MASTER_SOP — IVTT + TR-FRET (HTRF) Fast-Lane Binder Validation · WeaveSeq

MASTER_SOP — IVTT + TR-FRET (HTRF) Fast-Lane Binder Validation

Purpose: One bench protocol taking the top ~96 computationally ranked AffiniBind binder candidates (51-80 AA, mean 66.7, ~6-9 kDa) from amplified template to single-point TR-FRET screen and 12-point Kd titrations for the top 12 hits, executed by one technician. All work is BSL-1.

When to use: A ranking round has produced candidate_table.annotated.parquet (columns candidate_uid and sequence). Screen-only (hit/no-hit) = 2 bench days; screen + Kd = 4 bench days. End-to-end from sequence order ≈ 6-8 calendar days (plan 8-10 with contingencies).

Deep dives: every design rationale lives in the reference files in this folder — SOP 01 decisions + decision criteria · SOP 02 buy lists/storage/reader access · SOP 03 template design · SOP 04 IVTT · SOP 05 plate assay · SOP 06 reader · SOP 07 analysis · SOP 08 · SOP 09 · appendix catalog numbers. Essential numbers are all in this SOP; only consult the references for rationale.

2. PREREQUISITES & MATERIALS CHECKLIST

Check off before starting. Steps reference materials by the name in column 1. (verify) = catalog number or price is unverified in the sources — confirm at order time.

#	Item (name used in steps)	Vendor	Cat#	Qty	Used in steps
M1	gBlocks — 96 binder fragments, E. coli MFC codon-optimized	IDT	quote	96 tubes, 250-1,000 ng each	2, 7, 8, 9, 10, 11
M2	Control gBlocks — POS1 FLAG-core-streptavidin (~540 bp), CTRL FLAG-eGFP (~870 bp), spare POS1	IDT	quote	3 tubes	2, 16, 18
M3	Universal primer pair (F = T7, R = rev-comp terminator), 25 nmole	IDT	quote	1 pair	2, 9, 12
M4	Q5 2× Master Mix	NEB	M0492 (verify)	1	9
M5	DNA Clean & Concentrator-5	Zymo	D4013 (verify)	1 kit	10
M6	Qubit dsDNA HS Assay Kit	Thermo	Q32851 (verify)	1	10
M7	10 mM Tris pH 8.0 (IDTE) resuspension buffer	IDT	(verify)	1	7
M8	UltraPure DNase/RNase-free water	Thermo	10977015	500 mL	7, 9, 10, 15, 21
M9	Sanger amplicon sequencing	core facility	—	12-20 rxns	12
M10	PURExpress In Vitro Protein Synthesis Kit, 10 rxn × 25 µL	NEB	E6800S	1 (pilot only)	3, 14
M11	PURExpress In Vitro Protein Synthesis Kit, 100 rxn × 25 µL	NEB	E6800L	1 per campaign	3, 15, 21
M12	PUREfrex 2.0 kit (Sol I/II/III + DHFR control)	GeneFrontier/Cosmo Bio	PF201-0.25-EX	6 kits (backup/lean; price + lead (verify))	3, backup
M13	PUREfrex 2.1 (disulfide-forming variant)	GeneFrontier	PF213-0.25-EX	1 (contingency only)	§8
M14	PURExpress Disulfide Bond Enhancer	NEB	E6820S	1 (contingency only)	§8
M15	96-well full-skirt PCR plate	Axygen / Eppendorf	PCR-96-FS-C / 0030128648	2	15, 16, 21
M16	MicroAmp Clear Adhesive Film	Thermo	4306311	1 pkg	17, 20, 23
M17	Aluminum foil seals	VWR	60941-126 (verify)	1 pkg	alt. sealing
M18	Mini PCR plate spinner	Labnet	MPS 1001	1 (skip if plate-rotor centrifuge exists)	17, 19
M19	HTRF anti-FLAG M2-Tb cryptate donor, 20,000 tests	Revvity	61FG2TLB (renumbering 61FGBTLB — verify)	1 vial	4, 20, 23
M20	HTRF Streptavidin-XL665 acceptor, 20,000 tests	Revvity	610SAXLB	1 vial	4, 20, 23
M21	HTRF detection buffer (HEPES/EDTA/KF)	Revvity	62SDBRDF (200 mL); small pack 62SDBRDD (40 mL)	1	4, 19, 20, 22, 23
M22	384-well small-volume white flat-bottom plate, case of 40	Greiner	784075	1 case	4, 20, 23
M23	Biotin-FLAG peptide (biotin-DYKDDDDK)	NovoPro	319196 (confirm pack size)	1	4, 20, 23
M24	FLAG peptide DYKDDDDK	Sigma	F3290	4 mg	4, 18, 23
M25	MonoRab anti-DYKDDDDK [HRP]	GenScript	A01869 (verify vs A01428)	40 µg	4, 18
M26	Biotinylated target (sulfo-NHS-LC-biotin, 1-3 biotins/protein)	per antigen-prep SOP	—	pre-existing	20, 22, 23
M27	Homebrew-buffer reagents (optional cost lever): HEPES H3375, BSA A-420, Tween-20 P9416, KF (verify), EDTA (verify), Triton X-100 T8787, DTT DTT25	Sigma / GoldBio	as listed	optional	20
M28	PBS tablets	Sigma	P4417	100 tabs	18
M29	Nitrocellulose + ECL substrate + imager	lab stock	(verify)	—	18
M30	DNA LoBind 1.5 mL tubes	Eppendorf	022431021	250	10, 22, 23
M31	Safe-Lock 1.5 mL tubes	Eppendorf	022363204	500	general
M32	Racked universal 200 µL tips	United Scientific	P10148	2 boxes	general
M33	25 mL V-bottom reagent reservoirs	Axygen	14222430	1 case	20, 23
M34	96-well intermediate dilution plate	—	—	2	19, 20
M35	HTRF-certified plate reader (facility, pay-per-use)	facility	—	2 × 2 h blocks	5, 20, 24
M36	Mirror-orientation fallback pair: Streptavidin-Tb + anti-FLAG-d2	Revvity	610SATLB + 61FG2DLB	optional	§8

