How to Audit CFPS Vendor Quotes When Throughput Exceeds 200 Reactions per Week

When weekly reaction counts cross roughly 200, reagent economics stop being a line item, so they turn into the main cost driver for the whole CFPS program. The cheapest-looking vendor quote often buries the variables that decide real cost: reaction format, yield measurement conditions, and lot-to-lot consistency. This piece lays out the questions and calculations you need to turn an opaque quote into a verifiable, per-milligram cost comparison before you commit a budget to any single supplier.

Why high-volume CFPS procurement differs from low-volume procurement

A lab running only a handful of cell-free reactions a week can treat reagent spend as background cost. Cell-free protein synthesis skips cloning, transformation, colony screening, culture growth, and lysis, so a protein-engineering cycle that once took days can run in hours. CFPS is fast, so screening programs grow quickly, and once they pass roughly 200 reactions a week, the per-reaction price of a kit and the per-reaction yield decide whether the program's budget holds for the quarter.

Once a program is at that volume, a cost difference of a few cents per reaction, or a meaningful yield swing between lots, becomes a line item finance will ask about. Crude cell-free extracts carry real variability risks: contaminating nucleases and proteases, loosely controlled ratios between components, and batch-to-batch drift in performance. At low volume, a scientist can absorb a bad lot by rerunning a plate. At 200-plus reactions a week, that same bad lot means you have to re-optimize across the whole queue, at full reagent cost, and the week's schedule slips behind it. The question a buyer has to ask changes along with the volume. A scientist picking a kit for a single project just asks whether it works. If you run a weekly reaction count in the hundreds, you have to ask whether the quote can be verified against performance data, and whether that performance holds lot after lot. Answering that question requires a different kind of document than most vendors send out by default, and getting that document is the actual work of procurement at scale.

Vendor quote requirements

Most vendor quotes list a price per kit or per milliliter, and then they stop. They skip the four things that determine what a reaction actually costs: reaction volume, reaction format, yield, and whether that yield holds across lots. A price per milliliter tells a buyer almost nothing about cost per milligram of usable protein, which is the number that actually matters once a program is running hundreds of reactions a week.

Three gaps appear repeatedly in standard quotes. Yield claims arrive without a stated format, even though titers from CFPS reactions shift with oxygenation: batch reactions in tube format generally produce higher titers than the same reaction run in a 96-well plate, and 96-well plates generally outperform 384-well plates, most likely because of differences in oxygen exchange and surface area. A yield number with no format attached cannot be checked against anything and should be treated as incomplete. Lot-specific performance data goes missing just as often. If a vendor publishes only a single product-level specification, with no lot-by-lot quality control record, it is withholding the inter-lot variance that decides whether a program hits its reproducibility targets. None of this points to bad faith on the part of vendors. Opacity of this kind is close to standard in the market. A buyer running a high-volume program needs a checklist for evaluating vendors. Chaperone supplements, Mg²⁺ optimization reagents, and energy regeneration components also tend to get priced as add-ons while marketing language describes them as part of a complete system. Each of those needs to be itemized and priced separately before a true per-reaction cost can be built.

Converting opaque line items into a comparable per-reaction cost

Cost per milligram of protein produced is the only figure worth comparing across vendors. Cost per milliliter of reagent and cost per reaction both hide the one variable that moves the economics more than any other: how much usable protein actually comes out of the reaction. A reagent priced low per milliliter can still cost more per milligram if its yield lags behind a pricier competitor's.

Building that per-milligram figure takes four numbers from the vendor: the reaction volume per well or tube in microliters, the stated yield in micrograms per milliliter or per reaction (with the format that yield was measured in stated explicitly), the volume of each reagent component consumed per reaction (lysate, reaction buffer, energy mix, itemized rather than bundled), and the price per unit volume for each of those components. Skipping any one of the four means the resulting number is a guess dressed up as a calculation.

Format correction has to happen before any of this math means anything. Yield data generated in tube format cannot be applied to a quote built around 384-well plates. If a vendor quotes tube-format yields against a plate-based workflow, you end up with a cost-per-milligram figure that looks better than what the plate will actually deliver. Scale correction runs the other direction: most published CFPS work reports protein titers below 2 grams per liter, so a quote that implies cost parity with cell-based expression at milligram scale deserves a direct request for the underlying data before it's taken at face value.

Working the math makes the gap concrete. The nominally cheaper quote turns out to cost more once yield enters the calculation. That reversal is common enough that it should be checked on every quote.

Auditing yield claims: format, oxygenation, and the Mg²⁺ disclosure test

A yield claim only means something if it specifies three things: the reaction format it was measured in, the incubation conditions used, and the lot the data came from. Missing any one of the three leaves a number that cannot be checked against anything.

Format and oxygenation drive a meaningful share of that variability. If you batch 20-microliter reactions in 2-milliliter tube format, you generally get higher titers than the same chemistry run in 96-well plates, and 96-well plates generally beat 384-well plates. Two direct questions separate vendors with real format-specific data from vendors reciting a literature value: what format was the stated yield measured in, and what yield should the buyer expect in the format they actually intend to run. A vendor with genuine data answers the second question with a number. A vendor working from tube-format publications usually can't.

Mg²⁺ concentration deserves its own line of questioning, and it's one of the least appreciated levers in this whole audit. A direct question settles whether a vendor has done that work: does your lot-level quality control data include the optimal Mg²⁺ range for each lot? A vendor who can't answer that question is sitting on a known source of run-to-run variance, one that will cost both time and reagent once a program is running at 200-plus reactions a week. CFPS systems tolerate toxic protein targets and allow non-canonical amino acids to be incorporated directly into a reaction, but only when the surrounding reaction chemistry is configured correctly, which depends on formulation details the vendor has to disclose.

