Open problem arising from Tanaka and Kawai's partial sequencing of a single DNA molecule by scanning tunnelling microscopy (Nature Nanotechnology, 2009): extending label-free, single-molecule, base-resolved readout from the one base they could identify — guanine, via its characteristic electronic state on Cu(111) — to all four nucleobases in individual DNA molecules, with measurement protocols fast, robust and well-characterized enough to determine contiguous stretches of real DNA sequence. Base discrimination by electron tunnelling was achieved by 2010–2012 in electrode-gap configurations, and at single-base resolution by low-temperature tip-enhanced Raman spectroscopy for short designed strands in 2024; however, on-surface identification of cytosine and thymine within long DNA strands and practical throughput remain open, and the physical origin of the guanine-only signature on Cu(111) remains an open explanatory sub-question.

Background

A scanning tunnelling microscope (STM) images a surface by raster-scanning an atomically sharp conductive tip above it; the tunnelling current between tip and sample decays exponentially with their separation, so the microscope resolves structures at the atomic scale. In its spectroscopic mode, scanning tunnelling spectroscopy (STS), the differential conductance dI/dVdI/dV (the derivative of the tunnelling current I with respect to the sample bias V) measured at each point probes the local density of states (LDOS) — the distribution of electronic energy levels — of the surface and of any molecule adsorbed on it. Because the four DNA nucleobases (adenine, A; cytosine, C; guanine, G; thymine, T) differ in their electronic and vibrational structure, it has long been attractive to try to read the base sequence of a single DNA molecule directly, without the amplification, labelling and replication that conventional sequencing uses. Label-free single-molecule readout would additionally preserve epigenetic modifications and local molecular structure that biochemical processing destroys.

Attempts to image DNA by STM date to the technique's early years, but were plagued by unreliable sample preparation — long DNA chains are hard to deposit flat and intact on a conductive surface — by reproducibility problems, and by spurious graphite features that mimicked DNA images. Transverse STS of DNA bundles (Shapir et al., Nature Materials 7, 68–74, 2008) resolved molecular electronic structure but not a base sequence. Tanaka and Kawai (Nature Nanotechnology 4, 518–522, 2009) broke through the preparation barrier with an oblique pulse-injection method (OPI): a 1.5 ms pulse of aqueous DNA solution is fired at a Cu(111) surface inclined at about 45 degrees, which stretches and fixes single-stranded DNA (ssDNA) of the phage virus M13mp18 (7,249 bases) as extended linear chains under ultrahigh vacuum (UHV) at about 80 K. They found that guanine bases — whether in short designed oligomers or inside the deposited viral DNA — exhibit a characteristic peak in dI/dVdI/dV at a sample bias of about −1.6 V that the other three bases and the bare Cu(111) substrate do not show. Mapping this guanine-specific peak along individual molecules, and comparing images taken at two biases to reject contaminants that lack bias-dependent contrast, they matched the observed bright features to the known guanine positions of the M13mp18 sequence — partial sequencing of a single DNA molecule.

The physical picture behind this success was, however, incomplete. Density functional theory (DFT — the standard first-principles electronic-structure method) calculations by Kilina et al. (Journal of Physical Chemistry C 111, 14541–14551, 2007), cited in the source paper (whose acknowledgements thank Kilina and Balatsky), found that all four bases adsorbed on Cu(111) hybridize strongly with the metal: the calculated local density-of-states spectra show no purely localized molecular states for any of the four bases, but a base-dependent delocalization — smallest for adenine and thymine, largest for guanine and cytosine — offered as guidelines for interpreting DNA spectra measured through tunnelling in STM and for the possible differentiation of DNA nucleotide sequences by STM. UHV-STM studies of the four free nucleosides (base-plus-sugar units) by Xu, Endres and Arakawa (Small 3, 1539–1543, 2007) saw base-specific electronic signatures; yet in the in-strand experiment only guanine produced a clear characteristic state. Guanine's anomalously low ionization potential (the energy needed to remove its most weakly bound electron; Yang et al., PNAS 101, 17588–17592, 2004) plausibly places its highest occupied molecular orbital alone within the accessible bias window, but no quantitative account of why only guanine's in-strand state was resolved has been established; this origin question remains open background to the sequencing goal. The authors accordingly listed what remained open: identifying all four bases — suggesting vibrational spectroscopy via inelastic electron tunnelling spectroscopy (IETS, in which tunnelling electrons excite molecular vibrations, producing characteristic spectral fingerprints) as a possible route; the slowness of point-by-point spectroscopy (seconds to minutes per point, hour-scale maps) and the missing software, automation and faster protocols for practical sequencing; contamination discrimination and robust base identification; the incompletely elucidated fluid-mechanical mechanism by which OPI stretches and fixes the DNA; and instrumental limits such as tip–sample electric-field effects that can bend, migrate or decompose the molecule. A modern relative of the suggested vibrational route is tip-enhanced Raman spectroscopy (TERS), in which a plasmonic STM or AFM tip concentrates light to a subnanometre spot so that the Raman vibrational spectrum of a single molecule segment can be recorded.

