Template Generation: gBlocks, Codon Optimization, Universal PCR · WeaveSeq

Template Generation: gBlocks, Codon Optimization, Universal PCR

Objective

Produce one IVTT-ready linear dsDNA template per binder for the top ~96 in-silico candidates of a campaign: E. coli-codon-optimized, carrying T7 promoter, Shine-Dalgarno (SD), Kozak, N-terminal FLAG tag, double stop, and T7 terminator. Strategy: order 96 individual IDT gBlocks (Strategy B) rather than re-using the existing yeast oPool (Strategy A — unsafe in E. coli systems, Section 2) or ordering a fresh pool (Strategy C — keeps the deconvolution burden). Templates feed SOP 04 — IVTT Expression; the FLAG tag is the handle for the HTRF donor arm of the sandwich in SOP 05 — TR-FRET Plate Assay.

1. Strategy Selection

The binder DNA already exists as an IDT oPool of single-stranded oligos (yeast-codon-optimized CDS, no stop codon, pCTCON2 homology arms) built for the YSD lane — see YSD SOP 03 for its exact structure (5′ arm 50 bp + CDS + c-Myc 30 bp + TAATAG + 14 bp downstream). Its reuse for IVTT was evaluated and rejected for E. coli systems on codon usage (Section 2). Three strategies were compared:

	A — reuse existing oPool	B — 96 individual gBlocks	C — fresh E. coli oPool
DNA cost	primers $129-134 (112-mer Ultramer) or ~$20-40 (two-step PCR)	$2,150-2,765	$400-1,000 (verify; quote-only)
Deconvolution	Required: $750-2,000, +3-5 days	None (one tube per binder)	Required: same as A
Codons	Yeast MFC: AGA at every Arg — unsafe in E. coli IVTT (Section 2)	E. coli MFC: zero rare codons	E. coli MFC: zero rare codons
Lysate constraint	Rabbit reticulocyte only (mammalian tRNA pool); TnT L1170 discontinued	None (PURExpress/PUREfrex)	None
Synthesis turnaround	In hand	2-4 business days (≤750 bp tier)	4-7 business days + deconvolution
Verdict	Reject for the primary lane (documented as fallback, Section 7.1)	PRIMARY	Revisit only if panel grows to ≥384

1.1 Why amplify the gBlocks at all

gBlocks arrive at 250-1,000 ng each — enough for dozens of 10 µL mini-reactions directly. A single universal-primer amplification still pays because it: (1) banks a master stock so repeat campaigns cost ~$100 of PCR reagents instead of ~$2.8k of DNA (top cost lever, see SOP 09); (2) produces the exact linear-template format NEB documents for PURExpress (T7 promoter + RBS + start/stop + terminator, ≥6 bp after the stop); (3) yields a sequencing amplicon for Sanger QC.

2. Codon Caveat: Why the Yeast oPool Cannot Be Re-Primed

The existing oPool CDS is yeast-codon-optimized by AffiniBind's "most frequent codon" (MFC) strategy, and the yeast table maps R → AGA (_SCEREVISIAE_CODONS in affinibind/ordering/ysd/codon_optimizer.py, verified locally). Local analysis of 29 real campaign binders from candidate_table parquet:

45 AGA codons across 1,933 total codons (2.3%); 19/29 binders contain ≥1 AGA; 13/29 contain ≥2; 4/29 contain direct Arg-Arg (AGA-AGA) runs.
AGA/AGG are the rarest E. coli codons: ~5-8% of Arg codons in E. coli (Gurvich 2005; Kazusa K-12) vs 42.7% in human and 69.0% in yeast (Kazusa). They are <0.5% only as a fraction of all codons.
Chumpolkulwong et al. 2006 (PMID 16703415): AGA/AGG (with AUA, CCC, CGG) severely depress expression in E. coli — <5% of control yield in E. coli cells; the same paper shows cell-free yields are "appreciably higher" than in cells, so the exact IVTT penalty is unquantified, but severe reduction is well documented.
Tandem AGG-AGG can induce up to 50% +1 ribosomal frameshifting in highly expressed mRNAs (Spanjaard & van Duin 1988); in native contexts 1.5-11% (Gurvich 2005, PMID 15937165).