3. DAY 0 — ORDERING + FACILITY BOOKING (steps 1-6)

Step 1 — Select the campaign set. From candidate_table.annotated.parquet, take the top ~96 binders by ranking (candidate_uid + sequence). If this is the first campaign ever, also pick 3-4 of these for the pilot (step 14).

Step 2 — Order template DNA. 96 binder gBlocks (M1) + 3 control gBlocks (M2) + universal primer pair (M3) from IDT, tubes format. Cost $0.07-0.09/bp = $22-29/binder, $2,150-2,765 total for 96 (budget at $0.09/bp). ≤750 bp ships in 2-4 business days; 250-1,000 ng/fragment; median synthesis error 1:5,000 bp → expect ~6/96 fragments with ≥1 error (single-point screen tolerates this; dose-response set must be Sanger-clean, step 12). Fallbacks: Twist flat $35/fragment = $3,360, 2-4 BD; GenScript $0.07/bp min $35 = $3,360, 5-7 BD.

Construct per fragment, 5′→3′: T7 promoter TAATACGACTCACTATAGGG + GG + SD AGGAGG + Kozak GCCACC + ATG + FLAG GACTACAAAGACGATGACGACAAG + E. coli-MFC CDS + double stop TAATAG + GTCGAC + T7 terminator TAGCATAACCCCTTGGGGCCTCTAAACGGGTCTTGAGGGGTTTTTTG. Total 273-360 bp (mean 320). E. coli MFC codon table (zero rare codons): A=GCG, C=TGC, D=GAT, E=GAA, F=TTT, G=GGC, H=CAT, I=ATT, K=AAA, L=CTG, M=ATG, N=AAC, P=CCG, Q=CAG, R=CGT, S=AGC, T=ACC, V=GTG, W=TGG, Y=TAT. Do NOT reuse the yeast oPool — it carries 45 AGA codons across 29 sample binders; AGA/AGG severely depress and frameshift E. coli translation.

Step 3 — Order IVTT kits. PURExpress E6800L (M11, $2,774, 2,500 µL) for the campaign; E6800S (M10, $316) if the pilot (step 14) is still pending. US overnight delivery. In parallel, order PUREfrex 2.0 (M12, $166/250 µL — quote-based since Nov 2024, verify price; Japan lead time ~1 week) as backup and cost lever. Disulfide contingency (M13/M14) only if Cys-paired binders survive the design filters.

Step 4 — Order detection reagents (first campaign = one-time setup ≈ $6,581). anti-FLAG M2-Tb (M19, $2,051) + Streptavidin-XL665 (M20, $1,665) + detection buffer 200 mL (M21, $762) + Greiner 784075 plate case (M22, $417 budgeted; market $348.73-417.33/case) + positive controls $698 (M23 $196 + M24 $297 + M25 $205) + primers (M3) $40 + plate spinner (M18) $723 (skip if a plate-rotor centrifuge exists) + reader training ~$225. This stock enables ~24-32 campaigns (buffer-limited at 10,000 wells).

Step 5 — Book the facility reader. Confirm the unit passes the HTRF optics checklist: excitation 320-340 nm (337 nm), dual emission 620 + 665 nm, time-gated 50-100 µs delay / 400-600 µs integration, top read, white low-volume 384 plates, and exports raw 620/665 RFU per well; confirm the facility's HTRF Reader Control Kit (RCK) DeltaF high/low calibrator check is current (SOP 06 §4). Certified models: Revvity EnVision 2103/2104/2105/Nexus; BMG PHERAstar FSX; VICTOR Nivo 5S/5F ONLY (not 3S/3F); CLARIOstar / CLARIOstar Plus (filter configuration only); Tecan Spark (filter optics); Infinite M1000 PRO; SpectraMax iD5 (Enhanced TRF cartridge); Agilent Synergy Neo2; BioTek Synergy H1 (legacy Victor X4/X5 also runs HTRF — SOP 06 §3). Account + training lead time: 1-2 weeks self-use (unassisted $7-60/hr ≈ $2-6/plate) or 2-4 weeks staffed ($100-250+/hr). Reserve 2 × 2 h blocks on separate days (screen day + Kd day) — place the final booking only after step 18 QC passes; tentatively hold slots now.

Step 6 — Receipt QC. IVTT kit arrives on dry ice → store at −80°C within 1 h; 24-month shelf life from manufacture; ≤5 freeze-thaws (keep a written log of Solutions A/B; record lot in WeaveSeq). Conjugates lyophilized at 2-8°C (bulk 610SAXAC frozen <−60°C); log vial lots + reconstitution dates. Verify the Greiner 784075 case is white, flat-bottom. Re-confirm the reader booking.