Problem Statement

Complete the programme that Tanaka and Kawai's 2009 demonstration began: turn guanine-only electronic readout into a method that sequences individual DNA molecules by scanning tunnelling microscopy. For single DNA molecules prepared as extended, intact strands on a surface, this requires both of the following:

  1. Base identification. Provide a measurable, reproducible, base-specific signature — electronic or vibrational — by which each of the four nucleobases A, C, G and T can be identified at single-base resolution within the strand, either directly or through an explicitly validated indirect scheme, with quantified per-base assignment error rates validated on sequences held out from any calibration and including at least one natural (non-designed) DNA molecule, and with contaminant features demonstrably excluded.

  2. Usable readout. Do so with measurement and analysis protocols whose speed, stability and robustness — reliable sample preparation and stretching of single-stranded DNA on the surface, automated chain-finding, registration of spectroscopic maps to sequence position, tolerance of contaminants and of tip–sample electric-field effects — suffice to determine, in order, a contiguous stretch of sequence of at least 1,000 bases on a natural DNA molecule (with the source demonstration's 7,249-base viral strand as the reference target, not short designed oligomers) at a per-base assignment error rate below 5%, rather than isolated guanine clusters.

These two components serve one overarching objective — label-free, single-molecule DNA sequencing by scanning tunnelling microscopy: the base-identification signature is necessary but not sufficient, while the long-strand read already entails it. The physical origin of the guanine-only in-strand signature on Cu(111) and its relation to the DFT results of Kilina et al. (2007), together with the noted questions on deposition fluid mechanics and instrumental limits, remain background context and open sub-questions, not part of what the programme must achieve.

