In a quiet corner of computational biology, a small Japanese company and a university laboratory have asked a modest question of public data: whether any compounds found in seaweed might point in the direction opposite to the changes astronauts' blood shows after spaceflight. Working through three open databases with the help of generative AI, they flagged eight candidates, among them stigmasterol, fucoxanthin and vanillin, for later study. Their findings, shared at an algae conference in August 2026, are a map of where to look next rather than a verdict on what works, a reminder that the long
Space Seed and Gifu University Flag Eight Seaweed Compounds for Lab Follow-Up Using Spaceflight Gene Data
It does not show that any compound reversed a spaceflight change.
Mark
So what did they actually do here? It sounds like a lot of database work.
Mimi
That's the core of it. They matched seaweed compounds against cell-treatment data and compared that to astronaut blood changes. Eight compounds survived the matching, and they ranked them by how many of ten genes moved opposite to the spaceflight change.
Mark
And stigmasterol came out on top?
Mimi
Seven of ten in HEK293T cells. Fucoxanthin, phytol, protocatechualdehyde and vanillin hit six in at least one cell line.
Luke
Before anyone gets excited, that seven is a count of direction, not size. A gene nudged a hair counts the same as one that swings hard.
Mark
Why does it matter that the datasets are separate?
Mimi
Because the astronauts weren't given these compounds. The cell lines were treated in a different experiment entirely. So nothing here shows a compound reversed anything.
Luke
Also only six astronauts, and only 10 usable genes. That's a very small base for ranking anything.
Mark
And the AI part?
Mimi
It helped reconcile naming conventions and build the pipeline. The team says they kept the rules simple so steps stay traceable.
Luke
Worth noting the exact-match rule means candidates may have been missed. We don't know how many.
Mark
So what's the honest takeaway?
Mimi
A shortlist for lab work. Concentration and timing tests come next.
Luke
And whether the spaceflight direction holds across other datasets is still unchecked.
El Pulso
- Eight seaweed-derived compounds emerged from a search of public databases, but only as possibilities worth testing, not proven remedies.
- Stigmasterol led the field with 7 of 10 genes moving opposite to spaceflight changes, yet the score counts direction alone, not magnitude or effect.
- The analysis stitches together separate datasets, astronaut blood and cell-line experiments, so no one can yet say a compound reverses anything in a person.
- Next steps move into cell cultures to test concentration and timing, with blood-cell and endothelial studies possibly following.
- Any suggestion of food or supplement use remains distant, requiring separate checks of safety, purity, stability and acceptable intake.
In a quiet corner of computational biology, a small Japanese company and a university laboratory have asked a modest question of public data: whether any compounds found in seaweed might point in the direction opposite to the changes astronauts' blood shows after spaceflight. Working through three open databases with the help of generative AI, they flagged eight candidates, among them stigmasterol, fucoxanthin and vanillin, for later study. Their findings, shared at an algae conference in August 2026, are a map of where to look next rather than a verdict on what works, a reminder that the long human hope of finding protection for bodies that leave the Earth still begins with careful, humble questions.
A small team set out to see whether the ocean could offer something to people who leave it behind. Space Seed Holdings and Gifu University began with three public collections: a database of seaweed chemicals, a library of how compounds change gene activity in cultured cells, and records from six astronauts whose blood was sampled before launch and again after return. Because the names and labels in these sources do not match neatly, the team used generative AI to reconcile them and to build a traceable analysis pipeline.
The rules were kept deliberately strict. Only exact name matches were accepted, which the researchers acknowledge may have excluded some candidates. From the astronaut data, they focused on ten genes showing clear changes, and for each seaweed compound they counted how many of those genes moved in the opposite direction. That tally became the score. Eight compounds passed the filter, including apigenin, fucoxanthin, phytol and vanillin, with stigmasterol scoring highest at 7 of 10 in one cell line. Results shifted between cell types, and sterols like stigmasterol are common across living things, so they are not unique to seaweed.