Lot-to-lot variability as a hidden budget risk

An undisclosed lot change partway through a high-volume program isn't a minor scientific hiccup. At 200-plus reactions a week, it turns into a budget event: re-optimization runs burn reagent and schedule time at the program's full scale.

Surfacing that risk before it happens means demanding specific documents from any vendor under consideration: a Certificate of Analysis for each lot received, carrying lot-specific yield data against a defined reporter protein such as sfGFP or eYFP measured under stated conditions; inter-lot coefficient of variation data, since a vendor who can't produce this can't back up a reproducibility claim with anything beyond assurance; the lot-specific Mg²⁺ optimization range described above; and formal qualification records covering Certificate of Analysis verification, residual contaminant testing, and traceability for each lot.

The choice between a PURE system and a crude extract system carries a cost-benefit trade-off, and this kind of documentation makes it quantifiable. Fully reconstituted PURE systems remove contaminating RNases, proteases, and background host proteins by building the reaction from defined, individually purified components, which tends to produce steadier performance across batches. At high volume, that comparison belongs in the spreadsheet is before the first purchase order, not after a program-wide yield drop the diagnosis traces back to an unannounced lot change. The opacity built into most vendor quotes, missing lot-level QC, undisclosed formulations, format-agnostic yield claims, turns into a financial control risk once volume crosses that threshold. Vendors such as Sepia Bio publish lot-specific performance data and disclose full formulations with every shipment, which shows that transparency and competitive pricing don't have to conflict. They're both requirements for the reproducibility and cost predictability a high-throughput program actually needs.

Verifying automation compatibility claims

A reagent that hasn't been validated against the lab's own dispensing hardware doesn't count as compatible, regardless of what a vendor's product sheet says. At 200-plus reactions a week, most of that volume runs through acoustic liquid handlers such as the Echo 650, and these transfer nanoliter volumes of CFPS components directly into 384-well plates. A vendor claiming automation compatibility should be able to name the specific instruments its reagents have been tested on and produce the validation data behind that claim, not just an assertion that the reagent "works with standard liquid handlers."

Three questions separate a validated claim from a marketing one: which liquid-handling or acoustic dispensing instruments has the reagent actually been run on; what dead volume does the reagent's packaging require per plate run; and has the reagent been tested at the specific reaction volumes and plate densities the buyer intends to use, such as 1 to 5 microliters in 384-well format. Oxygenation differences between tube and plate formats are a documented variable in their own right, so plate density and format affect yield independent of reagent quality, and automation validation data has to be specific to the format in question. A vendor who says the same protocol, lysate, and reaction chemistry performs from a 50-microliter screening reaction up through preparative scale should be able to hand over side-by-side yield data at each scale, generated from the same lot, rather than asking the buyer to take the claim on faith.

Formulation transparency as an audit criterion, not a philosophical preference

A vendor that won't disclose its reaction buffer composition is asking a buyer to accept a black box at a volume where any unexplained yield failure costs real money and real time to diagnose. Reaction buffer recipes vary a lot across protocols and vendors, both in how many optional components they include and in the concentration of each one. If you don't have that recipe in hand, you can't compare one vendor's formulation against another's, you can't troubleshoot a sudden yield drop, and you can't standardize a protocol across multiple lots or multiple lab sites.

That opacity costs money in a few predictable ways. The direct test is straightforward: ask the vendor for a full ingredient list with concentrations for each component, not a category label like "energy regeneration system." If the vendor declines, the quote should be treated as carrying an unquantified troubleshooting cost that will come due eventually. Sepia Bio documents its OpenCFPS platform at the genome and buffer level so that formulation doesn't become a trust exercise, and buyers get what they need to troubleshoot, standardize, and scale a protocol without taking anything on faith.

Comparing vendor options at high volume: what to look for in each category

At 200-plus reactions a week, a defensible vendor comparison has to run through each of the categories above, not settle on a single headline price. On cost, the only number worth comparing is price per milligram of protein, calculated from reaction volume, format-matched yield, and itemized reagent consumption. Sepia Bio's OpenCFPS E. coli Core Kit lists at $80 per milliliter, and the company states pricing as low as $25 per milliliter against typical kit pricing around $1,330 per milliliter; these figures mean something only once they're converted to cost per milligram using a buyer's actual format and yield data.

On yield, you should favor vendors who state format explicitly and can tell you what titer to expect in your specific plate or tube configuration, since titers measured in tube format don't carry over directly to 384-well plates. On reproducibility, the deciding factor is whether a vendor supplies lot-specific Certificates of Analysis, inter-lot CV data, and a documented Mg²⁺ range per lot, the three pieces of evidence that separate a verifiable reproducibility claim from an assurance. If a PURE-based system is worth its premium, it should come with inter-lot CV data that actually shows the lower re-optimization frequency you're paying the added cost for, not just a general appeal to the purity of its components. And on formulation, the standard is a full, itemized ingredient list with concentrations, the kind of disclosure Sepia Bio builds into its OpenCFPS documentation when it publishes composition at the genome and buffer level alongside lot-specific QC data. Running every vendor quote through these same categories, cost per milligram, format-matched yield, lot-level reproducibility data, automation validation, and full formulation disclosure, is what turns a procurement decision at this volume from a guess into an audit.