workbench

CHOOSE a set of 4-base Golden Gate/MoClo junction overhangs, rather than scoring one you already have. Maximizes the fidelity of the set's WEAKEST junction against the same published T4-ligase ligation data golden_gate_fidelity scores with, subject to every member actually ligating well — an overhang can score a perfect ratio simply because nothing was ever measured cross-reacting with it, and 94% of the ligation matrix is zeros. Pin the overhangs your vector already commits you to with `fixed`, forbid others with `forbidden`, and give `junctions` when the junctions sit at real positions in real sequence and may only slide a few bases. Reports whether the search was exhaustive (provably the best available) or budget-limited (the best found).

Score a candidate set of 4-base Golden Gate/MoClo junction overhangs against real published T4-ligase ligation-count data: per-overhang specificity, the weakest link in the set, and any risky cross-reacting pairs. Optionally compare against a named published overhang set. This is SeqBench's own transparent scoring methodology — it does not reproduce NEB's/Potapov's own published aggregate fidelity percentages for named sets (their exact formula isn't disclosed anywhere accessible).

Golden Gate as the reaction runs: digest pre-domesticated part plasmids with a Type IIS enzyme and assemble them in the order their OVERHANGS dictate. The fragment released from each part is the one carrying no recognition site (the site goes out with the backbone, which is why a mis-ordered assembly is not re-cut), and the assembly order is an OUTPUT — a set whose overhangs do not close into a single cycle has no product, and the reason is the answer. Distinct from cloning_simulate's `goldengate` method, which does the other job: designing the primers that ADD the sites to BARE parts, assembled in the order you list them.

Parse an HGVS "c." variant description (by gene symbol, RefSeq NM_, or Ensembl ENST accession), convert it to genomic (g.) coordinates via a real, live-fetched Ensembl exon/CDS map (transcripts resolved through the bundled MANE RefSeq<->Ensembl crosswalk), apply 3'-rule normalization to any del/dup/ins, and predict the protein (p.) effect where that is safely computable. Also accepts a genomic "g." position on a chromosome or RefSeqGene (NC_/NG_, GRCh38 or GRCh37): it is validated and placed on GRCh38 by NCBI, projected onto the MANE Select transcript by Ensembl, and converted back by this tool's own engine, which must agree — so the answer rests on two independent conversions. And the "NC_/NG_(NM_…):c." form converts on the transcript in parentheses. A "c." on a RefSeq transcript outside MANE is placed by NCBI and converted on the gene's MANE Select transcript. A single amino-acid substitution ("TP53:p.Pro72Arg") is read as its codon change when exactly one single-base change produces it; otherwise every candidate is listed. Refuses cleanly — rather than guessing — for circular/mitochondrial genomes, RNA-level input, other protein-level input, uncertain/mosaic syntax, and splice-junction-adjacent or inversion protein effects.

Check a UniProt id-mapping job submitted via id_map_submit. Returns {status, ready:false} while still running; once FINISHED, also returns the mapped ids (normalized regardless of which target database was requested) and any ids that failed to map.

Submit up to 100,000 ids to UniProt's ID mapping service for a single confirmed-safe hop (e.g. Gene_Name -> UniProtKB-Swiss-Prot, or UniProtKB_AC-ID -> Ensembl/GeneID/RefSeq_Protein/Gene_Name). Returns a jobId immediately — poll it with id_map_poll.

Percent identity between every pair in a multiple alignment. Takes ALIGNED sequences — the output of multiple_sequence_alignment — because inferring an alignment here would bury the aligner's parameters inside a number that reads as a property of the sequences. Two percentages come back per pair: over columns where both have a residue (the usual figure), and over all columns including gaps. The gap between them is the warning — two sequences overlapping in 50 of 2,000 columns read 100% by the first measure and 2.5% by the second.

Predict PCR products for a template and a pair of primers (IUPAC-aware, allows mismatches, handles circular templates). Primers may carry a non-templated 5' tail — a restriction site, a Gibson arm, a Kozak, a tag: a primer primes on its 3' end, and the tail is carried into the product rather than required to match. start/end are the TEMPLATE-derived span, `length` is the whole product including tails, and `features` marks which product bases came from the oligos (present only when there is a tail). Each end reports annealedLength and tailLength. When two primers make NO product on a template up to 60,000 bp, `diagnosis` says why: where each primer does prime (with the mismatched bases), which copying mistake explains one that primes nowhere (reversed but not complemented, complemented but not reversed, a base dropped or doubled, a mismatched 3' base) with the sequence that would bind, whether the two point the same way or away from each other, and the product a linear reading missed across the origin of a circle.

