The NGS Market Was Hiding in PubMed
We build NGS market models the hard way: bottom-up, segment by segment, triangulated with surveys and operator interviews. The latest edition puts the worldwide NGS manufacturer market at about $7.3B in 2025, growing ~7% a year to ~$8.9B by 2028. That number costs real work to produce.
So here is a question worth asking: how much of it could you have seen for free, just by reading the literature?
Two instruments, one target
Separately from the market model, we built the NGS Diffusion Atlas — a read on how next-generation sequencing spread through science, classified across 98,417 papers by application, not by who funded them. It is a bibliometric instrument. It knows nothing about revenue. It only knows what scientists chose to work on, and when.
Point the two instruments at the same target and something nice happens: they agree.
The clinic, by two completely different measures
Express both in the same unit — the share of NGS that is clinical — and the lines nearly touch. By publications, clinical and diagnostic work rose from about 8% of NGS papers in 2008 to 49% in 2025. By dollars, clinical is now roughly $3.7B of the $7.3B market, or about 51%, and accounts for ~80% of all the growth we forecast through 2028.
Forty-nine percent and fifty-one percent. One number comes from a language model reading abstracts; the other from a bottom-up revenue build. They were not tuned to match. They match because they are measuring the same migration from two sides.
The chart on the Atlas makes the point that matters: the publication curve has been climbing for fifteen years. The dollars only crossed half recently. The literature was a leading indicator the whole time.
Where the publications went first, the money followed
Liquid biopsy is the cleanest example. In the Atlas, liquid-biopsy applications show up late — first papers around 2009 to 2013 — and then climb hard. In the market, that same wave is now the fastest-growing source of demand: Natera's Signatera oncology volume grew 54% year over year to roughly 211,000 tests in Q3 2025, and each MRD test consumes an order of magnitude more sequencing than a tissue panel. The papers called it before the revenue line bent.
The most useful disagreement: research publishes, the clinic pays
The two instruments do not agree everywhere, and the gap is the insight. Academic and population research are the oldest, highest-volume lanes in the Atlas — science's center of gravity for a decade. But academia is only about 25% of the manufacturer market and growing ~1% a year. Research generates papers out of proportion to the dollars it spends; the clinic generates dollars out of proportion to the papers it publishes. A bibliometric tool will always over-weight discovery. Knowing exactly how much is what lets you use it honestly.
That asymmetry is the whole NGS story right now. Sequencing data volumes grew 30–40% a year since 2022; reagent revenue grew about 3%. Reads got cheap; the clinical decisions they enable got valuable. And the value accrued accordingly: through 2025, Guardant Health was up 221% and Natera up 52% on the year, while Illumina — which sells the reads — was down 2%. The market is no longer a platform arms race. It is an infrastructure race, and the scoreboard is in the clinic.
What this is, and what it isn't
The Atlas measures scholarly diffusion — what the research community works on. The market model measures manufacturer revenue — instruments, reagents, and software, not the diagnostic services on top. They are different things. The point is not that one substitutes for the other. The point is that the cheap, public, real-time signal (publications) and the expensive, proprietary, bottom-up one (the market model) tell the same story — and the public one runs a few years ahead.
Which makes the obvious next question the interesting one: where are the publications lighting up now that the dollars haven't followed yet? In our data, early cancer detection and rare-disease whole-genome work are doing exactly what liquid biopsy did a decade ago. If the pattern holds, that is where the next $800M of incremental NGS demand is already being written, one abstract at a time.
You can read the market in PubMed. You just have to know which half to trust.
Market figures are from our 2025 NGS Market Report (11th edition), as shared with investors on a Cowen call. The Diffusion Atlas and its methodology are here.