4. DAY 1 — TEMPLATE PREPARATION (steps 7-12)

Step 7 — Resuspend gBlocks (M1, M2). Centrifuge each tube 10 s to collect the pellet. Resuspend in 10 mM Tris pH 8.0 (M7) to ~10 ng/µL (verify); pipette up/down 10×. If a pellet remains: heat 55°C 5 min, vortex, spin down. Equilibrate 30 min at RT. Store at −20°C in DNA LoBind tubes (M30); ≤3 freeze-thaws. ~40 min per batch. QC: no visible pellet; concentration ~10 ng/µL.

Step 8 — In-silico QC of all 96 fragments (hard gate, before PCR). Verify with the template script: (1) all 96 translate in-frame, ORF = M-DYKDDDDK-⟨binder⟩-STOP, 0/96 frame or premature-stop errors; (2) FLAG present 96/96 immediately after ATG; (3) double stop TAATAG 96/96 before the 3′ spacer; (4) rare-codon scan AGG/AGA/CGG/CGA/ATA/CTA/CCC/GGA/TTA = 0 hits; (5) internal T7 promoter/terminator motifs in CDS = 0 hits; (6) length 273-360 bp. All 6 checks must pass for all 96 before ordering was placed and before PCR proceeds. On receipt, (7) diff the vendor order manifest against the template table — 99/99 sequences match (SOP 03 §8).

Step 9 — Universal PCR. One primer pair (M3) amplifies all 96 fragments + controls. F = T7 promoter TAATACGACTCACTATAGGG; R = reverse-complement T7 terminator CAAAAAACCCCTCAAGACCCGTTTAGAGGCCCCAAGGGGTTATGCTA. Master mix per 50 µL reaction (96 binders + 2-3 controls + 1 no-template control (NTC) in a 96-well PCR plate, M15):

Component	Per 50 µL well
Q5 2× Master Mix (M4)	25 µL
Universal F primer, 10 µM	2.5 µL
Universal R primer, 10 µM	2.5 µL
gBlock (1-10 ng)	variable
Nuclease-free water (M8)	to 50 µL

Cycling: 98°C 30 s → 12-15 × [98°C 10 s / 60°C 15 s / 72°C 30 s] → 72°C 2 min → hold 4°C. Use the minimum cycle count giving robust yield (extra cycles accumulate polymerase errors). ~1-1.5 h.

Step 10 — Cleanup + quantitation + normalization. Clean with Zymo DCC-5 (M5), elute 20 µL nuclease-free water (M8). Quantitate with Qubit dsDNA HS (M6); A260/A280 = 1.8-2.0. Normalize every amplicon to a 100 ng/µL working stock in DNA LoBind tubes (M30), −20°C, ≤3 freeze-thaws. IVTT dose target: 100 ng per 10 µL reaction (step 15), equivalent to NEB's 250 ng/25 µL standard. Bank master stocks: repeat campaigns then cost ~$100 of PCR instead of ~$2,765 of DNA.

Step 11 — Gel QC. Agarose 1.5-2% (verify): single band at expected size ±10% (~320 bp mean). Multiple bands → raise annealing to 62-65°C (verify), cut to 12 cycles, gel-purify the correct band.

Step 12 — Sanger the dose-response set. Submit the top 12-20 binders (the dose-response cohort + screen outliers) with universal primers (M3), ~$4-8/sample (budget $72-160). Exact-match gate; any mismatch → re-order that fragment from the backup vendor or exclude from titration and flag in WeaveSeq. Results must be in hand before step 21 (Day 3 re-expression) — submit today.

5. DAY 2 — IVTT EXPRESSION + SCREEN (steps 13-20)

Step 13 — Thaw PURExpress (M10/M11). Solutions A and B from −80°C; thaw on ice 10-15 min (never at 37°C; do not store long-term at −20°C). Flick/pipette to mix. Keep on ice between uses. Record lot + freeze-thaw count in WeaveSeq. (PUREfrex M12, Solutions I/II/III: same handling.)

Step 14 — Pilot gate (first campaign only). Express 3-4 binders + POS1 + CTRL with 1 × E6800S (M10) before the full campaign. PASS = anti-FLAG dot blot positive (step 18 method) AND expression consistent with the 0.5-1 mg/mL expected yield. Do not start the full campaign on a failed pilot.

Step 15 — Prepare IVTT mini-reactions (10 µL/well). Per-well recipe; single master mix (no template) for 110 wells: 440 µL Solution A + 330 µL Solution B (7 µL/well):

Component	Per 10 µL well
Solution A	4 µL
Solution B	3 µL
Template DNA, 100 ng/µL (step 10)	1 µL (= 100 ng; range 10-400 ng)
Nuclease-free water (M8)	2 µL (NTC wells: 3 µL, no template)

Dispense 7 µL master mix + 1 µL template + 2 µL water per well in a 96-well PCR plate (M15). If amplification yield was <100 ng/µL, concentrate or increase template volume (compensate water), keeping template ≤4 µL/well. Expected yield 0.5-1 mg/mL = 5-10 µg/10 µL ≈ 60-125 µM crude (3-4 orders above the ~1-10 nM assay need). PUREfrex M12 recipe (SOP 04 §4.2): Solution I 5 µL + Solution II 0.5 µL + Solution III 1 µL + template (~100 ng, verify) + water to 10 µL per well.