The researchers were direct about what this cannot show. Joining separate datasets does not prove that any compound reverses a spaceflight effect, and the score ignores how large a change is. The genes involved relate to general stress responses rather than aging specifically, so no anti-aging claim follows. Kengo Suzuki, the company's chief executive, presented the work as a poster in Malaysia and singled out fucoxanthin given the company's interest in algae materials.
The path ahead is slow and experimental. The team plans to test how these gene changes respond to dose and timing in the same cell lines, then examine other cell types and multiple public datasets. Only after that would questions of food or supplement safety, content and purity arise.
Eight compounds, fifteen matching records, ten genes. Those are the numbers at the center of an exploratory analysis from Space Seed Holdings Inc. (SSHD) and Gifu University, which used three public databases to pick out compounds reported in seaweeds and compare how they affect gene expression with changes seen in astronauts' blood after spaceflight.
The source data came from three places. The Seaweed Metabolite Database (SWMD) holds 1,191 public records of metabolites found in seaweeds. The Chemical-Induced Gene Signature database (CIGS) contains expression data for 3,407 genes across 13,221 compounds, measured in two cell lines, HEK293T and MDA-MB-231. The third, NASA OSD-53 / GEO GSE47126, has whole-blood gene-expression data from six astronauts, comparing samples taken ten days before launch with samples taken immediately after return.
The databases name compounds and genes in different ways, and cross-referencing them normally takes specialist knowledge and a great deal of manual work. The team used generative AI to reconcile them and to build the analysis pipeline, from data acquisition through compound-name matching to scoring. They kept the selection rules simple so each step could be traced.
The method was deliberately strict. Compound names were standardized, and only exact matches were kept, not approximate ones, a choice the team notes may miss some candidates. That yielded 15 matching records and eight unique compounds. From the astronaut data, the team selected genes with an adjusted p-value below 0.05, which gave 11 genes, 10 of which were also measured in CIGS. For each compound and cell line, they then counted how many of those 10 genes moved in the opposite direction from the astronauts' pre- versus post-flight difference. That count is the score.
The eight candidates are apigenin, cholesterol, fucosterol, fucoxanthin, phytol, protocatechualdehyde, stigmasterol and vanillin. Stigmasterol scored highest, with opposite-direction changes in 7 of the 10 genes in HEK293T cells. Fucoxanthin, phytol, protocatechualdehyde and vanillin each reached 6 of 10 in at least one cell line. Results varied by cell line: vanillin scored 6 of 10 in MDA-MB-231 cells and 5 of 10 in HEK293T cells. Fucoxanthin is a carotenoid found in seaweeds. Stigmasterol and other sterols are widely found in living organisms and should not be treated as specific to seaweed.
The limits are stated plainly. The comparison joins separate datasets, astronaut blood and compound-treated cell lines, so it does not show that any compound reversed a spaceflight change. Nor does it show that any compound is effective or safe in humans, or that it can be added to foods. The score counts direction only, not the size of a change. The 11 genes relate to DNA repair, molecular chaperones and oxidative stress response. These are stress responses tied to spaceflight, not specific to aging, so the results cannot be read as evidence of anti-aging effects.
The results were presented as a poster at Global Algae Summit 2026, held at Sunway University in Malaysia on August 19, 2026. Kengo Suzuki, Representative Director and CEO of Space Seed Holdings, said the work identified eight compounds worth studying using only public data, and that generative AI made the cross-referencing far more efficient. He said fucoxanthin is of particular interest given the company's work on algae-derived materials, since it is characteristic of seaweed and scored relatively high.
Next, the team plans to test how gene expression changes with concentration and treatment time in the same cell lines as CIGS, measuring the 10 genes by qPCR or RNA-seq. Tests in cells such as peripheral blood mononuclear cells and vascular endothelial cells could follow. They also need to check, across multiple public datasets, whether the direction of the spaceflight changes holds and whether shifts in blood cell composition affect the results. Any food or supplement use would require separate evaluation of content, extract purity, stability, interactions and acceptable intake.
Citas Notables
Using only public data, the team identified eight seaweed-derived compounds to study further, and generative AI made cross-referencing the databases efficient.— Kengo Suzuki, CEO, Space Seed Holdings Inc.
Quieres la nota completa?
Lee el original en The Japan Times ↗
Technology