Why this is a false-negative trap here: detection is N-terminal FLAG, so premature stops or frameshifts anywhere in the CDS still report a signal — roughly two-thirds of the panel carries at least one AGA and would produce truncated protein that fails to bind, masquerading as non-binding. Rabbit reticulocyte lysate would translate the yeast CDS fine (AGA/AGG are common mammalian codons) but the standard RRL kit (TnT T7 Quick L1170) is discontinued and RRL hemoglobin interferes with TR-FRET (US Patent 7,150,973; see SOP 01). Conclusion: re-optimize the CDS for E. coli (Strategy B).

3. Construct Design (Strategy B)

Fragment layout 5′ → 3′ (mean 320 bp, range 273-360 bp for the 51-80 AA binders of the sample campaign; fixed flanks = 61 nt 5′ + ~59-60 nt 3′):

5'-TAATACGACTCACTATAGGG  GG  AGGAGG  GCCACC  ATG  GACTACAAAGACGATGACGACAAG  <binder CDS>  TAATAG  GTCGAC  TAGCATAACCCCTTGGGGCCTCTAAACGGGTCTTGAGGGGTTTTTTG -3'
   └─ T7 promoter (20 nt) ┘ └SD┘  └Kozak┘ └start┘└─ FLAG = DYKDDDDK (24 nt) ──┘ └─ E. coli MFC ─┘ └2×stop┘ └6 bp─┘ └───── T7 terminator (47 nt) ─────┘

The expressed ORF is M + DYKDDDDK + <binder AA sequence> — an N-terminal FLAG fusion whose detection handle is the anti-FLAG M2 donor arm of the HTRF sandwich (SOP 05). The CDS is back-translated from the sequence (AA) column of candidate_table.annotated.parquet using the E. coli MFC table (Section 3.2).

3.1 Design Rules (NEB PURExpress E6800 manual, verified verbatim)

Element	Requirement	This design
T7 promoter	20-100 nt upstream of the CDS	`TAATACGACTCACTATAGGG` (20 nt), immediately 5′ of the GG
Ribosome binding site (SD)	~6-8 nt upstream of the ATG	`AGGAGG` with `GCCACC` (6 nt) between SD and ATG → SD 3′ end sits 6 nt from the start codon (inside the window)
Kozak context	Not required by E. coli systems	`GCCACC` gives a vertebrate Kozak context, making the same construct work in rabbit reticulocyte lysate (dual compatibility; Section 7.1)
Start / stop codons	Required	`ATG` … `TAATAG` (TAA TAG tandem)
Post-stop spacer	≥6 bp downstream of the stop codon (PCR templates)	`GTCGAC` (6 bp)
T7 terminator	Recommended for PCR templates	47-nt canonical terminator (Section 3.3)
Template amount	Starting concentration 250 ng per 25 µL reaction; valid range 25-1,000 ng	SOP 04 scales this to 10 µL mini-reactions

3.2 E. coli MFC Codon Table

AA	Codon	AA	Codon	AA	Codon	AA	Codon
A	GCG	C	TGC	G	GGC	P	CCG
R	CGT	Q	CAG	H	CAT	S	AGC
N	AAC	E	GAA	I	ATT	T	ACC
D	GAT	K	AAA	L	CTG	V	GTG
F	TTT	M	ATG	W	TGG	Y	TAT

Re-optimization with this table produces zero rare codons — a scan for AGG/AGA/CGG/CGA/ATA/CTA/CCC/GGA/TTA across all 29 re-optimized sample CDSs returned 0 hits (local computation). Code note: codon_optimizer.py currently ships only the yeast table; the E. coli table must be added there as part of the template-builder script (Section 5.1).