Design KASP/ARMS allele-specific genotyping primers for a SNP: two allele-specific forward primers differing only at the 3' terminal base (one per allele), each with the standard KASP universal tail (FAM for allele A, HEX for allele B), a deliberate internal ARMS secondary mismatch near the 3' end whose strength complements that primer's own natural allele mismatch (strong↔weak), and one common downstream reverse primer sized to a chosen amplicon range. Because a forward primer reads the antisense strand, each primer's 3' base sits opposite the complement of the other allele, so the two primers get different mismatch classes and are reported separately (graded from the measured PCR yields in Kwok et al. 1990). Reuses the site's nearest-neighbor Tm engine.

Work out how many microlitres of vector and insert to pipette to hit a target molar ratio, from each part's length and stock concentration. Handles one insert or several with independent equivalents (Gibson, Golden Gate, MoClo), reports pmol and ng per part alongside the volumes, and flags the two things that actually go wrong on a bench: a volume below what a pipette measures reliably, and a plan whose DNA does not leave room for buffer and enzyme. A molar ratio is about moles, so a shorter insert at 3 molar equivalents goes in at LESS mass than the vector — that conversion is the point.

Write the differences a whole-plasmid sequencing consensus shows into the map, keeping the map's annotation: every feature is shifted, grown or shrunk with the edit, a feature whose boundary an edit cuts is trimmed (and named), and differences in a known nanopore error context (Dam/Dcm methylation sites, long homopolymers) are left out unless keepArtifacts is set. Returns the updated map as a GenBank record and as SnapGene .dna bytes (base64), plus a change log. Use it when the clone is what you want and the MAP is out of date — a change every clone carries, a redesigned primer, a different allele — not to absorb one clone's mutation. `positions` limits the update to those design positions (e.g. only the changes shared by every clone).

Primer/oligo melting temperature: nearest-neighbor (SantaLucia 1998) at the supplied reaction conditions, recommended from 14 nt up, with the Wallace rule for shorter oligos, a fixed-100 mM-Na+ Schildkraut-Lifson reference estimate, and molecular weights.

Find a motif NOBODY TYPED: given a set of sequences (multi-record FASTA), report the short pattern they share, as a consensus, a position-weight matrix and one site per sequence. This is the opposite question from motif_finder, which searches for a pattern you already have. Gibbs site sampler (Lawrence et al., Science 1993), one occurrence per sequence, both strands by default. ⚠️ READ `found` BEFORE THE CONSENSUS: a Gibbs sampler returns a motif from ANY input, including pure noise, because its objective has a maximum whether or not anything is there — so every run is scored against shuffled copies of the same sequences, which hold the same length and base composition and no arrangement. `found` is true only when the real alignment beats every shuffle outright AND clears the control mean by the margin in `control.zScore`; when it is false the consensus and sites are still returned, so you can see what was rejected. Measured: 0 of 20 pure-noise sets were reported as found, and a 10-mer planted in all 12 sequences was recovered exactly on 5 of 5 seeds. The sampler is deterministic — the same input gives the same answer, so a saved or shared result reproduces; pass a different `seed` to see whether the motif survives a different starting point, because one that does not is weak.

Find (overlapping) occurrences of an IUPAC motif on either strand, allowing mismatches.

Center-star multiple sequence alignment of a multi-FASTA input — nucleotide or protein, detected from the records and reported as `type` — with consensus and per-column conservation.

Choose one primer pair per target so the whole panel works in one tube: no cross-dimer between any two of the primers, every amplicon resolvable from every other on the gel you will run, and one annealing temperature that serves all of them. Searches combinations rather than picking each target's best pair in isolation, which is what makes panels fail — and when no compatible panel exists it names the target pairs that cannot be multiplexed at all, so you know which one to redesign.

