{
  "id": "transcription-start-site",
  "title": "Transcription Start Site",
  "lang": "en",
  "source_lang": "en",
  "summary": "A transcription start site is the DNA position at which RNA polymerase begins synthesizing an RNA transcript.",
  "category": "science",
  "subcategory": "biology",
  "type": "concept",
  "aliases": [
    "TSS",
    "transcription initiation site",
    "transcriptional start site"
  ],
  "generated_at": "2026-11-05",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/transcription-start-site/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/transcription-start-site/",
      "json": "https://aiwiki.page/zh/transcription-start-site.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "A transcription start site (TSS) is the position in DNA corresponding to the first nucleotide incorporated into an RNA molecule during transcription. It marks where RNA polymerase begins RNA synthesis, rather than where it first binds DNA or where protein synthesis begins. A gene may have one predominant TSS or multiple start sites whose use varies among cell types, developmental stages, and environmental conditions."
    },
    {
      "heading": "Position and coordinate conventions",
      "text": "In conventional gene diagrams, the first transcribed nucleotide is designated +1. Positions upstream of it are numbered negatively, such as −1 and −10, whereas subsequent transcribed positions are +2, +3, and so forth. This relative numbering convention has no position zero.\n\n“Upstream” and “downstream” refer to the direction of transcription, not necessarily to increasing coordinates in a genome assembly. For a gene transcribed from the reverse strand, downstream positions have decreasing genomic coordinates. A genomic TSS annotation therefore requires both a coordinate and a strand assignment.\n\nRNA polymerase reads the DNA template strand in the \\(3′ \\rightarrow 5′\\) direction and synthesizes RNA in the \\(5′ \\rightarrow 3′\\) direction. The RNA sequence corresponds to the non-template, or coding, strand, with uracil replacing thymine. A TSS is commonly reported as the genomic position corresponding to the first RNA nucleotide.\n\nA TSS must be distinguished from several related features:\n\n- Promoter: the DNA region that directs initiation; the TSS is a position associated with that region.\n- Translation start site: the position at which a ribosome begins translation, usually at a start codon.\n- Annotated gene boundary: a database feature that may reflect incomplete evidence, merged transcript models, or annotation conventions.\n- Mature RNA 5′ end: an RNA end that may have been generated by processing rather than transcription initiation.\n\nFor many protein-coding genes, the interval between the TSS and the translation start site contributes to the 5′ untranslated region of the messenger RNA. In eukaryotes, this interval can contain introns that are subsequently removed by RNA splicing."
    },
    {
      "heading": "How the start site is selected",
      "text": "Start-site selection emerges from interactions among promoter sequence, RNA polymerase, initiation factors, and the local DNA environment. It is not determined by a universal sequence present at every TSS.\n\nBacteria\n\nIn bacteria, a sigma factor associates with RNA polymerase and helps recognize a promoter. Many promoters recognized by the principal sigma factor have characteristic sequence elements approximately 10 and 35 nucleotides upstream of the TSS. These positions are approximate, and promoter architectures vary across sigma factors and species.\n\nPromoter recognition positions the polymerase so that it can open a short region of DNA and begin RNA synthesis. Local sequence and the availability of initiating nucleotides can influence which nearby position is used. A bacterial promoter may therefore produce transcripts beginning at more than one nucleotide.\n\nEukaryotes\n\nIn eukaryotes, start-site selection differs among RNA polymerases and promoter classes. For RNA polymerase II, which transcribes protein-coding genes and many noncoding RNAs, general transcription factors and polymerase assemble into an initiation complex at a core promoter.\n\nSome core promoters contain a TATA box, an initiator element overlapping the start site, or other sequence features. None is universal. Chromatin accessibility and nucleosome organization also affect where initiation can occur. The relative contributions of promoter recognition, DNA opening, and scanning for a suitable initiation position differ among organisms.\n\nRNA polymerases I and III use distinct initiation systems. Their recognition elements are not always confined to DNA upstream of the transcribed region; some polymerase III promoters contain important internal elements.\n\nArchaea\n\nIn archaea, transcription initiation uses factors related to components of the eukaryotic initiation machinery, including TATA-binding protein and transcription factor B. Promoter sequence and initiation-complex geometry help position RNA polymerase relative to the TSS."