Verification (2026-08-18). The table was checked against the primary references; no codon changes were warranted, and the rationale is documented here so future edits are evidence-bound:

The table reproduces the Kazusa E. coli K-12 whole-genome MFC table on 19/20 amino acids (K-12/MG1655, taxid 316407, as bundled in python_codon_tables e_coli_316407.csv; kazusa.or.jp/codon).
Arg CGC vs CGT is a near-tie genome-wide (CGC 0.40 vs CGT 0.38 in K-12/MG1655; 20.8 vs 20.2 ‰ in O157:H7), but CGT is the clear winner in the classic Hénaut & Danchin (1996) class II (highly expressed genes) table (64.25 vs 32.97; values per Maduro, UCR). The current R = CGT is therefore kept — it is the better-cited of the two defensible choices.
The class II table disagrees with this table on 8 amino acids (F→TTC, I→ATC, V→GTT, S→TCT, Y→TAC, H→CAC, D→GAC, G→GGT). These are not adopted, because: (a) for the wobble-pair families (F, I, Y, H, D, and GGC/GGT) both codons are decoded by the same tRNA, so there is no tRNA-availability advantage in PURE; (b) for V and S the two references are split and no PURE-specific per-codon data favours the class II choice; (c) the PUREfrex vendor's own worked example uses whole-genome "major codons" (GAA/GTG/CAG/CTG — GeneFrontier ConBio2017 handout) that match this table exactly.
PURE-system basis for an E. coli table at all: PUREfrex uses a tRNA mixture prepared from E. coli, so each tRNA level mirrors E. coli (GeneFrontier FAQ Q12 — "design template DNA with codons matching the codon frequency of E. coli"). NEB's PURExpress FAQ warns that many rare codons (or a skewed amino-acid composition) leave full-length product a minor species and that supplementing the "missing" tRNA may help. Li et al. 2017 found the PURE system is tRNA-limited (adding tRNAs raised yield ~1.5-2×), and Doerr et al. 2021 measured 5-50× elevated per-codon processivity loss in PURE — both consistent with "avoid rare codons", which this table does.

Rare-codon list validation (2026-08-18). The §5.2 scan list (AGG/AGA/CGG/CGA/ATA/CTA/CCC/GGA/TTA) was checked against the cell-free literature. Chumpolkulwong et al. 2006 measured five of them as severely depressing E. coli cell-free expression: AGA, AGG, ATA, CCC, CGG. The remaining four (CGA, CTA, GGA, TTA) were tolerated in their S30 extract but are inefficient in cells and depend on minor tRNAs — and the PURE system is more tRNA-limited than an S30 extract (Li et al. 2017; Doerr et al. 2021) — so they are retained as conservative exclusions (excluding them costs nothing, since the MFC table never emits them). TGT and GGG were considered and not added: Cys is decoded by a single wobble tRNA (so TGT carries no tRNA penalty) and the MFC table emits neither codon. The 9-motif list is therefore kept unchanged.

3.3 Fixed Element Sequences (Exact)

Element	Sequence (5′ → 3′)	Notes
T7 promoter	`TAATACGACTCACTATAGGG`	Doubles as the universal forward primer
GG spacer	`GG`	2 nt; keeps the T7 +1 transcript start clean
SD	`AGGAGG`
Kozak	`GCCACC`
Start	`ATG`
FLAG	`GACTACAAAGACGATGACGACAAG`	Canonical FLAG epitope DYKDDDDK; no dibasic OmpT cleavage site
Double stop	`TAATAG`
3′ spacer	`GTCGAC`	6 bp
T7 terminator	`TAGCATAACCCCTTGGGGCCTCTAAACGGGTCTTGAGGGGTTTTTTG`	47 nt
Universal R primer	`CAAAAAACCCCTCAAGACCCGTTTAGAGGCCCCAAGGGGTTATGCTA`	Reverse complement of the terminator
Optional c-Myc	`GAACAAAAGCTTATTTCTGAAGAGGACTTG`	EQKLISEEDL (30 bp), inserted between CDS and TAATAG as a free second detection tag; adds ~$2.7/fragment at $0.09/bp (sequence identical to the YSD 3′ arm, YSD SOP 03)

Because all 96 + control fragments share identical 5′ and 3′ flanks by construction, one universal primer pair amplifies the entire panel.

