Science1 distinct publisher3 min readPublished
Roughly 300 of 6,749 archived genomes carry the Calvin-Benson gene set, a 4.4% hit rate. Because the hits are catalogued strains with recorded growth conditions, the shortlist can be ordered and assayed.
The Scientist · Science desk

science
Sex becomes a setting in cloning: a CRISPR cut makes female clones of a male mouse1 distinct publisher
Compiled by The ScientistSomething wrong?How this is made
Having the genes for the Calvin-Benson cycle is one thing; running the cycle is another, and that gap is where the two years went. The genomes were already in hand; the slow work was cross-referencing each genomic hit against a scattered literature to see which organisms had ever been observed fixing carbon [4][5]. Phys.org reports that the genomic data and that literature had never been systematically linked [13], which is the unglamorous gap that lets a collection this size hold candidate carbon fixers nobody had listed. "To be honest, this was the kind of research I'd been hoping someone else would take on," postdoctoral researcher Arisa Nishihara told phys.org [12].
The hit-rate arithmetic is worth doing. Roughly 300 gene-set carriers out of 6,749 sequenced genomes is 4.4% [1]. It is a screen rather than a census: those same 6,749 genomes are 45% of the roughly 15,000 prokaryotic strains the Japan Collection of Microorganisms distributes [2], so more than half the shelf has not been read this way. The sequenced set is also lopsided, with 487 archaeal genomes against 6,262 bacterial ones, or 7.2% archaea [3], so anything the survey implies about archaeal carbon fixation rests on a thin base.
One number needs care. The published account puts potential CO2-fixing strains at 173 [3] while describing roughly 300 genomes with the gene set [1], and the narrative does not set out the criteria that take one figure to the other [4]. The literature cross-check is the plausible filter. Until that funnel is visible, 173 is the conservative count and 300 the permissive one.
A gene inventory tells you which organisms carry the pathway, not how fast any of them runs it. Nothing reported here says how much CO2 any strain fixes per hour, which electron donor it needs, or whether it does either at a density a vessel could use. Senior Research Scientist Shingo Kato frames the appeal plainly: plant photosynthesis fixes CO2 but requires light, many microorganisms fix it in the dark, and harnessing that "in places where light doesn't reach" is the stated aim [11]. That is grounds for looking further, short of a demonstrated process.
What lifts this above a spreadsheet is provenance. Much of the microbial world has never been grown in a laboratory at all, which is what "microbial dark matter" refers to [14], so carbon-fixation candidate lists usually begin with organisms nobody can obtain. These strains grow, and the collection records both the environment each was isolated from and the conditions under which it grows [9], inside an operation built around supplying other labs reproducibly [8]. Tsukuba houses the freezers [15], and the strains inside them are available to order.
So the claim the evidence earns is narrow and real. On the order of 173 archived strains now have a genomic and published case for CO2 fixation strong enough to justify an assay, and each arrives with a culture recipe attached. Whether any of them fixes carbon fast enough to interest a process engineer is the next experiment; this paper only sets up that question.
Ranked by verification strength, evidence, and original report placement.
Researchers in the Microbe Division of the RIKEN BioResource Research Center analyzed the genomes of approximately 6,700 microbial strains and found that roughly 300 possess a set of genes involved in the fixation of CO2.
Phys.org reports the result as microbial 'dark matter' yielding 173 potential carbon dioxide-fixing strains.
The work is published in the journal Microbes and Environments.
The team examined the genomes of 6,749 JCM strains for genes associated with the Calvin-Benson cycle, and at the same time cross-referenced the data with the literature to determine whether the microorganisms were actually performing CO2 fixation.
The project to discover CO2 fixation capabilities was a painstaking two-year analysis process.
Of the 15,000 prokaryotic strains JCM makes available, full genomic data have already been analyzed for 6,749 of them: 6,262 bacteria and 487 archaea.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · September 3, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Peer-reviewed screen, one retelling
The counts are refreshingly specific — 6,262 bacteria, 487 archaea, 306 hits across 147 genera, split 74 studied against 73 unstudied — and they come from a paper in Microbes and Environments. Two things hold the score down. Everything we have is Phys.org's single account, which opens with rounded figures it never squares with the exact ones it gives later. And the central finding is an inference from the presence of genes: not one strain in the shortlist has been shown to take up carbon in culture.
Orderable, not yet ordered
What the reporting genuinely establishes is a delivery route: JCM already ships more than 4,000 strains a year and keeps each one's habitat and growth conditions on file, so a shortlisted candidate can be requested rather than merely read about. What it does not show is anyone requesting one. No follow-up culture, no assay, no group outside RIKEN yet working with the 173.
Careful text, generous headline
The prose is honest — 'potential', 'may', 'suggest' do a lot of work — and then the headline sells 173 carbon dioxide-fixing strains and the motivation reaches all the way to a low-carbon society, which is a long way from a gene list. The other bit of stretch is arithmetic theatre: readers meet roughly 300 hits in the second paragraph and 173 in the title, and only deep in the piece learn that the smaller figure is the interesting subset rather than a walk-back.
An archive demonstrating its own worth
Subject and promoter are the same institution. Phys.org's opening calls RIKEN BRC 'unique, world-class', ties the centre's quality control to the credibility of science generally, and states outright that the project set out to make full use of the JCM collection. That is a legitimate scientific result and simultaneously a showcase for a resource whose funding case is demand for its strains. Every voice in the piece is on the team.
Plausible, unchecked, single channel
The claims are the kind that are usually right — a gene-presence screen over a catalogued collection is hard to get badly wrong — but our confidence is capped by having one publisher, one set of interested sources, a paper we see only through summary, and internal number drift left unexplained. Enough to act on if you are ordering strains; not enough to quote as settled biology.