Build a set of new genetic parts that match a degenerate (IUPAC) template and share more than a chosen length with nothing — not each other, not themselves, not a background sequence you supply. This is how a toolbox of promoters, RBSs, terminators or sgRNA handles is made large without making an assembly unstable. Honours a GC range and excluded motifs (restriction sites) during the search rather than filtering afterwards. Deterministic given a seed: the same inputs give the same toolbox. Returns fewer parts than asked, with a reason, when the constraints leave no room — it never invents a repetitive one to hit the count.

Given a toolbox of genetic parts, return the largest subset in which no two parts share more than a chosen length of sequence — on either strand. Parts that share a long stretch recombine into each other in a multi-part assembly and are the single largest cause of DNA synthesis failure, and neither shows up when the parts are checked one at a time. Reports every conflicting pair and why each dropped part was dropped. Deterministic: no model, no rate, no score. Use nonrepetitive_parts_design to build new parts instead of selecting from existing ones.

Convert an OD600 reading to a cell density and a total cell count, correcting for the optical path and any dilution, and flagging the two things that silently make the number wrong: a reading above the linear range (a dense culture reads LOW), and a plate-reader path length assumed to be a cuvette's 1 cm. Also plans a back-dilution to a target OD. The cells-per-OD factor is a convention, not a constant — it varies with strain, growth phase and instrument, so pass factorPerOd when you have calibrated against plate counts.

Full oligo analysis: nearest-neighbor Tm/ΔG/ΔH/ΔS plus hairpin and self-dimer screening with base-pair diagrams and warnings.

Minimum-free-energy structure and ΔG for one oligo (hairpin) or two oligos together (homo/heterodimer), using ViennaRNA's published loop model at a temperature you choose — DNA parameters (Mathews 2004) by default, RNA (Turner 2004) on request. Reports each strand alone, the duplex, and the interaction ΔG the two gain by pairing with each other rather than folding alone, which is the number a primer-dimer screen wants. Unlike oligo_analysis's fast stack-sum screen this is a full loop model with bulge, internal-loop and dangling-end terms; the two are on different parameter sets and must not be compared. PREDICTED, NOT MEASURED. No skill statistic is claimed for predicting whether a PCR fails. Loop-model MFE folding reproduces measured structure well for short duplexes and progressively worse with length; the ΔG itself carries roughly kcal/mol-scale uncertainty and the MFE structure is one structure out of an ensemble — request `partition` for the ensemble free energy, which is the more honest single number when several structures compete. Valid for: short oligos, at most 200 nt per strand, at the temperature given. It models two strands in isolation at no particular concentration: it does not know your primer concentration, salt, or cycling program, so it cannot say whether a dimer will actually form in your tube.

Check a list of oligos against the plasmid or template they were designed from, before ordering them. For each oligo: where it primes (3'-anchored, so cloning tails are allowed), on which strand, and which bases disagree with the map. When an oligo primes nowhere it says WHICH copying mistake explains it — reversed without being complemented, complemented without being reversed, one base dropped or doubled, a wrong last base — and gives the sequence that was meant, but only when exactly one such mistake reproduces an oligo that binds. It simulates every product the order can make (up to 15,000 bp) with the in_silico_pcr engine, and for pairs the names declare (GFP_F / GFP_R) says why they make none, including which of the two is written as the wrong strand. Also flags one sequence under two names, one name on two sequences, a primer saved in the map under the same name that sits elsewhere, and 3' self-dimers/hairpins. Up to 96 oligos.

Screen a whole set of oligos you already have — every pair for cross-dimers, every oligo for its own hairpin and self-dimer, and the set for duplicates and Tm spread — and get back the conflicts ranked rather than a table of every combination. This is the pool-level answer cross_dimer gives one pair at a time: 51 primers is 1,275 pairs, which is 1,275 separate calls done by hand and one call done here. Not to be confused with multiplex_panel_design, which DESIGNS primers from templates; this takes the primers you have already ordered. A pairing that involves an oligo's 3' END is judged at a weaker ΔG than one that only pairs internally, because that end is where extension starts — the same two-bar rule the multiplex panel designer uses. Every number is a nearest-neighbor calculation over the sequences supplied, not a prediction of what the reaction will do.