4. Materials

Item	Vendor	Catalog #	Price (2026)	Notes
gBlocks gene fragments, 125-3,000 bp (tubes)	IDT	n/a (order item)	$0.09/bp (budget; "from $0.07/bp" all-inclusive launch pricing may apply to eBlocks)	250-1,000 ng/fragment; 2-4 business days for ≤750 bp (5-8 BD >1,000 bp); median error rate <1:5,000 bp; plate format has a 24-fragment minimum — use tubes
Gene Fragments, 0.3-0.5 kb	Twist Bioscience	n/a	$35 flat/fragment ($0.07/bp applies to 0.5-1.8 kb only)	Backup; express 2-4 BD; error rate 1:7,500
Gene Fragments, 200-500 bp	GenScript	n/a	$0.07/bp, $35 minimum charge for <500 bp; 5-7 BD	Fallback; institutional contracts ~20% off (PSU 2024-25 contract)
oPools oligo pool (Strategy C only)	IDT	n/a	quote-only; budget $400-1,000 (verify)	40-350 nt, 4-7 BD, error <1:2,000; minimum pool size ambiguous (~100 per third-party comparison vs "no minimum" per IDT)
Ultramer 112-mer (Strategy A forward primer)	IDT	n/a	~$1.15-1.20/base at 4 nmol (McMaster facility resale) → ~$129-134	Direct IDT Ultramer pricing: (verify)
Standard 25 nmole DNA oligos (universal F/R pair)	IDT	n/a	$0.25-0.80/base (verify); ~$20-40 per pair	IDT list pricing is login-based
Q5 2× Master Mix	NEB	TODO(verify)	TODO(verify)	High-fidelity PCR enzyme for universal amplification
DNA Clean & Concentrator-5	Zymo	TODO(verify)	TODO(verify)	Post-PCR cleanup
Qubit dsDNA HS Assay Kit	Thermo Fisher	TODO(verify)	TODO(verify)	Template quantification
Sanger sequencing	core facility	n/a	$4-8/rxn	UChicago core $4.95 internal / $8.47 external; UTHSC $5; GenHunter $8
Amplicon-EZ barcoded NGS (deconvolution option, Strategy A/C)	Azenta/Genewiz	n/a	from ~€49/sample	8 business days
DNA LoBind tubes, 1.5 mL	Eppendorf	022431021	$57.79 / 250	For fragment/stock dilutions
UltraPure DNase/RNase-free water	Thermo Fisher

Full bill of materials and one-time vs per-campaign split: SOP 02 and Appendix.

5. Step-by-Step Protocol (Strategy B)

5.1 Generate the Template Table (AffiniBind Integration)

Input: the campaign's candidate_table.annotated.parquet, which contains the candidate_uid and sequence (AA) columns (verified in affinibind/schema/_candidates.py — CANDIDATE_TABLE_ANNOTATED_SCHEMA; affinibind/ordering/ysd/cli.py already requires the sequence column and back-translates AA → DNA for oligo ordering).

Proposed script affinibind/scripts/trfret/build_templates.py — does not exist yet (planned module; scripts/ currently contains design/export/filter/ fingerprint/generation/ordering/prepare/publishing/rescore/shared — the scripts/trfret/ subdirectory follows that convention):

Read candidate_table.annotated.parquet (see affinibind/ingest/_reader.py for discovery of candidate_table.\{annotated,complex,base\}.parquet).
Back-translate the sequence column with the E. coli MFC table (Section 3.2 — add the table to affinibind/ordering/ivtt/codon_optimizer.py).
Assemble the full construct per Section 3.
Run the in silico QC checks (Section 5.2).
Emit template_table.tsv — columns: candidate_uid, aa_len, fragment_len, fragment_sequence, gc_pct, rare_codon_count — plus an IDT order manifest (Pool name | Sequence convention, as in affinibind/ordering/ysd/idt_writer.py).

