provider or Sequence Ontology definition.
  • When exact amino-acid or HGVS notation matters, use the current authoritative sequence and nomenclature standard.
  • Separate predicted consequence from functional evidence. A predicted stop-gained consequence does not by itself demonstrate transcript degradation, protein truncation, disease mechanism, or phenotype.

Common failure modes

Protein change derived from gene name alone

A gene symbol does not define one coding sequence. Alternative transcripts can change codon position, reading frame, or whether the genomic variant is coding at all.

Ensembl IMPACT treated as pathogenicity

HIGH, MODERATE, LOW, and MODIFIER are annotation categories used for predicted consequence severity. They are not clinical variant classifications.

Synonymous treated as no effect

No amino-acid change does not exclude splice, regulatory, RNA, or translational effects.

Nuclear code applied to mitochondrial DNA

Several mitochondrial codons differ from the standard nuclear genetic code. Using the wrong code can assign the wrong amino acid or stop consequence.

Authoritative standards

Use NCBI genetic-code resources for translation tables, HGVS for exact sequence-variant nomenclature, Sequence Ontology for formal consequence terms, and the current annotation-provider documentation for provider-specific consequence categories.

Examples

Coding-strand example

DNA coding strand:  ATG GAA TGG
mRNA:               AUG GAA UGG
protein:            Met Glu Trp

This simple correspondence assumes that the displayed DNA is the coding strand and that the reading frame is already established.

Molecular consequence is not clinical classification

A stop_gained annotation states that the selected transcript model predicts creation of a premature termination codon. Clinical interpretation still requires evidence about gene-disease mechanism, transcript relevance, nonsense-mediated decay, population frequency, segregation, functional data, and other applicable evidence.

Sources

  • NCBI genetic codes: https://www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi
  • HGVS Sequence Variant Nomenclature: https://hgvs-nomenclature.org/
  • Sequence Ontology: http://www.sequenceontology.org/
  • Ensembl calculated variant consequences: https://www.ensembl.org/info/genome/variation/prediction/predicted_data.html