Build a polycistronic operon from a promoter, a list of CDSs with their RBSs, spacers and a terminator — optionally recoding every CDS for a host — then scan the ASSEMBLED molecule for internal promoters, Shine-Dalgarno sequences, terminators, out-of-frame start codons and repeats. Scanning the product rather than the parts is the point: these elements are very often created BY THE JOIN between two parts, so checking each part alone finds nothing. Returns the sequence, an annotated GenBank file, and every element found. Assigns no translation rate — use rbs_library_design to choose RBSs, then rbs_predict to rank the result.

Scan a multi-gene construct, on both strands, for the sequence that quietly breaks operons: promoter-like -35/-10 pairs (including ones pointing backwards, which make antisense RNA), Shine-Dalgarno sequences positioned in front of an internal start codon, terminator-shaped hairpins with a U-tract, out-of-frame start codons inside a declared CDS, and exact direct repeats. Reports what MATCHED and how far it sits from consensus — it does not score a match or claim it transcribes. For an estimated promoter strength use promoter_predict; for a translation rate use rbs_predict.

Look up the orthologous (or paralogous) gene for up to 50 gene symbols in a target species, via Ensembl's homology-by-symbol REST endpoint. Symbols with no homology record are reported in `unmapped`, never silently dropped.

Decompose one Sanger trace into fractions over a set of candidate molecules — the intended construct, the empty backbone, a double insert, a flipped part — instead of onto a generic indel ladder. Non-negative least squares against the candidates' own sequences, so no molecule is ever assigned a negative share. Reports the R² of the fit, which is what says whether the tube holds anything outside the candidate set, and GROUPS candidates the read cannot tell apart rather than splitting their share between them. Feed it the alternatives from assembly_outcomes. PREDICTED, NOT MEASURED. Every run reports its own R²: how much of the observed per-position composition the candidate basis explains. That is a measured adequacy on YOUR trace, and a low value is a statement that the tube holds something the candidate set does not contain. No accuracy against a reference method is published for this implementation, and none is quoted. Exact recovery of synthetic mixtures is deliberately NOT offered as validation: it tests the arithmetic, not whether a real capillary trace behaves like the model. Valid for: One read, from one primer, over a set of candidate molecules that all contain that primer's site and differ from each other within the read. Fractions are reported per GROUP of candidates the read cannot tell apart, and that grouping is part of the answer. NOT valid when the read's anchor identity to a candidate is low (it does not share the primer region), nor when R² comes back low, nor for telling apart candidates that differ only outside the read.

Global (Needleman-Wunsch), local (Smith-Waterman) or semi-global/fitting pairwise alignment of two sequences, with match/mismatch scoring and affine gap costs (Gotoh).

Parse a GenBank flat file into its locus, definition, features and sequence. An EMBL / ENA flat file is accepted too — the same INSDC record in a different layout — and reported with sourceFormat "embl".

Read a GFF3 annotation file: every feature with its 1-based inclusive coordinates, strand, phase, score and decoded attributes, plus any sequences from an embedded ##FASTA section. GFF3 usually carries NO sequence — it annotates a separate FASTA, joined on the first column — so the result says which of the two it was handed. Give `sequence` (or let an embedded ##FASTA supply it) to also get the record as GenBank, readable by every other tool here. Percent-escapes are decoded, `.` is read as absent rather than zero, and a feature repeated across lines keeps both lines.

Decode a Sanger ABIF (.ab1 / .abi) chromatogram: base calls, per-base quality, the four dye-channel traces, peak locations, and the run's own labels (sample name, well, plate, instrument, run start).

Read a SnapGene .dna file: sequence, topology, every feature with its span, strand, display color and qualifiers (spliced and origin-spanning features kept as such), and the saved primer list.

Read a PDB or mmCIF structure file — an RCSB PDB entry, or a model from AlphaFold DB, AlphaFold Server or ColabFold — and report what is in it: every polymer chain with its molecule type, name and one-letter sequence (all chains also as FASTA), how many residues have coordinates and which stretches of the sequence do not, bound ligands and waters, and the experimental method and resolution. When the file itself marks its B-factor column as a confidence score (a pLDDT metric, an AlphaFold title, or one value per residue with no experimental method), each chain's mean pLDDT and its stretches below 50 are reported too; otherwise B-factors are left alone. Only the first model of a multi-model file is read. Reads the text; nothing is fetched.

