You saved the screenshot somewhere. Type a few words from it, and it comes up.

• Federal Circuit opinion, 10X Genomics v. Parse Biosciences, Nos. 2025-1199, 2025-1618 and 2025-1619, decided August 19. Quote the opinion, not the coverage.

• The three patents from the appeal: 10,155,981, 10,240,197 and 10,697,013. Check which claims the board cancelled.

• The August 28 verdict: confirm US$4.8 million and that it covers Evercode Whole Transcriptome sales.

• The Spatial Genomics and Caltech complaint, filed August 5 in Delaware. Get the five patent numbers.

• Ask 10x and Parse for comment before Friday.

Single-cell sequencing tells you which genes are active in each cell, but the cells are taken out of the tissue first, so you lose where they were. Spatial transcriptomics keeps that map: it measures gene activity in a thin slice of tissue and records where each reading came from.

Some methods capture RNA on a slide printed with position barcodes and then sequence it. Others image the RNA inside the tissue under a microscope. Xenium, the platform in the Caltech case, is one of the imaging kind.

As Cátia Moutinho puts it, the hard part is often the tissue, not the technology.

Three replies to the last issue asked versions of the same things. Answer them in a box after the three cases.

• Should I stop using a kit I already have? No court has ordered a product off the market so far. 10x has said it will ask for an injunction against Parse, so say what that would mean.
• Who decides whether a patent is valid, the court or the patent office? Both can. Explain inter partes review in two sentences.
• Is Xenium affected by the August rulings? Only the Spatial Genomics and Caltech case names it, and there is no decision in that case yet.

Between August 5 and August 28, three decisions landed in the fight over who owns the tools for reading single cells.

Spatial Genomics and Caltech sued 10x Genomics in Delaware over five patents they say the Xenium platform uses. The Federal Circuit agreed with the patent office that three 10x patents on single-cell barcoding were obvious. Nine days later, a Delaware jury found that Parse had infringed three patents 10x acquired with Scale Biosciences, and awarded US$4.8 million.

If you run a core facility or are about to buy an instrument, none of this changes what works at the bench this week. It may change what you can buy next year, and on what terms.

Cátia Moutinho ends her August roundup with advice I want to build on: stay informed, and ask the company before you buy. It sounds obvious, but most labs I talk to choose a platform on data quality and price and never ask about pending cases.

Her post points to NanoString’s CosMx, now sold by Bruker, as an earlier case where a patent dispute touched what labs could buy. Andrew P. Han at Ion Genomics reported that a judge sent the Scale patents case against Parse to a jury instead of throwing those patents out.

The IPWatchdog summary of the appeal is worth reading slowly. The court found the barcoding claims were an obvious combination of earlier work. That is a narrow legal point, not a verdict on whose chemistry is better.

What I want readers to leave with: a short list of questions for a sales rep, and a clear sense of which rulings are final and which are not.

Call on September 18 with the manager of a university genomics core. Asked not to be named.

They run droplet single-cell and one spatial platform. The choice came down to sample types, the service contract and what their users already knew. Nobody raised patents during the purchase.

Two PIs asked their vendor reps about the lawsuits after August. Both got a general answer and nothing in writing.

The worry is less the instrument than the kits. If a kit line is withdrawn or repriced, a study that is half done has to change chemistry.

Quote to use: “We budget reagents three years ahead. A court case can outlast that.”

Subject lines to test for issue 42. Pick one by Thursday.

• Three rulings, one market
• Before you sign the quote
• Who owns the barcode?
• The patent fights, in plain terms

Practical titles did better than clever ones last time, so the second is the favourite.

Name the product, not only the company. Readers buy Xenium or Evercode, not “10x” or “Parse”.

Say which court and what stage. An appeal decision, a jury verdict and a new complaint are three different things, and a reader should never have to guess which one a paragraph is about.

Link the primary source first, then the coverage. One chart per issue, with its source printed on it. No predictions about who will win.