The template table is a local working artifact; record fragment lot numbers and QC outcomes in the WeaveSeq experiment notes when results are pushed (SOP 07).

5.2 In Silico QC Before Ordering (Mandatory)

Check	Method	Gate
Reading frame	Translate all 96 fragments in silico	0/96 frame or premature-stop errors; ORF = M-DYKDDDDK-<binder>-STOP
FLAG present	Sequence scan	96/96 at the N-terminus, immediately after ATG
Double stop	Sequence scan	96/96 `TAATAG` before the 3′ spacer
Rare codons	Scan AGG/AGA/CGG/CGA/ATA/CTA/CCC/GGA/TTA	0 hits
Internal motifs	Scan CDS for T7 promoter/terminator runs	0 hits
Length sanity	Expected range for 51-80 AA binders	273-360 bp

5.3 Order the Fragments

Order 96 binder gBlocks + 2-3 control gBlocks (Section 6) from IDT, tubes format, 250-1,000 ng each. Cost: ~320 bp × $0.09/bp ≈ $28.8/binder; 96 fragments ≈ $2,765 (range $2,150-2,765 across the $0.07-0.09/bp tier). Budget at the $0.09/bp end (per the Elegen comparison, March 2026, standard gBlocks list at $0.09/bp).
Obtain a Twist quote in parallel as backup/leverage (flat $35 per 0.3-0.5 kb fragment → $3,360 for 96, but the lower error rate 1:7,500 and express 2-4 BD make it the preferred re-order vendor); GenScript as second fallback (~$3,360 list; ~$2,688 with the PSU-style 20% contract discount).
Plan for synthesis errors: at 1:5,000/bp, a ~320-bp fragment has ~6% probability of ≥1 error → expect ~6 of 96. Single-point screen wells tolerate this; dose-response binders must not (Section 5.7).

5.4 Resuspend the Fragments

Briefly centrifuge each tube (10 s) to collect the pellet.
Resuspend in 10 mM Tris pH 8.0 (IDTE) to ~10 ng/µL (verify).
Pipette up and down 10×; if a visible pellet remains, heat 55°C 5 min, vortex, spin down; equilibrate 30 min at room temperature.
Store at −20°C in DNA LoBind tubes; aliquot to avoid >3 freeze-thaw cycles.

5.5 Universal PCR Amplification

One primer pair amplifies all 96 + controls. Reaction (50 µL, per fragment):

Component	Volume
Q5 2× Master Mix	25 µL
10 µM universal F primer (T7: `TAATACGACTCACTATAGGG`)	2.5 µL
10 µM universal R primer (rev-comp terminator, Section 3.3)	2.5 µL
gBlock (1-10 ng)	X µL
Nuclease-free water	to 50 µL

Thermocycler program:

Step	Temp	Time	Cycles
Initial denaturation	98°C	30 s	1
Denaturation	98°C	10 s
Annealing	60°C	15 s	12-15
Extension	72°C	30 s
Final extension	72°C	2 min	1
Hold	4°C	∞

Use the minimum cycle count that gives robust yields (≤15 — every extra cycle accumulates polymerase errors and, for mixed templates, bias; each well here is a single fragment, so bias is not the issue, error accumulation is). Run in a 96-well PCR plate with 96 fragments + 2 controls + 1 no-template well per plate.

5.6 Cleanup, Quantification, Normalization

Clean up with Zymo DNA Clean & Concentrator-5; elute in 20 µL nuclease-free water.
Quantify by Qubit dsDNA HS; optionally check A₂₆₀/A₂₈₀ = 1.8-2.0.
Normalize to a common working concentration and bank master stocks (−20°C, DNA LoBind tubes, ≤3 freeze-thaws). The IVTT dosing target is 250 ng per 25 µL reaction (25-1,000 ng range) — SOP 04 scales this to 10 µL mini-reactions (equivalent 10 ng/µL in-reaction).
Re-amplification economics: the banked amplicon stocks make repeat campaigns ~$100 of PCR reagents instead of ~$2.8k of new DNA (SOP 09).