Search a parts list harvested from the annotated features of the vector library — promoters, terminators, RBSs, polyA signals, origins, selection markers, affinity tags, reporters, linkers/MCSs — by name, kind or length. Nothing here is transcribed: every part is the exact sequence a GenBank record annotated, and each hit carries the accession and 1-based span it was cut from, plus every other library vector the same part was found in. Parts whose location is spliced or approximate are excluded, because their sequence is not fully determined.

A neighbour-joining tree from a multiple ALIGNMENT, as Newick. Distances are corrected pairwise identities over the columns where both rows have a residue — Jukes-Cantor for nucleotides, Poisson for protein, the same corrections MEGA applies by default — and the tree is rooted at the midpoint of its longest tip-to-tip path. Branch support is Felsenstein's bootstrap over alignment columns, written into the Newick as the usual internal-node labels and returned per split; it needs four or more sequences and is refused with a reason past the work one call will do, rather than returned from fewer replicates under the same name. Takes the output of multiple_sequence_alignment: inferring an alignment here would bury the aligner's parameters inside a tree.

Auto-detect common cloning features (promoters, tags, origins, resistance markers, MCS, primers) on both strands. Signatures under 20 bp must match exactly; longer ones tolerate up to ~10% mismatches so point mutants still annotate — each feature reports its own `mismatches` count and an `exact` flag.

Annotate a plasmid against pLannotate's open-source feature library — a much larger signature set (GenoLIB parts + Swiss-Prot + FPbase + Rfam, cross-referenced against ~195k Addgene-deposited plasmids) than plasmid_annotate's built-in curated list, and it reports partial and low-identity hits as graded alignments rather than the pass/fail signature match plasmid_annotate does (that one is not exact-only either — signatures of 20 bp or more tolerate up to ~10% mismatches — but it reports a hit or nothing, with a `mismatches` count and an `exact` flag). Each feature here carries its percent identity, reference coverage and a fragment flag so you can judge a weak hit. Runs a multi-second search on a shared service and is therefore rate limited (see 429/503); use plasmid_annotate for an instant, unmetered first pass.

Answer the question someone holding an unlabelled tube actually has, rather than returning another feature list: which antibiotic to select on, which host will carry it and at roughly what copy number, which curated backbone it resembles, and what on it will bite you later (duplicate markers, two origins for the same host, internal Type IIS sites that stop it being a Golden Gate part, long direct repeats a recA+ host can recombine out). Reads features from an annotated GenBank record when you paste one, otherwise runs the built-in signature scan. EVERY ABSENCE TRAVELS WITH THE VOCABULARY IT WAS DECIDED AGAINST: an empty marker list means nothing matched the families this screen knows, never that the plasmid cannot be selected — run plasmid_deep_annotate, whose library is far larger, before concluding that.

One combined view of 'what is this plasmid': recognized common features (from plasmid_annotate), backbone identity / possible chimera (from plasmid_identify), and — the two crossed together — any region that neither a curated backbone nor a recognized common feature explains. That last list is a triage signal (an unusual insert, an unannotated part, or worth a closer look), not a defect finding: a real gene-of-interest legitimately has no curated-feature match.

Screen a query plasmid against a small curated set of common backbones (cloning vectors, expression vectors, BACs — see referencesChecked for the exact list) to identify which one(s) it resembles, separate an unmatched region (normal — your own insert) from a POSSIBLE CHIMERA (a region matching a different known backbone than its neighbor), and report per-match %identity/%coverage. NOT a search against Addgene's ~100k-plasmid catalog or PlasmidScope's 850k+ — a curated-set screen only.

Design SpCas9 prime-editing pegRNAs for a substitution, insertion, deletion, or small replacement: for each usable NGG PAM it builds the spacer, a primer-binding-site (PBS) length sweep targeting a ~30 C melting temperature, the reverse-transcriptase template (RTT) that encodes the edit, and the full 3' extension, plus PE3 nicking-sgRNA suggestions 40-90 bp away on the opposite strand. Designs where the edit destroys the pegRNA's own PAM (preventing re-nicking of the edited allele) are ranked first. Coordinates: every pegRNA coordinate (protospacer span, nick position, editStart/editEnd) is 1-based inclusive in the submitted PRE-EDIT target's frame — the protospacer+PAM search runs on the unedited sequence, because Cas9 has to bind the allele you actually have. The one exception is edit-dependent PE3b nicking guides, which exist only once the edit is installed; each nickingGuides entry therefore carries a `coordinateFrame` field of "target" or "editedSequence" naming the frame its own start/end/nickToNickDistance are measured in, and for a length-changing edit the two frames differ downstream of the edit. Off-target activity is not evaluated (no in-browser reference genome).

