Your first iPSC experiment, from line choice to first result

A principal investigator has decided the laboratory needs a human iPSC model. Nobody in the group has cultured a pluripotent line. The first order is about to be placed, and the most likely outcome, by a wide margin, is that eight months later there is a freezer full of vials, no authenticated line, and a differentiation that works some weeks and not others.

That outcome is a sequencing problem rather than a skill problem: the expensive, hard-to-reverse decisions get made first, and the cheap ones that would have informed them get made later.

This page reorders them. The short version: for a first experiment, do not buy an iPSC line. Buy the differentiated cells you actually need, run the experiment, and only take on pluripotent culture if the experiment tells you that you have to.

Step 1, Establish whether you need an iPSC line at all

This is the decision worth the most money and it takes ten minutes.

The test: write down the exact cell type and genotype you need. Then search three catalogue suppliers for it. If it exists as a vial, buy the vial. In our catalogue work, cortical neurons, motor neurons, astrocytes, microglia, cardiomyocytes, hepatocytes, endothelial cells and myoblasts are all catalogue items with published prices from at least one supplier, in normal and in several named disease genotypes.

The buy-versus-build arithmetic in detail, with the failure rate included, is in buy versus differentiate: an iPSC neuron cost model.

If you still need the line, keep reading. If you do not, skip to Step 5, the QC and receipt sections apply to catalogue cells too, and they are where first experiments most often go wrong.

Step 2, Choose the source, and understand what the price tier means

Four routes to a line, with published fees where they exist.

Three things to take from that.

The academic/commercial multiplier is real, published and large. Coriell’s NIGMS tier is exactly 1.58× and RIKEN’s is exactly 2.00×. If you are a commercial buyer, that is your price and there is no negotiating it down. If you are academic, confirm your eligibility in writing before you budget at the academic rate, eligibility is defined by the repository, not by your self-description. See academic versus commercial pricing.

Reprogramming your own material is roughly three times the cost of buying a line, and it is the correct choice exactly when the donor is the point. If the science is about a specific patient, there is no catalogue substitute. If it is not, you are paying three times for a line you could have bought.

Repository lines come with a public record, and that is worth more than it looks. A deposited line has an identifier other laboratories can cite, a documented provenance, and in several cases matched material from the same donor, Coriell’s record for GM23338 cross-references the parent fibroblast GM23248, a lymphoblast line GM20431, and DNA preparations from the same donor. If your programme will ever need a matched fibroblast or matched DNA, buying the matched set from one repository at the outset is much easier than reconstructing it later. See /catalog/matched-donor-material-sets.

Order the isogenic control at the same time

If your experiment compares a disease line against a control, the control is a separate SKU with separate stock and a separate lead time, and buyers routinely end up holding a mutant line and no control.

Published disease-line pricing shows the premium: iXCells lists motor neurons at $739.00 for the normal line, $1,037.00 for a sporadic ALS line and $1,537.00 for a SOD1 A4V homozygous line, so the disease genotype roughly doubles the price. Neural stem cell equivalents run $1,040.00 normal against $1,155.00 for the disease variants. Disease and reporter lines have lumpy, unadvertised availability; a line can be listed and still be months out.

Reserve both lines at once and get the reservation in writing. The comparison you are designing is only valid if both arms exist. See isogenic control cell lines.

Step 3, Read the licence before you standardise anything on the line

This is the failure that costs the most and is discovered latest.

Research use is granted as standard by essentially every supplier and repository. The right to provide a fee-for-service using the material, and the right to sell a product derived from it, generally are not. They are separately licensed, separately priced, and in some cases must be negotiated with an upstream depositing laboratory rather than with the party selling you the vial.

One vendor publishes the clearest statement we have found on any product page: the underlying iPSC technology is licensed from iPS Academia Japan, and commercial use of the cells is not allowed without an appropriate licence from iPS Academia Japan. That is on a product page for a differentiated product, and the same structure sits behind many iPSC-derived catalogue items.

Ask three questions in writing before the first order, not before the first paper:

1. Does this licence permit internal research only, or also fee-for-service work for third parties?
2. Does it permit selling a product derived from the material, and if not, who grants that?
3. Are there donor consent restrictions that constrain sequencing or data deposition?

The full treatment is in cell line licensing for commercial use. Start it at specification time. It runs in parallel with everything else and it is the only item on this page that can invalidate the entire programme retrospectively.

Step 4, The QC panel: what it costs and when each test becomes non-optional

A laboratory new to iPSC work systematically under-buys QC, because none of it produces a result. It is the cheapest insurance available.

The finding worth acting on: the same STR assay ranges from $65 to $250 per sample. The spread has three causes and only one is about quality, institutional eligibility, the number of loci amplified, and whether database comparison and expert interpretation are bundled with the number. A core that returns an allele table has not done the same work as a service that compares against a curated database and interprets stutter and off-ladder alleles.

Karyotype is over half of a defensible release panel and it is the one people skip. The reason not to skip it is published: a large screen of ethnically diverse human pluripotent stem cell lines identified a recurrent chromosome 20 minimal amplicon conferring a growth advantage in culture. Culture-acquired karyotypic change is a known, characterised phenomenon, not a theoretical risk, and an assessment of established techniques for determining developmental and malignant potential of human pluripotent stem cells exists precisely because the field needed to agree what the tests actually establish.

Full price detail, including sterility and endotoxin, is in what cell QC testing costs.

Line-to-line variability is the reason all of this matters

Two published findings should shape your experimental design from day one.

Common genetic variation drives molecular heterogeneity across human iPSC lines, the donor background is a variable in your experiment whether or not you treat it as one. And published guidelines on addressing variability in iPSC-derived models of human disease exist specifically to promote reproducibility, because the problem was widespread enough to warrant them. The ISSCR standards for human stem cell use in research, and the published work on implementing them, are the current reference point for what rigorous practice looks like.