5.7 Sanger QC

Sequence the top ~12-20 candidates (the dose-response set plus any screen outliers) using the universal primers, ~$4-8/sample (budget $72-160).
Gate: exact match to the designed sequence. Any mismatch → re-order that fragment from the backup vendor (Twist) or exclude the binder from the titration set and flag it in WeaveSeq.
Optional full-panel QC: 96 × ~$6 ≈ $576.

6. Control Templates

Order as 2-3 extra gBlocks with the same flank design (~$100-130 total):

Control	Content	~Size	Purpose	~Cost
POS1	FLAG + core streptavidin CDS	~540 bp	Binds the biotinylated target directly (core streptavidin retains biotin binding, Kd ~10⁻⁷ M — Sano & Cantor, PNAS 1995) → maximal HTRF sandwich signal; validates expression, FLAG detection, and the full FRET geometry end-to-end	~$38-49
CTRL	FLAG + eGFP CDS	~870 bp	Expression and donor-arm control; negative control for target binding	~$61-78
(spare)	duplicate POS1	~540 bp	Re-order insurance for the most-used control	~$38-49

Assay-level negative wells — no-template lysate, FLAG-eGFP + biotinylated target, free FLAG peptide (DYKDDDDK) competition at 100 µg/mL, and biotin-BSA acceptor-only — are plate controls defined in SOP 05.

7. Fallback Strategies (Documented, Not Primary)

7.1 Strategy A — Re-Prime the Existing oPool (Rabbit Reticulocyte Only)

Only viable if fragment synthesis is impossible and an RRL system is procured. Caveats: TnT T7 Quick L1170 is discontinued at Promega (replacement TnT T7 Quick for PCR DNA L5540, 40 rxns, price login-gated — TODO(verify)), and RRL hemoglobin interferes with TR-FRET readouts (US Patent 7,150,973; SOP 01).

Frame math (mandatory 1-nt spacer). FLAG (24 nt) + the pCTCON2 5′ arm (50 nt, affinibind/ordering/ysd/oligo_builder.py DEFAULT_5PRIME_ARM) = 74 nt ≡ 2 mod 3 → the CDS would be out of frame. Inserting a single G spacer between FLAG and the arm restores the frame; the 50-nt arm then translates as a ~17-AA GS-rich flexible linker. Verified in silico: 0/29 frame or stop errors with the spacer (local computation).

112-nt forward Ultramer (T7 + GG + SD + Kozak + ATG + FLAG + G spacer + 5′ arm):

TAATACGACTCACTATAGGG  GG  AGGAGG  GCCACC  ATG  GACTACAAAGACGATGACGACAAG  G  CTAGTGGTGGAGGAGGCTCTGGTGGAGGCGGTAGCGGAGGCGGAGGGTCG
└─ T7 (20 nt) ──┘  └SD┘ └Kozak┘ └ATG┘ └── FLAG (24 nt) ──┘  └1-nt┘ └──── pCTCON2 5′ arm (50 nt) ────┘

Reverse primer: reuse the existing RSPE primer from the YSD campaign (already in hand): TATCAGATCTCGAGCTATTACAAGTCCTCTTCAGAAATAAGCTTTTGTTC (YSD SOP 03).
The oPool oligo already carries the TAATAG double stop 14 bp before its 3′ end → the ≥6 bp post-stop requirement is satisfied without modification.
Extension/overlap PCR: 112-mer Ultramer (~~$129-134 at 4 nmol via McMaster pricing, $1.15-1.20/base; direct IDT pricing (verify)) + RSPE on the pooled ssDNA; a two-step PCR with short 25-nmole primers (~~$20-40 total) is the budget alternative (verify).
Deconvolution (unavoidable): oPools ship as one mixed dry pellet. Limiting-dilution single-molecule PCR + Sanger of ~~150-400 occupied wells (~~$4-8 each) — budget $750-2,000, +3-5 days — or bulk PCR + cloning + one barcoded Amplicon-EZ run (from ~€49/sample). Poisson note: recovering all 96 species of a 384-oligo pool at 0.3 molecules/well needs ~3,800 wells.
Cost framing (per-reaction vs per-µL): at standard 50 µL scale TnT is ~$23/rxn (L1170 $929.79/40 rxn), vs NEBExpress ~$20-24.6/rxn and PURExpress ~$27.7-31.6/rxn — no meaningful per-reaction premium. Per µL, TnT is ~~$0.46/µL ($929.79/2,000 µL) — cheaper than PURExpress ($1.11-1.26/µL) and comparable to NEBExpress (~~$0.49/µL); at the 10 µL mini-rxn scale TnT would cost ~$4.65 vs ~$2-9 for E. coli lysate systems at that scale. The real costs of Strategy A remain the Ultramer, deconvolution, and the hemoglobin-background risk — not the lysate.