References

  1. Tanaka, H. & Kawai, T. Partial sequencing of a single DNA molecule with a scanning tunnelling microscope. Nature Nanotechnology 4, 518–522 (2009). doi:10.1038/nnano.2009.155, https://doi.org/10.1038/nnano.2009.155
  2. Porath, D. A DNA sequence scanned (News & Views). Nature Nanotechnology 4, 476–477 (2009). doi:10.1038/nnano.2009.212, https://doi.org/10.1038/nnano.2009.212
  3. Shapir, E., Cohen, H., Calzolari, A., Cavazzoni, C., Ryndyk, D. A., Cuniberti, G., Kotlyar, A., Di Felice, R. & Porath, D. Electronic structure of single DNA molecules resolved by transverse scanning tunnelling spectroscopy. Nature Materials 7, 68–74 (2008). doi:10.1038/nmat2060, https://doi.org/10.1038/nmat2060
  4. Kilina, S., Tretiak, S., Yarotski, D. A., Zhu, J.-X., Modine, N., Taylor, A. & Balatsky, A. V. Electronic properties of DNA base molecules adsorbed on a metallic surface. Journal of Physical Chemistry C 111, 14541–14551 (2007). doi:10.1021/jp070805u, https://doi.org/10.1021/jp070805u
  5. Xu, M., Endres, R. G. & Arakawa, Y. The electronic properties of DNA bases. Small 3, 1539–1543 (2007). doi:10.1002/smll.200600732, https://doi.org/10.1002/smll.200600732
  6. Yang, X., Wang, X.-B., Vorpagel, E. R. & Wang, L.-S. Direct experimental observation of the low ionization potentials of guanine in free oligonucleotides by using photoelectron spectroscopy. Proceedings of the National Academy of Sciences USA 101, 17588–17592 (2004). doi:10.1073/pnas.0405157101, https://doi.org/10.1073/pnas.0405157101
  7. Tsutsui, M., Taniguchi, M., Yokota, K. & Kawai, T. Identifying single nucleotides by tunnelling current. Nature Nanotechnology 5, 286–290 (2010). doi:10.1038/nnano.2010.42, https://doi.org/10.1038/nnano.2010.42
  8. Chang, S., Huang, S., He, J., Liang, F., Zhang, P., Li, S., Chen, X., Sankey, O. & Lindsay, S. Electronic signatures of all four DNA nucleosides in a tunneling gap. Nano Letters 10, 1070–1075 (2010). doi:10.1021/nl1001185, https://doi.org/10.1021/nl1001185
  9. Huang, S., He, J., Chang, S., Zhang, P., Liang, F., Li, S., Tuchband, M., Fuhrmann, A., Ros, R. & Lindsay, S. Identifying single bases in a DNA oligomer with electron tunnelling. Nature Nanotechnology 5, 868–873 (2010). doi:10.1038/nnano.2010.213, https://doi.org/10.1038/nnano.2010.213
  10. Tsutsui, M., Matsubara, K., Ohshiro, T., Furuhashi, M., Taniguchi, M. & Kawai, T. Electrical detection of single methylcytosines in a DNA oligomer. Journal of the American Chemical Society 133, 9124–9128 (2011). doi:10.1021/ja203839e, https://doi.org/10.1021/ja203839e
  11. Ohshiro, T., Matsubara, K., Tsutsui, M., Furuhashi, M., Taniguchi, M. & Kawai, T. Single-molecule electrical random resequencing of DNA and RNA. Scientific Reports 2, 501 (2012). doi:10.1038/srep00501, https://doi.org/10.1038/srep00501
  12. Ribot, J. C., Chatterjee, A. & Nagpal, P. Measurements of single nucleotide electronic states as nanoelectronic fingerprints for identification of DNA nucleobases, their protonated and unprotonated states, isomers, and tautomers. Journal of Physical Chemistry B 119, 4968–4974 (2015). doi:10.1021/acs.jpcb.5b01403, https://doi.org/10.1021/acs.jpcb.5b01403
  13. Di Ventra, M. & Taniguchi, M. Decoding DNA, RNA and peptides with quantum tunnelling. Nature Nanotechnology 11, 117–126 (2016). doi:10.1038/nnano.2015.320, https://doi.org/10.1038/nnano.2015.320
  14. Tanaka, H. & Taniguchi, M. Sequencing of adenine in DNA by scanning tunneling microscopy. Japanese Journal of Applied Physics 56, 08LB02 (2017). doi:10.7567/JJAP.56.08LB02, https://doi.org/10.7567/JJAP.56.08LB02
  15. Afsari, S., Korshoj, L. E., Abel, G. R., Khan, S., Chatterjee, A. & Nagpal, P. Quantum point contact single-nucleotide conductance for DNA and RNA sequence identification. ACS Nano 11, 11169–11181 (2017). doi:10.1021/acsnano.7b05500, https://doi.org/10.1021/acsnano.7b05500
  16. Abel, G. R., Korshoj, L. E., Otoupal, P. B., Khan, S., Chatterjee, A. & Nagpal, P. Nucleotide and structural label identification in single RNA molecules with quantum tunneling spectroscopy. Chemical Science 10, 1052–1063 (2019). doi:10.1039/c8sc03354d, https://doi.org/10.1039/c8sc03354d
  17. Rodríguez-Galván, A. & Contreras-Torres, F. F. Scanning tunneling microscopy of biological structures: an elusive goal for many years. Nanomaterials 12, 3013 (2022). doi:10.3390/nano12173013, https://doi.org/10.3390/nano12173013
  18. Hou, B., Zhang, T., Yang, H., Han, X., Liu, L., Li, L., Grazioli, C., Wu, X., Jiang, N. & Wang, Y. Advances in probing single biomolecules: from DNA bases to glycans. Interdisciplinary Materials 2, 511–528 (2023). doi:10.1002/idm2.12106, https://doi.org/10.1002/idm2.12106
  19. Jena, M. K. & Pathak, B. Development of an artificially intelligent nanopore for high-throughput DNA sequencing with a machine-learning-aided quantum-tunneling approach. Nano Letters 23, 2511–2521 (2023). doi:10.1021/acs.nanolett.2c04062, https://doi.org/10.1021/acs.nanolett.2c04062
  20. Han, Y., Dong, L., Zhu, L.-Y., Hu, C.-R., Li, H., Zhang, Y., Zhang, C., Zhang, Y. & Dong, Z.-C. Real-space spectral determination of short single-stranded DNA sequence structures. Journal of the American Chemical Society 146, 33865–33873 (2024). doi:10.1021/jacs.4c12393, https://doi.org/10.1021/jacs.4c12393
  21. Zhu, L.-Y., Han, Y., Zhang, Y.-F., Li, H., Dong, L., Zhang, C., Zhang, Y., Zhang, Y. & Dong, Z.-C. Algorithm-assisted structure identification of individual nucleobases in single DNA chains through tip-enhanced Raman spectromicroscopy. JACS Au 5, 4835–4844 (2025). doi:10.1021/jacsau.5c00791, https://doi.org/10.1021/jacsau.5c00791
  22. Maurya, D., Mittal, S., Jena, M. K. & Pathak, B. Machine learning-driven quantum sequencing of natural and chemically modified DNA. ACS Applied Materials & Interfaces 17, 20778–20789 (2025). doi:10.1021/acsami.4c22809, https://doi.org/10.1021/acsami.4c22809
  23. Han, Y., Zhu, L.-Y., Dong, L., Li, G., Zhang, Y., Zhang, Y. & Dong, Z.-C. Probing the identification of single modification sites in short RNA strands. Journal of Raman Spectroscopy 57, 1008–1014 (2026). doi:10.1002/jrs.70094, https://doi.org/10.1002/jrs.70094