Step 16 — Plate layouts.

Plate 1 (screen IVTT)	Plate 2 (QC/controls IVTT)
b01-b96 row-major in ranked order: A1-A12 = b01-b12, B = b13-b24, C = b25-b36, D = b37-b48, E = b49-b60, F = b61-b72, G = b73-b84, H = b85-b96	A1-A4 = NTC; A5-A6 = POS1; A7-A8 = CTRL; A9-A12 = spare NTC; B-H empty (day 3: 12 × 25 µL re-expressions go here)

Step 17 — Incubate. Seal (M16, or foil M17), press firmly around every well. Incubate 2 h at 37°C, static (no shaking; up to 4 h max — over-incubation costs yield to RNA decay) in an incubator or thermal cycler with heated lid 37-40°C. Keep the seal on throughout. Then: plate on ice 5 min; spin 1,000 × g 2 min in the plate spinner (M18) to collect condensate/pellet aggregates. Proceed within ~1 h.

Step 18 — Expression QC: anti-FLAG dot blot. Spot 1 µL of each reaction diluted 1:10 in PBS (M28) per binder + all Plate 2 controls + a FLAG-peptide titration (M24, e.g. 50/10/2 ng) as loading standard. Air-dry 10 min. Block 3% BSA in PBS-T (PBS + 0.1% Tween-20, M27) 30-60 min RT. Anti-FLAG-HRP (M25) 1:2,000-1:5,000 in block buffer, 1 h RT. Wash 3 × 5 min PBS-T. Develop with ECL (M29), image. ~3 h total (≈2.5-3 h wall clock: 10 min dry + 30-60 min block + 1 h antibody + 3 × 5 min washes; SOP 04's ~1 h figure is hands-on time only). PASS: clear signal above NTC in ≥80% of binder wells AND POS1/CTRL strong. Flag and exclude failing wells from the screen; record in WeaveSeq. (FLAG is N-terminal — truncated products still light up; if truncation is suspected, Sanger the template.) Book the facility slot (step 5) now that QC passed.

Step 19 — Dilute lysates. Dilute neat IVTT reactions in detection buffer (M21) so the FINAL assay dilution is 1:25-1:50 → binder ≈ 1.2-5 µM in the well (60-125 µM crude ÷ 25-50). The 1-10 nM regime SOP 01 §6.2 targets for titrations needs a ~1:5,000-1:12,000 final dilution — apply that larger dilution for titration plates (step 27). The lysate occupies 5 µL of the 20 µL final volume, so the added pre-dilution must be 4× more concentrated (e.g., 1:20 final = pre-dilute 1:5). Prepare in the intermediate dilution plate (M34) after a plate spin (M18). Dilute ALL lysates identically, including NT, CTR- and POS1 control reactions. Doubles as buffer exchange — no purification (PURE-type lysates are colorless/defined). Diluted lysate is good for same-day screening (hold on ice); overnight 4°C only for confirmation re-reads of already-diluted samples; never freeze-thaw diluted lysate for quantitative Kd data.

Step 20 — Screen plate S1: setup, incubation, read. 384-well Greiner 784075 (M22); 20 µL final; 240 assay wells.

Conjugate 4× stocks (fresh per batch, in M21): anti-FLAG M2-Tb (M19) 4 nM → 1 nM final (vendor range 1-5 nM); Streptavidin-XL665 (M20) 160 nM → 40 nM final (~50 ng/well). Add as separate 5 µL additions — do NOT premix the two conjugates.
2× target stock: 400 nM in M21 → 200 nM final. (Optional dual-concentration upgrade: 2× stocks at 40 nM and 1,000 nM → 20/500 nM finals on two plates.)
Fill the buffer moat FIRST: 20 µL M21 into all unused wells (rows K-P).
Add samples: 5 µL diluted lysate per well per the map; POS/BUF wells get 5 µL buffer (POS peptide is added in the next addition).
Add target: 5 µL 2× target (400 nM) to binder, NT, CTR-, POS1 wells; 5 µL buffer instead to NEG, POS, dBk, aBk, COMP wells. (For POS: 5 µL of 100 nM 2× biotin-FLAG peptide, M23.)
Pre-incubate: seal (M16); 30-60 min RT, dark.
Add donor: 5 µL 4× M19 to ALL wells except dBk (buffer instead).
Add acceptor: 5 µL 4× M20 to ALL wells except aBk (buffer instead). Final = 20 µL everywhere.
Incubate: sealed, 1-2 h RT, dark.
Read at the facility (settings §DAY 3 step 24). Read 1-3 h after conjugate addition (target; absolute max <6 h plate-to-read).