    },
    {
      "heading": "Focused initiation, dispersed initiation, and alternative promoters",
      "text": "A promoter need not specify a single invariant start position. Two broad patterns are often distinguished:\n\n- Focused initiation: most initiation events occur at one nucleotide or within a narrow cluster.\n- Dispersed initiation: initiation occurs across a broader region containing multiple start positions.\n\nThese are descriptive categories along a continuum rather than an absolute division. In vertebrates, broad initiation regions are frequently associated with CpG-rich promoters, whereas some focused promoters contain strongly positioned core-promoter elements. These associations are not universal rules.\n\nA gene can also use alternative promoters, producing transcripts with different first exons or different 5′ untranslated regions. This is distinct from small shifts among adjacent start positions within one promoter, although the boundary can be operational rather than sharp.\n\nAlternative initiation can change:\n\n- The timing, location, or abundance of gene expression.\n- RNA stability, localization, or translation efficiency.\n- The selection of downstream splice sites and transcript isoforms.\n- The encoded protein, when promoter choice changes the coding sequence or available translation start sites.\n\nMany alternative TSSs produce the same protein sequence. A change in start-site usage therefore does not by itself establish a change in protein identity or biological function."
    },
    {
      "heading": "Experimental identification",
      "text": "TSS mapping requires distinguishing the original start of an RNA molecule from ends created by cleavage, degradation, or incomplete experimental recovery.\n\nTargeted methods\n\nPrimer extension maps an RNA 5′ end by extending a complementary primer toward that end and measuring the resulting product. S1 nuclease protection uses hybridization and nuclease digestion to identify transcript boundaries.\n\nRapid amplification of cDNA ends (5′ RACE) recovers sequences near RNA 5′ ends. Its ability to identify an authentic TSS depends on the protocol, RNA integrity, and whether primary transcript ends are specifically enriched.\n\nGenome-wide methods\n\nCap analysis of gene expression (CAGE) sequences tags derived from capped RNA 5′ ends and maps them to a reference genome. It is widely used to characterize promoter activity and clusters of initiation positions in eukaryotes. Other cap-enrichment and nascent-RNA methods provide complementary measurements.\n\nIn bacteria, many primary transcripts initially carry a 5′ triphosphate, whereas many processed RNAs carry a 5′ monophosphate. Differential RNA sequencing and related enrichment methods exploit such differences to help identify primary transcript starts. RNA-end chemistry is nevertheless more varied than this simple distinction suggests.\n\nConventional RNA sequencing can support transcript reconstruction, but standard fragmented libraries usually do not identify exact TSSs reliably. Long-read sequencing can connect transcript starts with downstream exon structures, although incomplete 5′ recovery remains a concern."
    },
    {
      "heading": "Interpretation and annotation limitations",
      "text": "An observed RNA 5′ end is not automatically a TSS. Processing and degradation can generate internal ends, and technical biases can alter their apparent abundance. Even capped ends require contextual interpretation because RNA processing and recapping can complicate their relationship to initiation.\n\nA reported TSS may represent an individual nucleotide, the most frequently observed start, or a representative coordinate for a cluster. Different clustering thresholds can produce different promoter counts from the same data.\n\nTSS annotations also depend on the sampled tissue, developmental stage, growth condition, and sequencing depth. A site absent from one dataset may be active elsewhere. Conversely, low-level initiation does not by itself demonstrate production of a stable or functional transcript.\n\nIn functional genomics, TSS maps help define promoter regions, interpret regulatory mutations, and compare initiation across conditions. They are most informative when combined with transcript structure, chromatin measurements, and direct evidence of transcriptional activity rather than treated as fixed, context-independent gene boundaries."
    }
  ],
  "links": [
    "dna",
    "nucleotide",
    "rna",
    "transcription-biology",
    "rna-polymerase",
    "gene",
    "genome",
    "ribosome",
    "translation-biology",
    "codon",
    "messenger-rna",
    "rna-splicing",
    "bacteria",
    "promoter-biology",
    "eukaryote",
    "transcription-factor",
    "chromatin",
    "nucleosome",
    "archaea",
    "gene-expression",
    "protein",
    "rna-sequencing",
    "mutation"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}