7.2 Strategy C — Fresh E. coli-Optimized oPool

96-100 oligos × ~310 nt at 50 pmol: quote-only, budget $400-1,000 (verify); 4-7 business days; error <1:2,000; minimum pool size ambiguous (~100 per third-party comparison vs "no minimum" per IDT's launch PR).
Still requires the full deconvolution of Section 7.1 → strictly dominated by Strategy B at 96-plex. Revisit only if the panel grows to ≥384 binders.

8. QC Checkpoints

Stage	Method	Gate
Before ordering	In silico translation of all 96 (Section 5.2)	0 frame/stop errors; FLAG + TAATAG present; 0 rare codons
Order manifest	Diff vendor file vs template table	99/99 sequences match
Post-PCR	Qubit dsDNA HS	Stock sufficient for 250 ng/25 µL IVTT dosing per binder (verify)
Post-PCR	Agarose gel 1.5-2% (verify)	Single band at expected size ±10%
Sanger (top ~12-20)	Core facility	Exact match to design; else re-order/exclude
Storage	−20°C, LoBind tubes	≤3 freeze-thaw cycles

9. Costs and Turnaround

Strategy B all-in DNA cost (~$27-34 per binder):

Line item	Cost
96 gBlocks (~320 bp × $0.09/bp)	$2,150-2,765
2-3 control gBlocks	$100-130
Universal primer pair	$20-40 (verify)
PCR reagents (Q5 MM + cleanup + tips/plates)	~$150-200
Sanger QC, top 12-20	$72-160
Total	~$2,600-3,300

The cost-model scenario B line is $3,117 for template DNA (96 gBlocks $2,765 + controls $130 + PCR reagents $150 + Sanger top-12 $72) — see SOP 09 for the full campaign budget.

Turnaround (matches SOP 09 §2-3 — template arrival to Kd answer = 4 days; Sanger is post-hoc QC with a veto on the Day-3 re-expression):

Milestone	Day
Order gBlocks (+ control fragments)	0
Fragments arrive (2-4 business days, ≤750 bp tier)	2-4
Fast-lane Day 1: universal PCR + cleanup + QC + Sanger submitted (top 12-20)	arrival + 1
Fast-lane Day 2: 96-well IVTT + single-point TR-FRET screen (SOP 04, SOP 05)	arrival + 2
Fast-lane Day 3: re-express top 12-20 (25 µL) + 12-point Kd titrations (Sanger veto in hand)	arrival + 3
Fast-lane Day 4: Kd fits + WeaveSeq push	arrival + 4

End-to-end sequence → answer: ~6-8 calendar days (fragments 2-4 BD + 4-day fast lane); plan ~8-10 with contingencies. Sanger results land on fast-lane Day 2-3 — use them to veto any erroneous dose-response template before the Day-3 re-expression.

Optional pilot: order 3-4 binder gBlocks + POS1 + CTRL first and validate them end-to-end through SOPs 04-05 before committing the full 96 (~+2-3 days). Recommended for the first campaign; skippable thereafter.