Screen plate S1 map — matches SOP 05/SOP 07 §6 exactly (rows A-H = binders, rep 1 cols 1-12, rep 2 cols 13-24; rows I-J = QC; K-P = buffer moat). Binder order is column-major within 8-binder blocks (b01, b09, b17, … per the SOP 07 plate designer) — NOT the row-major order used on IVTT Plate 1 (step 16):

     1    2    3    4    5    6    7    8    9   10   11   12 |  13   14   15   16   17   18   19   20   21   22   23   24
A   b01  b09  b17  b25  b33  b41  b49  b57  b65  b73  b81  b89 | b01  b09  b17  b25  b33  b41  b49  b57  b65  b73  b81  b89
B   b02  b10  b18  b26  b34  b42  b50  b58  b66  b74  b82  b90 | b02  b10  b18  b26  b34  b42  b50  b58  b66  b74  b82  b90
C   b03  b11  b19  b27  b35  b43  b51  b59  b67  b75  b83  b91 | b03  b11  b19  b27  b35  b43  b51  b59  b67  b75  b83  b91
D   b04  b12  b20  b28  b36  b44  b52  b60  b68  b76  b84  b92 | b04  b12  b20  b28  b36  b44  b52  b60  b68  b76  b84  b92
E   b05  b13  b21  b29  b37  b45  b53  b61  b69  b77  b85  b93 | b05  b13  b21  b29  b37  b45  b53  b61  b69  b77  b85  b93
F   b06  b14  b22  b30  b38  b46  b54  b62  b70  b78  b86  b94 | b06  b14  b22  b30  b38  b46  b54  b62  b70  b78  b86  b94
G   b07  b15  b23  b31  b39  b47  b55  b63  b71  b79  b87  b95 | b07  b15  b23  b31  b39  b47  b55  b63  b71  b79  b87  b95
H   b08  b16  b24  b32  b40  b48  b56  b64  b72  b80  b88  b96 | b08  b16  b24  b32  b40  b48  b56  b64  b72  b80  b88  b96
I   POS  POS  POS  POS  NEG  NEG  NEG  NEG  NT   NT   CTR  CTR | aBk  aBk  dBk  dBk  POS  POS  POS  POS  NEG  NEG  NEG  NEG
J   POS  POS  POS  POS  NEG  NEG  NEG  NEG  NT   NT   CTR  CTR | aBk  aBk  dBk  dBk  POS  POS  POS  POS  NEG  NEG  NEG  NEG
K-P BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF | BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF  BUF

Control legend (used on S1 and T1): POS = 50 nM biotin-FLAG + both conjugates, no lysate (maximal bridge signal; validates both conjugates + full sandwich). POS1 = IVTT FLAG-core-streptavidin lysate + target + conjugates (end-to-end geometry). NEG = lysate + both conjugates, NO target (baseline for ratio/deltaF). NT = no-DNA IVTT diluted identically + target + conjugates (lysate-only background). CTR- = FLAG-eGFP lysate + target + conjugates (must read ≈ NEG). aBk = POS peptide + donor only (donor bleedthrough). dBk = POS peptide + acceptor only (acceptor cross-excitation). COMP = POS + 100 µg/mL free FLAG peptide (M24) (must drop to ≈ NEG). BUF = buffer only.

6. DAY 3 — DOSE-RESPONSE TITRATIONS + FACILITY READ (steps 21-24)

Step 21 — Re-express the top 12 screen hits at 25 µL. Fresh day-3 lysate (resets the lysate-age clock) in the empty rows of Plate 2: per well 10 µL Solution A + 7.5 µL Solution B + 250 ng template (step 10 stock) + water to 25 µL. Incubate/QC as steps 17-18 (12 × 25 µL = 300 µL; yields 12-25 µg per binder). Sanger veto (step 12) must be in hand first — exclude any erroneous template from the titration set. 13+ binders → second T1 plate.

Step 22 — Prepare the 2× target master series. Serial 1:1.6 dilutions of biotinylated target (M26) in detection buffer (M21) in LoBind tubes (M30), 13 tubes:

Tube	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13
2× (nM)	2000	1248	780	486	304	189.6	118.4	73.8	46.2	28.8	18.0	11.2	0 (buffer)
Final (nM)	1000	624	390	243	152	94.8	59.2	36.9	23.1	14.4	9.0	5.6	0

Optional 6,000 nM 2× top point (3,000 nM final) for suspected weak binders. Do NOT substitute a 3-fold series — 12 true 3-fold steps from 1,000 nM run off the bottom at 5.6 pM, below assay range. The series brackets the expected 5-300 nM Kd window.

Step 23 — Titration plate T1: setup, incubation. 384-well Greiner 784075 (M22); 12 binders × 13 points × 2 replicates = 312 binder wells + 32 QC + 16 BUF wells (rows N-O) + moat (matches the 584-well cost model, §9).

Moat first: 20 µL M21 in row P.
Target: 5 µL of the appropriate 2× point into every well of rows A-M — each 2-column binder block gets the full series top-to-bottom (P1 in row A … P13 in row M).
Binder lysates: 5 µL per well — binder b1 in columns 1-2, b2 in 3-4, … b12 in 23-24; both replicates. First titration of an unknown binder: plate at two lysate dilutions (e.g., 1:1,000 and 1:5,000 final — SOP 05's 1:50/1:200 example leaves µM-range binder, step 19) on separate plates.
QC wells (rows N-O): POS ×8, NEG ×8, NT ×4, CTR ×4, aBk ×4, dBk ×4, BUF ×16 per the control legend (step 20). NT/CTR- get 5 µL 2× 400 nM target (200 nM final); POS gets 5 µL 100 nM biotin-FLAG (M23); COMP, if included, gets 5 µL 200 µg/mL 2× free FLAG (M24).
Pre-incubate 30-60 min RT dark (sealed, M16).
Conjugates: 5 µL 4× M19 to all wells except dBk; 5 µL 4× M20 to all wells except aBk (stocks from step 20).
Incubate 1-2 h RT dark, sealed.

