Science1 publisher2 min readPublished
Osaka chemists count mirror-image amino acids one molecule at a time
University of Osaka researchers used a gold nanogap and AI to tell L- from D-amino acids one molecule at a time, at over 80% accuracy. On meteorite and desert extracts it matched standard methods, so a compact life-detection instrument is still a hope.
The Scientist · Science desk
Drafted by a language model from the sources cited here and checked against its claim ledger before publication. How we use AISend a correction
What happened
- The team passed molecules through a gap between two gold nanowires, where each mirror form produced a distinct tunneling-current waveform that let them count molecules directly.
- Conventional chirality methods work by measuring large groups of molecules at once, an approach the phys.org account says brings practical challenges.
- The team tested the technique on extracts from Australia's Murchison meteorite and on soil from Chile's Atacama Desert, both complex natural mixtures.
- The work is published in Nature Communications as a paper on chiral discrimination of amino acids via single-molecule nanogap conductance.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Per-molecule misreads pull a true L excess toward 50:50. A flight version would need its error rate calibrated before a lopsided count could be offered as evidence of biology.
- decision For mission planners, the meteorite comparison supports parity with standard methods, so choosing this sensor would turn on size and simplicity, and those have not been demonstrated in hardware.
- capability Because the ratio comes from tallies of individual molecules, its statistical precision should in principle improve as more molecules are passed through the gap.
Telling an L-amino acid from its D mirror image is a two-way call. Guessing scores 50%, so the Osaka classifier's "over 80%" sits about 30 points above chance [1][1]. "By combining our nanogap tunneling technique with artificial intelligence, we were able to distinguish between the L- and D-forms of amino acids with over 80% accuracy," lead author Takahito Oshiro said [7][13]. He called it "the first discrimination of amino acid chirality at the single-molecule level" [8].
For life detection, the misreads matter as much as the hits. Living things use almost only L-amino acids, while nonliving chemistry makes the two forms in equal amounts, so a mission is looking for a lopsided L/D ratio [2][3]. Suppose the 80% applies to each molecule and the classifier errs equally in both directions. A sample of pure L molecules would then tally about 80:20, an apparent ratio of 4 to 1, while a 50:50 mix would still tally 50:50 [2]. The two cases stay apart. A real excess, though, would look smaller than it is until the error rate is measured and corrected for. The phys.org account does not say which amino acids were tested, how many molecules sit behind the 80% figure, or whether the errors run evenly.
The meteorite and desert samples were the harder test. Astrobiological material holds many kinds of molecules, so the team checked whether the method could still pick out amino acids in a mixture [14]. The result they report is parity on broad composition. "Our method was comparable to traditional methods, as both were capable of capturing the major features of amino acid composition," senior author Masateru Taniguchi said [10][13].
That supports the sensor on messy natural extracts. It does not yet show greater sensitivity than the instruments a mission carries now [10].
I'd give the most weight to the bench chemistry. If Oshiro's single-molecule first holds up under review, it is a result that exists today [8]. The flight instrument is a separate matter. The team hopes the work leads to compact, electrically based instruments for detecting extraterrestrial life [11]. The case for such a device rests on properties that have been suggested for electrical detection in general: simpler operation, less sensitivity to vibration, and no chemical reagents [5].
What to watch
- Whether the full paper reports accuracy for each amino acid and whether misreads run evenly in both directions, which sets how far a measured L/D ratio can be trusted.
- A blind test on samples with a known L/D ratio, showing whether the single-molecule counts recover the true excess after error correction.
- A prototype with size, mass and power figures, the evidence the compact-instrument claim currently lacks.