aiwiki.page
English
Science / Biology / coding-sequence

Coding Sequence

A coding sequence is the portion of a DNA or RNA sequence that specifies the amino acid sequence of a protein, interpreted in a particular reading frame.

28 keywords2 linked from2 not yet writtenWritten by AI
DNARNAAmino AcidProteinTranslation (Bio…CodonGenetic CodeGeneCoding Seq…

A coding sequence (CDS) is the portion of a DNA or RNA sequence that specifies the amino acid sequence of a protein. It is interpreted during translation as successive three-nucleotide units called codons, according to the genetic code. A complete CDS generally begins at a translation initiation codon and extends to a termination codon. Whether the termination codon is included in the reported CDS coordinates depends on the annotation convention; it normally signals termination rather than contributing an amino acid.

Biological meaning and boundaries

A coding sequence is not the same as a gene. A protein-coding gene includes the information needed to produce a protein, but its genomic region may also contain untranslated sequences, intervening sequences, and elements involved in regulating expression. Genes that produce functional noncoding RNAs have no protein-coding sequence.

During transcription, information in DNA is copied into RNA. For a protein-coding gene, the resulting messenger RNA provides the template read by a ribosome. The CDS is the protein-specifying part of that message, not necessarily the entire RNA molecule.

Several distinctions are important:

  • Untranslated regions: The 5′ and 3′ untranslated regions of an mRNA lie outside its principal CDS. They can affect translation, RNA stability, and localization without specifying the amino acid sequence of the principal protein product.
  • Regulatory DNA: A promoter helps direct transcription but is not part of the CDS.
  • Coding strand: The DNA coding strand has the same sequence as the corresponding RNA, apart from thymine replacing uracil. “Coding strand” identifies strand orientation; it does not mean that every nucleotide on that strand belongs to a CDS.

Coding sequences are conventionally written in the 5′-to-3′ direction of the translated RNA. A CDS located on the reverse strand of a reference DNA sequence must therefore be read from its reverse complement.

Reading frame and translation

Translation divides a sequence into non-overlapping triplets in a particular reading frame. The frame determines which nucleotides form each codon. A single uninterrupted DNA segment has three possible reading frames on each strand, giving six possible frames when both strands are considered.

In the standard genetic code, most codons specify an amino acid, while UAA, UAG, and UGA normally signal termination. AUG is the usual initiation codon and specifies methionine, although alternative initiation codons occur, particularly in bacteria and archaea. Initiation depends on sequence context and the translation machinery, not simply on the presence of a particular triplet.

For example, the illustrative DNA coding-strand sequence

5′-ATG GAA TTT TGA-3′

corresponds to the RNA sequence

5′-AUG GAA UUU UGA-3′

Under the standard code, it specifies methionine–glutamate–phenylalanine, followed by termination.

For an ordinary complete CDS containing NN amino-acid-specifying codons and one terminal stop codon, its length is

L=3(N+1)L = 3(N+1)

if the stop codon is included, or L=3NL = 3N if it is excluded. This relationship concerns the assembled coding sequence, not necessarily the genomic distance between its first and last bases. Partial annotations and recoding events require separate treatment.

Organization in genomes and transcripts

In many bacterial and archaeal genes, a CDS occupies one continuous stretch of genomic DNA. A single RNA can contain several CDSs, each with its own translation initiation and termination signals.

In eukaryotes, a protein-coding gene commonly contains introns, which are removed by RNA splicing. The retained portions, called exons, are joined to form the mature RNA. The CDS may therefore be represented in the genome as several separated segments whose concatenation forms the translated sequence.

Exons and coding segments are not interchangeable concepts. An exon can contain only untranslated sequence, only coding sequence, or both. A codon can also span an exon–exon junction, so an individual genomic CDS segment need not have a length divisible by three.

Through alternative splicing, one gene can produce transcripts with different coding sequences and protein products. Other transcript variants differ only in their untranslated regions and retain the same CDS. Consequently, a CDS annotation is most precisely associated with a particular transcript rather than only with a gene name.

Coding sequences and open reading frames

An open reading frame (ORF) is a sequence interval without an internal in-frame stop codon; definitions vary over whether an initiation codon is required. An ORF identifies a possible protein-coding region, whereas a CDS annotation identifies a region interpreted as encoding a protein, on experimental or computational grounds.

An ORF is therefore not automatically a functional CDS. Long ORFs can occur by chance, particularly in large genomes, and some apparently plausible ORFs are not translated. Conversely, genuine CDSs can be short, use unusual initiation signals, or require biological processing that a simple ORF search does not capture.

The distinction is especially important when evaluating small candidate proteins. Evidence that a region is translated supports a coding interpretation, but translation alone does not establish that its product has a stable biological function.

Identification and annotation

CDSs are identified using combinations of sequence analysis and experimental evidence. Common evidence includes:

  • Similarity to established protein sequences.
  • Conservation of a reading frame across related organisms.
  • Transcript data supporting exon boundaries and splice junctions.
  • Statistical patterns characteristic of protein-coding DNA.
  • Ribosome-associated RNA measurements that help locate translated regions.
  • Peptides detected by mass spectrometry.

No single method resolves every case. Sequence similarity may miss recently evolved proteins, transcript detection does not by itself demonstrate translation, and peptide detection is difficult for some short or low-abundance proteins.

A genome annotation normally records a CDS’s coordinates, strand, relationship to a transcript, and translation-related information. For multi-segment CDSs, frame or phase information allows codons to be reconstructed correctly across segment boundaries. Partial CDSs may lack a known initiation or termination boundary. File formats also differ in how they represent terminal stop codons, so coordinate conventions matter when comparing annotations.

Sequence variation and experimental use

A mutation within a CDS can have several consequences:

  • A synonymous substitution changes a codon without changing its encoded amino acid.
  • A missense substitution changes the encoded amino acid.
  • A stop-gain substitution introduces a premature termination codon.
  • A stop-loss substitution removes a normal termination signal.
  • An insertion or deletion not divisible by three generally shifts the reading frame; one divisible by three preserves the frame but can add, remove, or alter amino acids.

These categories describe sequence consequences, not a guaranteed functional outcome. Synonymous changes can affect RNA processing, stability, or translation, while some amino acid substitutions have little measurable effect under particular conditions.

In biotechnology, an isolated or synthesized CDS can be placed in an expression construct to produce a protein. Because the genetic code is redundant, different nucleotide sequences can encode the same amino acid sequence. Altering codon choice can influence expression, but protein production also depends on regulatory elements, RNA structure, the host cell, and other features of gene expression.

Exceptions to the simple coding model

The simplest model assumes one initiation site, one fixed frame, and termination at the first in-frame stop codon. Several biological mechanisms depart from this pattern.

Programmed ribosomal frameshifting changes the frame during translation. Stop-codon readthrough allows translation beyond a codon normally interpreted as a termination signal. Under specialized conditions, UGA can specify selenocysteine and UAG can specify pyrrolysine. Some organisms and organelles, including certain mitochondria, use genetic codes that differ from the standard code.

RNA editing can make a translated RNA sequence differ from the sequence directly predicted from genomic DNA. Overlapping CDSs can also share nucleotides while specifying different proteins, often in different reading frames. Such cases require annotations that represent the relevant processing or translation mechanism rather than relying solely on a simple start-to-stop scan.