        1 2 | 3 4 | 5 6 | 7 8 | 9 10 | 11 12 | 13 14 | 15 16 | 17 18 | 19 20 | 21 22 | 23 24
  A     P1 P1 (b1) | P1 P1 (b2) | ...                                                      (b12)
  B     P2 P2      | ...
  C     P3 ...     |   (P4..P12 in rows D-L)
  M     P13 P13 (0 nM blank)
  N-O   POS POS NEG NEG NT CTR aBk dBk BUF... (32 QC + 16 BUF across both rows)
  P     buffer moat

Keep critical points off the outer ring; both plates' QC wells sit in rows N-O (already interior).

Step 24 — Read at the facility. Reader settings template: format 384-well white low-volume (M22), top read, 20 µL/well; excitation 337 nm laser or UV flash 320-340 nm; emission 620 nm donor + 665 nm acceptor (simultaneous dual emission if available); delay/integration per model — EnVision 50/400 µs, VICTOR Nivo (5S/5F only) 70/400, CLARIOstar (filter build) 60/400, Tecan Spark (filter) lag 100 µs/integration 400 µs (gain 80-220), Infinite M1000 PRO 60/500, SpectraMax iD5 (Enhanced TRF) 100/600, Synergy Neo2 150/500, PHERAstar FSX as certified; 30-200 flashes/well; endpoint reads only (no kinetics, no injectors). Export raw 620 and 665 RFU per well (plus computed ratio if offered) as CSV. Transport: plates read as-is (no wash, no stop). Covered/foil-wrapped, dark, room temperature, NO ice; leave the seal on until the read head is ready; do not centrifuge the assay plate after the final incubation unless the facility requests it (SOP 05's pre-read spin conflicts with SOP 06 — follow SOP 06). Schedule the read to land 1-3 h after conjugate addition (hard max <6 h). Read time ≤5 min/plate (14 s flying-mode on PHERAstar); ~3 reads total in Scenario B. After the read, export raw data and hand to Day 4 — do not use the reader's on-board curve fitting.

7. DAY 4 — ANALYSIS + WEAVESEQ PUSH (steps 25-29)

Step 25 — Compute ratios. Archive raw files unmodified at campaign_root/trfret/raw/<plateId>_<reader>_<date>.<ext> (plate IDs IVTT-TRFRET-<campaign-slug>-<plate#>); record plate ID, reader model, delay/integration, reagent lots, incubation time, operator. Then per well: HTRF ratio R = (665 nm / 620 nm) × 10,000 — never fit raw 665 counts (the ratio compensates pipetting and mild interference). deltaF% = (R − R_NEG) / R_NEG × 100 (R_NEG = plate's own NEG mean). Assay window (titrations) = max ratio / min ratio.

Step 26 — Plate QC gates, then call screen hits. Apply per plate; on fail, see §8.

Gate	Pass	Fail →
Z′ (POS vs NEG, ≥4+4 wells; 8+8 recommended)	≥ 0.5 (0.4-0.6 acceptable with caution — do NOT hard-reject on Z′ alone)	flag, investigate control prep; re-read/repeat
Control CV (POS + NEG replicates)	≤ 10%	fix pipetting, repeat control block
POS bridge deltaF%	≥ 200%	conjugate problem — do not trust the plate
620 nm donor cps (donor-loading QC; not expression)	20,000-80,000 (accept ≥ 5,000); flag wells outside 0.5-2× donor-control median	donor dilution or reader-gain error
Assay window (titration plates)	≥ 3 (gate ≥ 2)	re-titrate wider range
NT & CTR- vs NEG	within ~2× of NEG ratio	lysate background — check dilutions, raise BSA/Tween
COMP (first plate of campaign)	drops to ≈ NEG	donor binds non-tagged material
Titration top points	P1 ≥ P3 (top points can hook from free-biotin competition for the fixed acceptor; exclude hooking P1/P2 from the fit)	exclude from fit

Well-level flags: exclude/down-weight a well if 620 cps outside 0.5-2× donor median, duplicate ratio deviation > 30%, or 665 nm above aBk median + 5σ with 620 normal. Screen hit call (mean of n=2): hit = R ≥ μ_NEG + 5σ AND deltaF% ≥ 50%; strong hit = deltaF% ≥ 100%. Rank by deltaF%; promote top 12-20 (plus design-priority binders) to titration.

Step 27 — Fit the quadratic 1:1 model. Geometry: fixed FLAG-binder, titrated biotinylated target; donor on binder, acceptor on target; signal increases with target. Model in total concentrations, binder concentration B free: AB = [ (B + X + Kd) − sqrt( (B + X + Kd)² − 4·B·X ) ] / 2 ; Y = bg + S·AB Midpoint = Kd + B/2 in the depletion regime; reduces to Langmuir when B ≪ Kd. Commands (proposed helper affinibind/scripts/trfret/fit_kd.py — scripts do NOT exist yet in the repo; use scipy/lmfit directly):

from scipy.optimize import curve_fit   # weights 1/y; 1/y² for the 4PL cross-check
# 8 steps: import CSV → compute R → plate QC → well QC → fit quadratic (ALWAYS primary)
# → cross-check 4PL on log10(target) + Langmuir when fitted B < Kd/3
# → bootstrap 500-1,000 well resamples for 95% CI on Kd → acceptance gates → emit JSON + PNG