Predict per-pegRNA prime-editing efficiency for one edit with PRIDICT2.0, and return the top-scoring pegRNA designs ranked by it. Takes the target as context, the edit in brackets, then context — ACGT...(A/G)...ACGT, with roughly 100+ bp each side — and enumerates PBS/RTT length combinations, scoring every one in HEK293 and K562. Each candidate comes back with both scores, its percentile against the training library, its rank, the spacer, PBS and RTT lengths, the full pegRNA, and Golden Gate cloning oligos. Use it to CHOOSE between designs; the number is not a promised editing percentage. PREDICTED, NOT MEASURED (Spearman ρ = 0.85 on held-out data from the libraries it was trained on). Spearman rho of about 0.85 for intended edits on held-out library data — the best-validated figure of any model in this registry, and roughly double OSTIR's 0.39 on independent data. That figure is still within the library and cell lines it was trained on. Valid for: human sequence, and efficiency ranking within one locus. It is parameterized on HEK293 and K562; your cell type, delivery method, and chromatin context will all move the absolute efficiency, chromatin alone by severalfold. Nothing here is predicted for a non-human host or for editors outside the PE2/PE3 architecture the training libraries used.

Design a twinPE pegRNA pair (Anzalone et al. 2022) for a replacement too large for a single pegRNA's RTT: a left pegRNA nicks the + strand at/before the replacement window and a right pegRNA nicks the - strand at/after it, each synthesizing a new 3' flap; both flaps are truncated at a shared overlap in the middle of the new sequence so they anneal and resolve the edit without an HDR donor. Coordinates: both pegRNAs' protospacerStart/protospacerEnd/nickPosition are 1-based inclusive in the submitted PRE-EDIT target's frame (the PAM search runs on the unedited sequence, on both sides), while replaceSpan is the span of the new content in the returned editedSequence. Off-target activity is not evaluated (no in-browser reference genome).

De-novo PCR primer design (Primer3-style penalty picker): enumerate and score candidate primer pairs against length/Tm/GC/3'-clamp/structure constraints.

Fold the template around each place a primer binds, at the annealing temperature and under DNA parameters, and report how much of the primer's own footprint sits inside a helix. Primer design tools score the oligo — its Tm, its hairpin, its dimers — and leave the template unexamined, while a binding site buried in a stable stem is an ordinary reason a well-designed primer does not amplify. Returns the folded window, its free energy, the paired fraction of the footprint, and how many of the 3'-terminal five bases are paired.

Self-hosted e-PCR-style screen for off-target amplicons predicted by a primer pair against the background chosen with `background`: the bundled lab genomes (the default — E. coli K-12 MG1655, B. subtilis 168, human mitochondrion rCRS, Mycoplasma hyorhinis SK76), or the human or mouse TRANSCRIPTOME, one transcript per gene, where an off-target product means a different GENE. See genomesChecked in the response for what actually ran. No background covers the nuclear genome, so an off-target in an intron, a promoter, an intergenic region or genomic-DNA carry-over is out of scope in every mode. Amplicons are 1-based inclusive on the plus strand; a product across a circular genome's origin reports an end lower than its start and sets wraps: true. This checks background/host-genome specificity, NOT whether the primers hit your intended target — pair it with in_silico_pcr against your own template for that. Each off-target end reports its 3' ANCHOR — the primer's unbroken run of matched bases at the extending end — with that anchor's nearest-neighbor ΔG and a margin against the intended, fully matched reaction, so a site can be told apart by WHERE its mismatches fall rather than only how many there are: one mismatch at the 5' end leaves a site nearly as strong, and one at the 3' base leaves it unable to prime at all. Batchable over candidate REVERSE primers against one fixed forward primer (screen many candidates against a shared partner) — not independent primer-pair batching, which this tool doesn't support. A primer may carry a non-templated 5' tail (a restriction site, a Gibson arm, a tag): the screen looks for a 3'-anchored annealing region as well as a full-length match, so a tailed cloning primer is screened rather than silently matching nothing. Each end's `anchor` is the annealed run, which is the length that matters for extension, and `start`/`end` are measured on the ANNEALED footprints — the bases each primer actually pairs with on the genome — so `length` (the product, tails included) equals end - start + 1 only for untailed primers. Screening a TAILED primer without `intendedTemplate` inflates every margin by the tail's own free energy, because nothing about an oligo says where its non-templated part ends; `intended.basis` reports which footprint the margins rest on.