10. Troubleshooting

Symptom	Likely Cause	Action
No amplicon for an individual fragment	Synthesis error (~6% per fragment at 1:5,000/bp) or lost pellet	Verify resuspension (spin, 55°C); re-order from the backup vendor; mark binder "excluded — template QC" in WeaveSeq
Multiple bands on gel	Primer mis-priming on GC-rich CDS	Raise annealing to 62-65°C (verify); reduce to 12 cycles; gel-purify the correct band
Low expression across the whole panel (SOP 04 QC)	Construct design error	Re-run the Section 5.2 checks; confirm FLAG is in frame and the codon table matches Section 3.2
Truncated products on anti-FLAG QC blot	Frameshift from a synthesis error	Sanger that binder; re-order or drop it
A₂₆₀/A₂₈₀ outside 1.8-2.0	Cleanup carryover	Re-clean with DNA Clean & Concentrator-5; blank against the elution buffer
CTRL (FLAG-eGFP) expresses but most binders do not	Unexpectedly — check for residual rare codons or strong secondary structure in the CDS	Re-run the rare-codon scan; redesign problem CDSs with the MFC table or switch to the Twist sequence

11. References

NEB PURExpress E6800 manual — template requirements: T7 promoter 20-100 nt upstream, RBS ~6-8 nt upstream of ATG, start/stop codons, ≥6 bp spacer after stop, T7 terminator recommended; 250 ng/25 µL (25-1,000 ng range). Manual PDF
NEB PURExpress protocol — 37°C, 2 h. Protocol
Chumpolkulwong et al. (2006) — AGA/AGG (with AUA/CCC/CGG) severely depress E. coli expression. J Struct Funct Genomics 7:31-36. PMID 16703415
Spanjaard & van Duin (1988) — tandem AGG-AGG induces up to 50% +1 frameshifting. PNAS. Europe PMC PMC282334
Gurvich et al. (2005) — E. coli AGA/AGG usage and expression penalties in native contexts. J Bacteriol 187:4023-4032. PMID 15937165
Kazusa codon usage database — E. coli K-12, H. sapiens, S. cerevisiae tables. kazusa.or.jp/codon
IDT gBlocks product page — 125-3,000 bp, 2-8 business days, 250-1,000 ng, error <1:5,000. Page
IDT press release (2025-07-10) — all-inclusive gene fragments from $0.07/bp. BusinessWire
Elegen vendor comparison (March 2026) — IDT gBlocks listed at $0.09/bp. Page
Twist Bioscience gene fragments — flat $35 for 0.3-0.5 kb, $0.07/bp for 0.5-1.8 kb, error 1:7,500. Page
Penn State / GenScript 2024-2025 contract — gene fragments $0.07/bp, $35 minimum <500 bp, 5-7 BD, 20% discount. PDF
IDT oPools — 40-350 nt, 4-7 business days, error <1:2,000, quote-only. Page
McMaster Genomics Facility price list — Ultramer 4 nmol $1.15-1.20/base (60-200 nt). Prices

AA	Codon	AA	Codon	AA	Codon	AA	Codon
A	GCG	C	TGC	G	GGC	P	CCG
R	CGT	Q	CAG	H	CAT	S	AGC
N	AAC	E	GAA	I	ATT	T	ACC
D	GAT	K	AAA	L	CTG	V	GTG
F	TTT	M	ATG	W	TGG	Y	TAT

AA	Codon	AA	Codon	AA	Codon	AA	Codon
A	GCG	C	TGC	G	GGC	P	CCG
R	CGT	Q	CAG	H	CAT	S	AGC
N	AAC	E	GAA	I	ATT	T	ACC
D	GAT	K	AAA	L	CTG	V	GTG
F	TTT	M	ATG	W	TGG	Y	TAT

AA	Codon	AA	Codon	AA	Codon	AA	Codon
A	GCG	C	TGC	G	GGC	P	CCG
R	CGT	Q	CAG	H	CAT	S	AGC
N	AAC	E	GAA	I	ATT	T	ACC
D	GAT	K	AAA	L	CTG	V	GTG
F	TTT	M	ATG	W	TGG	Y	TAT