Fit acceptance: n ≥ 8 points; ≥ 4 points within 3-fold of fitted Kd; curve span (top/bottom) ≥ 1.5×; 95% CI span < 10-fold; R² > 0.9. Depletion flag: fitted B ≥ Kd/3 → apparent Kd is an UPPER BOUND → re-run the lysate more dilute — the 1:25-1:50 screen dilution leaves ~1.2-5 µM binder, so reach the 1-10 nM regime with a ~1:5,000-1:12,000 final dilution (step 19). Quantitative floor ~2-5 nM; tighter binders are reported as upper bounds. Flat curve (span < 1.5×) → report "Kd > 1 µM". Fitted B < 1 nM (reachable only in the 1-10 nM titration regime) → low-expression flag (retest with 25 µL re-expression or drop). Optional dual-concentration screen (ranking only): R = ratio(500 nM)/ratio(20 nM) (SOP 01 §6.1/SOP 07 §6 text writes the reciprocal ratio(20)/ratio(500) — see footer, verify); apparent Kd ≈ C_hi·C_lo·(R−1)/(C_hi − R·C_lo); valid only when binder ≪ Kd and 1 < R < 25; noise-sensitive — triage, not quantitative.

Step 28 — Hit matrix + final report. Confirmed binder = fitted Kd ≤ 1 µM with all fit-acceptance gates; Strong binder = Kd ≤ 100 nM (priority for soluble-scale validation); Weak = 1-3 µM; Non-binder = "Kd > 1 µM"; flags set for depletion/low-expression. Merge onto the annotated candidate table (join on candidate_uid — NOT uid), adding trfret_kd_nm, trfret_kd_ci_lo_nm, trfret_kd_ci_hi_nm, trfret_binder_total_nm, trfret_depletion_flag, trfret_screen_deltaf_pct; write candidate_table.trfret.parquet + a human-readable CSV sorted by classification then Kd. Report = per-binder PNG of the fitted curve + the QC gate table (step 26) + the hit CSV + plate IDs/lots.

Step 29 — Push to WeaveSeq. One campaign-level screen experiment row (proteinCandidateId: null) + one titration row per candidate. Code changes (do once): add TR_FRET to enum ExperimentType in prisma/schema.prisma + migrate; add \{ value: "TR_FRET", label: "TR-FRET (HTRF)" \} in ExperimentForm.tsx; add results?: unknown to CreateExperimentInput; in db-loader.ts set results: data.results ?? null (currently silently dropped). POST https://<weaveseq>/api/experiments requires a WeaveSeq session cookie/token (mechanism still open — check with the app admin). Results JSON, schemaVersion: "trfret-1.0":

{
  "schemaVersion": "trfret-1.0",
  "kind": "titration",
  "assay": "trfret_htrf_ivtt",
  "plateId": "IVTT-TRFRET-SEB-001",
  "reader": { "model": "PHERAstar FSX", "excitationNm": 337, "delayUs": 60, "integrationUs": 400 },
  "qc": { "zprime": 0.72, "controlCvPct": 6.4, "assayWindow": 14.2, "deltaFposPct": 2400,
          "donor620MedianCps": 41200, "negMeanRatio": 118 },
  "candidateUid": "cand-0001",
  "fit": { "model": "quadratic_1to1", "kdnm": 156, "kdCi95LoNm": 102, "kdCi95HiNm": 240,
           "binderTotalNm": 8.2, "htrfRatioMax": 4520, "htrfRatioMin": 210,
           "nPoints": 13, "nReplicates": 2, "rSquared": 0.985, "curveSpan": 21.5,
           "depletionFlag": false,
           "targetNm": [1000, 624, 390, 243, 152, 94.8, 59.2, 36.9, 23.1, 14.4, 9.0, 5.6, 0] }
}

Screen rows use kind: "screen" with targetConcNm: 200 and per-candidate ratioMean, ratioSd, deltaFPct (schema for the per-candidate array is not fully pinned in the sources — verify against SOP 07 §11).

8. TROUBLESHOOTING QUICK HITS

Escalation ladder (in order): re-read the plate → re-pipette affected wells → Sanger the template → re-express at 25 µL → re-order the fragment → drop with a recorded reason (drop only when two independent causes are excluded).