Build a set of sigma-70 promoters that (a) share no more than a chosen length of sequence with each other, so the library does not recombine with itself once integrated, and (b) span a range of predicted transcription rates, picked as an evenly log-spaced ladder. Variants are constructed deterministically from an IUPAC template holding the -35 and -10 consensus; their strengths are then estimated by the Promoter Calculator model. Rungs with no variant near them are reported as gaps rather than filled with the nearest thing. PREDICTED, NOT MEASURED: strengths carry R^2 = 0.45-0.60 against independent in vivo data, so treat the ladder as a ranked set to screen, not as calibrated numbers. PREDICTED, NOT MEASURED. R^2 = 0.45 and 0.60 against the two INDEPENDENT in vivo datasets the authors tested (Hossain et al., 4,350 promoters, Spearman rho = 0.69; Urtecho et al., 10,898 promoters, rho = 0.67). The widely quoted R^2 = 0.80 is a held-out tenth of the authors' OWN in vitro transcription data and is not the number to plan against: a promoter in a cell is the in vivo case, where between a third and a half of the variance is unexplained. Valid for: sigma-70 (housekeeping) promoters in E. coli. NOT valid for another sigma factor, another organism, a promoter under activator or repressor control, or anything about mRNA stability or translation — rbs_predict is the translation half, and neither speaks to the other.

Scan DNA for E. coli sigma-70 promoters and estimate each one's transcription initiation rate, with the free-energy terms it is built from: the -35 and -10 boxes, the spacer, the discriminator, the extended -10 and the initial transcribed region. Pairs with rbs_predict — together they separate 'nothing is transcribed' from 'it is transcribed and not translated', which no single measurement on the sequence does. Both strands by default, because a promoter reading into your insert from the other strand is still a promoter. PREDICTED, NOT MEASURED. R^2 = 0.45 and 0.60 against the two INDEPENDENT in vivo datasets the authors tested (Hossain et al., 4,350 promoters, Spearman rho = 0.69; Urtecho et al., 10,898 promoters, rho = 0.67). The widely quoted R^2 = 0.80 is a held-out tenth of the authors' OWN in vitro transcription data and is not the number to plan against: a promoter in a cell is the in vivo case, where between a third and a half of the variance is unexplained. Valid for: sigma-70 (housekeeping) promoters in E. coli. NOT valid for another sigma factor, another organism, a promoter under activator or repressor control, or anything about mRNA stability or translation — rbs_predict is the translation half, and neither speaks to the other.

In-silico protease/chemical digestion: cleave a protein and report each peptide's position, length and neutral mass. CNBr masses assume terminal Met becomes homoserine lactone; the sequence retains M and the response labels the modification.

Check an InterProScan job submitted via protein_annotate_submit. Returns {status, ready:false} while still running; once FINISHED, also returns the parsed domain architecture, per-match details, deduplicated GO terms, and `topology` — the MEMBRANE TOPOLOGY and SIGNAL PEPTIDE that EBI's default run includes (Phobius, TMHMM and SignalP), as {kind, start, end, source} with kind one of signal-peptide / transmembrane / cytoplasmic / non-cytoplasmic. Two predictors reporting nearly the same span are listed separately rather than merged, because a merged boundary is one neither produced. ⚠️ EBI runs every SignalP ORGANISM model whatever the protein is, so a human sequence comes back with gram-positive and gram-negative hits too; the source is named in full on every feature and Phobius — which is organism-agnostic and the only one that returns a full inside/outside topology — sorts first.