The practical consequence for a first experiment: one line is an anecdote. If the finding matters, you will eventually need multiple donor backgrounds or a properly isogenic pair. Plan the budget knowing that, rather than discovering it at review.

Step 5, MODELLED first-experiment budget

This is a model. Every unit price is real, sourced and dated. Every quantity is our assumption. It is not a quotation.

Route A, Buy differentiated cells (recommended for a first experiment)

A 96-well cortical neuron experiment, one plate, no pluripotent culture at all.

Route B, Buy a line and differentiate it yourself

The same experimental endpoint, starting from a pluripotent line.

Read those two subtotals against each other. Route A delivers a result for roughly $4,000 in materials and two to four weeks. Route B spends roughly $6,200 in materials before a single neuron exists, and the excluded lines, differentiation reagents and four to eight months of technician time, are larger than everything shown.

Route B should be a decision, taken because the genotype or the derivative-cell-type breadth requires it, rather than a default taken because “we need iPSCs”.

Step 6, The timeline

The MTA is the one people forget, and it runs in series. Two to six weeks of institutional paperwork sits between your purchase order and your vial, and no amount of urgency compresses another institution’s legal review. Start it the day you choose the line.

For a first-ever iPSC experiment, add contingency on top of all of it. The published phase durations assume a laboratory that already knows how to do this. See realistic lead times and, if there is a hard date, sourcing under a grant deadline.

Step 7, The day it arrives

Do all of this on delivery day. It takes an afternoon and it is the difference between a recoverable problem and an unexplained one.

• Photograph the package before opening. Record the coolant state and read any temperature logger.
• Verify the lot against the certificate of analysis. Check the passage number and the counting convention on the label.
• Count viability by the vendor’s stated method, not by a method you prefer. Trypan blue reads generously on debris-laden post-thaw samples and AO/PI does not; a vendor specification measured by trypan blue will not agree with your AO/PI count, and yours will be lower. See post-thaw viability.
• Seed an attachment control and count at 24 hours. That number predicts your experiment better than any figure on the datasheet.
• Send the mycoplasma sample. Three to five business days at $167.00 is cheap relative to finding out later.
• Raise any discrepancy inside the claim window, which is usually short, starts on delivery, and lives in the general terms rather than on the product page. Find it before the shipment arrives, not after.

The fifteen items to have asked for by this point are in the cell supplier QC checklist.

What goes wrong for first-time laboratories

Buying the mutant without the control. Separate SKU, separate stock, and the comparison is worthless without both. Reserve them together.

Planning cell numbers on the label rather than the specification floor. A vial labelled “1 × 10⁶ cells, ≥ 80% viability” on a freeze basis gives you 800,000 usable cells, not a million. The full correction set is in vials to wells.

Ordering cells without media, supplement and coating on the same purchase order. These are frequently separate SKUs with separate lead times, and the shortest path to a working experiment is set by the slowest line item.

Letting a lot boundary run through the middle of the experiment. Two lots of the same catalogue number can differ enough to break a comparison. Reserve the whole quantity from one lot and get it stated on the order confirmation.

Not recording passage number from day one. Passage number and population doublings are not interchangeable, and a line that has been through an undocumented number of doublings before it reached you carries a history you cannot reconstruct. See passage number and population doublings.

Treating one line as a result. Donor background is a variable. Design for a second line or an isogenic pair before a reviewer asks for it.

Leaving the licence to the end. It is the only failure on this list that cannot be fixed retrospectively.

The specification to send

The line, or the differentiated cells

• Exactly what you need: [cell type, subtype, genotype]. If differentiated cells exist as a catalogue item for this specification, say so and ask both routes to be quoted
• Source class: commercial catalogue / public repository / contract reprogramming
• Genotype and, if a disease line, the matched isogenic control as a named separate line, reserved at the same time
• Donor constraints: sex, age, ethnicity, and whether matched material from the same donor (fibroblast, lymphoblast, DNA) is required now or may be later
• Reporter requirement: none / GFP / GCaMP / other
• Quantity and pack size, with the price per million at each pack size offered
• Counting convention: total at freeze or viable post-thaw, stated explicitly
• Lot: entire quantity from one lot, reserved, stated on the order confirmation
• Passage number at shipment, and the population doubling history if known

QC deliverables at this price

• Certificate of analysis
• Post-thaw viability specification floor, method and timepoint
• STR identity profile, included or chargeable
• Mycoplasma status and method
• Karyotype, included or chargeable, and how recently performed
• Pluripotency panel for a pluripotent line, markers and method
• Sterility status

Paperwork and rights

• Is an MTA required, and can it be provided now for parallel institutional review
• Use rights: internal research / fee-for-service / commercial derivative
• Any upstream licence required from a third party, named
• Donor consent restrictions affecting sequencing or data deposition
• Import permit requirements, for an international source

Logistics

• Requested delivery date and ship day; no Friday deliveries
• Shipping mode and cost, quoted as a separate line
• Media, supplement and matrix: included or excluded, quoted separately if excluded
• Claim window in days, and the remedy for a lot failing its viability specification

Pricing tier

• State your institution class explicitly. Published repository multipliers of 1.58× and 2.00× between academic and commercial tiers are not negotiable, but eligibility is worth confirming in writing before you budget

Where we fit

We put that specification to named suppliers and repositories, normalise the responses onto one unit and one counting convention, and check MTA requirements, lot availability and the isogenic control’s stock position before you raise a purchase order.

For most first experiments the right advice is also the one that generates us the least work: buy the differentiated cells, run the experiment, and take on pluripotent culture only when the science demands it.