Single PCR fragment fails → likely ~6% synthesis-error rate (1:5,000/bp) or lost pellet → re-spin, re-resuspend in M7, re-PCR; still silent → re-order from backup vendor; flag "excluded — template QC".
All PCR fails / multiple bands → check universal F/R sequences (step 9) and 60°C anneal vs primer Tm; raise annealing to 62-65°C (verify), cut to 12 cycles, gel-purify correct band; replace Q5 master mix (M4).
Plate-wide IVTT silence → kit exceeded ≤5 freeze-thaws or 24-mo shelf life, or wrong temperature → check lot/log, thaw fresh aliquots, confirm 37°C for 2 h (max 4 h); verify master mix contains Solutions A AND B; confirm DNA within 10-400 ng/10 µL; suspect RNase → nuclease-free water + filter tips.
Subset/single binder silent → rare codons (AGA/AGG — yeast oPool signature) or SD not 6-8 nt upstream of ATG → re-run rare-codon scan, redesign first 10-15 codons with the MFC table, re-order.
Ladder of truncation bands → frameshift/indel from synthesis error or hidden stop → Sanger; never trust screen results from an erroneous template.
Turbid lysate / curve won't saturate at high lysate input → aggregation → dilute 1:25-1:50 into buffer (step 19), reduce lysate input, re-titrate more dilute; Cys-paired binder → M13 or M14 contingency.
No assay window (ratio ~flat) → run the POS biotin-FLAG bridge (M23) fresh: dead conjugate or free-biotin poisoning → replace whichever conjugate restores signal; POS works but binders flat → check dot blot + fitted B (620 cps is donor-loading, NOT expression); verify target biotinylation (1-3 biotins/protein).
High background from lysate → keep final lysate share ≤ 5-10% of volume; dilute 1:25-1:50; fresh tips per well; spin the IVTT plate (M18) before transfer. (PURExpress ≈ PUREfrex ≈ colorless lowest-background systems; never rabbit reticulocyte lysate — hemoglobin absorbs ~320-450 nm.)
Edge-effect noise → fill the moat first, seal throughout incubation, keep critical titration points off the outer ring, keep incubations 1-2 h RT.
Low Z′ / high control CV → pipetting imprecision, edge wells, evaporation → check control CV ≤ 10%, moat + seal, re-read; small window → reagent concentrations or dead conjugate → check POS deltaF% ≥ 200%.
Bad Kd fits → midpoint shifting with dilution = depletion regime (B ≥ Kd/3) → refit quadratic with B free, re-run at ~1:5,000-1:12,000 final dilution (the 1:25-1:50 screen dilution leaves µM-range B — step 19); CI span ≥ 10-fold → check series arithmetic (3-fold × 12 steps overshoots to pM; use the 1.6-fold series), add replicates or re-center on provisional Kd; Kd < ~2-5 nM with exploding CI = depletion floor → report upper bound.
Reader wrong build → CLARIOstar/Spark monochromator configs, VICTOR Nivo 3S/3F, or iD5 without the Enhanced TRF cartridge are not HTRF-capable → request the filter-optics/5S-5F/cartridge build; facility exports only computed ratio → request raw 620/665 RFU per well (needed for the donor QC gate).

9. COST REFERENCE

Scenario	Deliverable	Wells	Campaign cost	Per well	Per binder	First campaign (incl. ~$6,581 setup)
A	Screen only (hit/no-hit)	240	$4,700	$19.6	$49	≈ $11,300
B (recommended)	Screen + 12-pt Kd, top 12	584	$6,418	$11.0	$67	≈ $13,000
C	Screen + 8-pt Kd, top 48	1,232	$6,874	$5.6	$72	≈ $13,500
D	Full 12-pt Kd, all 96	2,752	$8,009	$2.9	$83	≈ $14,600

Scenario B line items: IVTT 1× E6800L $2,774 · DNA (96 gBlocks $2,765 + 3 controls $130 + PCR $150 + Sanger top-12 $72) $3,117 · conjugates 584 wells × $0.186 $109 · buffer 11.7 mL $45 · 2× Greiner plates $21 · PCR plates + seals $26 · tips 2 boxes $110 · tubes/reservoirs $28 · water $26 · positive controls $50 · reader 3 reads ~$12 · dry-ice shipping ~$100. One-time setup ≈ $6,581 (conjugates $3,716 + buffer $762 + plate case $417 + controls $698 + primers $40 + spinner $723 + training ~$225) ≈ 24-32 campaigns. Cost levers (biggest first): PUREfrex 6 kits $996 instead of E6800L $2,774 (−$1,778) · 6× E6800S $1,896 (−$878) · negotiate $0.07/bp gBlocks (−$400-600) · re-amplified fragment stocks make repeat campaigns ~~$100 of PCR (~~$2,700 saved) · homebrew HEPES/KF/BSA/Tween buffer cuts $0.076 → ~$0.005/well (validate against M21 first; total ~$30-50 over 10,000 wells (verify)).

Every value in this SOP is sourced from the 11 protocol files in this folder; where the sources disagree, the majority-source value is used and flagged (verify). Known conflicts resolved this way: titration cohort fixed at 12 (Scenario B, T1 layout) with Sanger scope 12-20; 13 wells per curve (12 concentrations + 0 nM blank); Greiner 784075 budgeted at $417.33/case; anti-FLAG-HRP = A01869 (verify vs A01428); donor part 61FG2TLB (verify 61FGBTLB renumbering); PUREfrex price/lead (verify, quote-based since Nov 2024); E6800L consumption 1,260 µL (960 + 300); conjugate pre-mixing forbidden; read window 1-3 h target / <6 h max; staffed reader $100-250+/hr; n=2 replicates (n=3 optional for top-5 Kd quality); dual-point R = ratio(500)/ratio(20) (SOP 01 §6.1 and SOP 07 §6 literally write ratio(20)/ratio(500) — contradicted by SOP 07's own worked example Kd = 100 nM → R = 5 and its 1 < R < 25 validity condition — verify); lysate dilution range 1:25-1:50 (SOP 05/08/09 say 1:10-1:50; used the narrower range); S1 plate map = SOP 05/SOP 07 §6 layout (column-major binder order, QC rows I-J).