Engine
Codon optimization
Paste a coding sequence, choose a host, and get back a rewrite that suits it — with the constraints you set honoured, the before and after adaptation shown side by side, and a record of what changed and why.
65
host organisms
DNA Chisel
optimization engine
14.1 Mbp
designed DNA
Free
to run
01
What the feature does
A coding sequence that works in one organism is not automatically a sensible thing to put in another. Codon preferences differ, and so do the sequence features a host would rather not encounter. Adapting for that by hand is possible, slow, and almost impossible to document afterwards in a way that survives a handover.
This engine is built on DNA Chisel. You give it a coding sequence and a target host, set the constraints that matter to you, and it returns a rewritten sequence that preserves the protein while shifting codon usage toward what the host prefers. Every constraint it satisfied is listed next to the result, which is the part that makes the output defensible six months later.
02
In the interface
Bring in the sequence
Paste a coding sequence into the editor, upload GenBank or FASTA, or pull one from a project or from the catalog by ID. Imported records keep their annotation, so features you care about are already marked before optimization starts.
Choose the host
Pick from the 65 supported organisms. The host selection sets the codon usage table the rewrite works against, and the interface shows which table is in use rather than leaving it implicit.
Set the constraints
GC content windows, sites to avoid, repeat and homopolymer limits, and regions to leave untouched — a tag, a signal, a junction you have already settled. Constraints are declared up front and the engine works inside them rather than around them.
Run it
Optimization is a CPU job and finishes while you are still looking at the page. Longer sequences and heavier constraint sets take longer; nothing here consumes credits at any length.
Read the before and after
The result view puts the original and the rewrite side by side with a codon-adaptation comparison, a codon usage distribution for each, and a positional view of what moved. Constraint checks are listed with their outcomes, including any the engine could not fully satisfy.
Accept and export
Accepting the rewrite writes it into the project as a new version with the original kept intact. Export as GenBank, FASTA, SBOL, or an IDT or Twist order sheet when the sequence is ready to be made.
03
A worked example
- Input
- Coding sequence from construct SS-0421
- Target host
- Selected from the 65 supported organisms
- Protein sequence
- Unchanged
- GC constraint
- Windowed, host-appropriate range
- Avoided sites
- 3 declared
- Adaptation index, this run
- 0.61 → 0.88
- Constraints satisfied
- All declared
- Engine class
- CPU — no credits
The numbers above describe one run, not a guarantee; what matters in practice is the comparison view. Seeing the codon distribution move from the original toward the host’s own profile, with the avoided sites gone and the protected regions visibly untouched, is the check most people actually run before accepting a rewrite.
SS-0421 is one entry in a catalog of 1,000 constructs spanning 70 series, six tiers from Pepper through Tier 5, and 14.1 Mbp of designed DNA. Every one of those sequences passed through this engine on its way to a host, which is why the catalog can be filtered by organism at all.
04
Who runs it, and what it costs
Cross-species work leans on it hardest: moving a pathway between organisms means running every coding sequence in the set against the new host, and the batch queue turns that into one job with one results table. Teams designing for synthesis use it last, immediately before export, because vendor-friendly constraints are easiest to apply once everything else is settled.
Codon optimization is a CPU engine and free to run, with no cap on sequences or exports. Prepaid credits — $25 for 250 credits, one credit for roughly one Standard job — cover GPU engines such as structure prediction and variant scanning, and are never spent here.
Ready to run it
Try this in the Splice Platform.
CPU engines are free to run. GPU engines run on prepaid credits.
Related