Genetic Circuit

1. 01Turn a logic function into a construct: give it a named gate or a truth table and it builds the smallest NOR/NOT network that computes it
2. 02Assign real characterized repressor gates from the published Cello libraries — six libraries, 75 gates, across E. coli, B. thetaiotaomicron and S. cerevisiae
3. 03Predict the dynamic range: the dimmest ON output divided by the brightest OFF, across every input combination
4. 04Search every assignment exhaustively rather than sampling — and refuse, naming what to narrow, when the space is larger than the budget
5. 05A full truth table showing both the logic value each node intends and the level in RPU it actually reaches, so a design that computes the right function at the wrong levels is visible
6. 06Emit the construct's DNA: each device assembled from the library's own part sequences — ribozyme, ribosome binding site, repressor coding sequence, terminator
7. 07Report which physical gate of a repressor family won each row, so a family whose best member changes between rows is caught rather than hidden
8. 08Name the host, medium, temperature and growth protocol every parameter was measured under, beside the prediction that inherits them
9. 09State what the prediction assumes — insulated parts, no metabolic burden, steady state — with every result
10. 10Report the placement constraints the library declares and this tool does not apply, rather than presenting an arbitrary device order as a plasmid map
11. 11Export the construct as FASTA, one record per device

packages/ports/expression/circuit ports the biological stages of Cello 2 (CIDAR Lab / Voigt lab, Apache-2.0) and is graded against Cello itself at every stage: its UCF reader over every record of all six libraries, its gate evaluator over 600 swept response points, its circuit scorer over 24 circuits, and its exhaustive assignment search over 2,160 assignments. Logic synthesis is ours rather than ported, because Cello uses Yosys — a C++ framework that does not belong in a browser and is the least biological of the four stages. 371 tests.

Everything runs client-side. Files are parsed in the browser and projects are saved on this machine, so an unpublished sequence is never uploaded.

Why does an AND gate cost three gates?

Because repression is inverting. A repressor turns its promoter OFF when it is present, which is a NOT. The only gates a transcriptional library gives you are NOT and NOR, so an AND is built as a NOR of two NOTs, and an exclusive-OR costs six gates. That cost is the real constraint on what a library can express: every gate in the circuit has to be a different repressor, or two gates would silence each other.

Why does the choice of gates matter so much?

Because it decides whether the circuit works at all. On one library's AND circuit the worst assignment scores 1.0 — the output is the same whether the circuit is ON or OFF, so it is not a switch — and the best scores 19.6. Same logic, same library, same parts. Only the pairing differs, which is why the assignment is searched rather than picked.

What is a dynamic range of 19.6 telling me?

That the dimmest ON state is 19.6 times brighter than the brightest OFF state, across every input combination. It is a worst case rather than an average, and it is taken across outputs too, because a circuit is only as good as its weakest readout. Below about 2 there is no usable separation.

Where do the gate parameters come from?

They are measured, from the published Cello libraries: the output each repressor makes when fully off and fully on, the input level at which it switches, and how sharply. Nothing here estimates them. Each library also records the organism, genotype, medium, temperature and induction protocol they were measured under, and those are shown beside the prediction because it inherits them.

What is RPU?

Relative Promoter Units — expression measured against the library's own reference standard rather than in raw fluorescence. It is what makes a number from one lab comparable with a number from another, and it is the unit every level in the truth table is reported in.

Does it give me a finished plasmid?

Not quite. The parts and each device's cassette are correct and complete, but the ORDER of the devices along the construct is set by placement rules that need a constraint solver this tool does not carry. It tells you how many constraints it left unapplied rather than presenting an arbitrary order as a map.

Does it account for burden?

No, and it says so with every result. Gate behaviour measured in isolation is not gate behaviour in a circuit competing for the same ribosomes and polymerases, so the prediction assumes no load — which is optimistic in the direction that costs an experiment.

Is anything uploaded?

No. The libraries ship with the app and the whole design runs in the browser.