DNA is a four-letter text, and almost everything a cell does starts with someone reading it three letters at a time. The trouble is that the reading can begin at any of three positions, on either of two strands. Pick the wrong start and you get a different protein, or none at all. A new browser tool, DNA Lab, lets you paste a sequence and watch that happen.
Prefer more room? Open DNA Lab full screen. Nothing you paste leaves your browser.
Reading frames: the same letters, six different messages
The genetic code is a lookup table with 64 entries, because four bases read three at a time give 4 × 4 × 4 combinations. Sixty-one of those codons specify an amino acid. Three, UAA, UAG and UGA, tell the ribosome to stop. AUG usually serves as the start signal and is read as methionine (in bacteria, as formylmethionine).
Because codons do not overlap, the starting position matters. A stretch of single-stranded DNA read 5′ to 3′ has three possible frames; double-stranded DNA has six, three on each strand, since the opposite strand is read in the opposite direction. An open reading frame runs from a start codon to an in-frame stop. The Translate tab shows all six frames, colours each amino acid by chemical class, and lists every ATG-to-stop stretch it finds.
CRISPR guides: finding the NGG
The most widely used Cas9, from Streptococcus pyogenes, is steered by a guide RNA whose 20-nucleotide spacer matches the DNA target. The enzyme will only bind next to a short motif called the PAM, which for this Cas9 is NGG (the N is any base). It cuts about three base pairs upstream of that motif. Jinek and colleagues showed in Science in 2012 that this two-part RNA-guided system could be programmed to cut chosen DNA, and Emmanuelle Charpentier and Jennifer Doudna shared the 2020 Nobel Prize in Chemistry for the work. For the underlying biology, see how CRISPR works and how base and prime editing differ.
The CRISPR guides tab scans both strands for 20-nucleotide protospacers followed by NGG. Run on the built-in example, it returns 20 candidate sites. Each one gets its GC content and a few plain flags: GC outside 40–70%, a run of four Ts (which can end transcription from a U6 promoter), long homopolymer runs, and a first base that is not G.
Be clear about what the tool does not do. It performs no off-target search and predicts no cutting efficiency, and it does not check your genome of interest. Those steps need real alignment against a reference genome and experimental validation, a problem covered in how off-target effects are measured. Treat the output as a list of places a guide could go, not as validated guides.
The helix view and the example sequence
The Helix tab draws the first 48 bases as a rotating double helix with A–T and G–C pairs coloured differently. The twist uses a spacing of roughly 10.5 base pairs per turn, a standard figure for B-form DNA. It is a schematic, not a structural model.
The example sequence is synthetic. It was written for this tool as a back-translated peptide with a start and a stop codon, and it is not a record from any sequence database. Paste your own sequence to replace it; FASTA headers, spaces and digits are ignored, and U is treated as T.
Sources
- Genetic code, Wikipedia
- Reading frame, Wikipedia
- Cas9, Wikipedia
- Jinek M. et al., “A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity,” Science, 2012.
