Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 51 min ago
Peer-reviewedCC BY 4.0Life Science Alliance

An unsupervised explainable AI reveals oligonucleotide-usage zones matching 2,000 chromosome bands

  • Toshimichi Ikemura1,2
  • Yuki Iwasaki1
  • Kennosuke Wada1
  • Yoshiko Wada1
  • Takashi Abe3
  1. 1Faculty of Bioscience, Nagahama Institute of Bio-Science and Technology
  2. 2Association for Propagation of the Knowledge of Genetics, Shizuoka, Japan
  3. 3Graduate School of Science and Technology Electrical and Information Engineering, Niigata University, Niigata, Japan
Open at sourceDOICite

Abstract

Unsupervised and explainable AI can uncover genomic features beyond human expectations. Oligonucleotides such as penta- and hexanucleotides often function as core binding motifs for regulatory proteins, and their usage provides powerful tools in functional genomics. We applied an unsupervised, explainable AI to their odds-ratio profiles of all 1-Mb euchromatic fragments in the human genome. Clustering using penta- or hexanucleotide odds ratios (1,024 or 4,096 variables) unexpectedly revealed nearly 2,000 distinct zones. CG-containing penta- and hexanucleotides (244 and 1,185 variables) produced comparable ∼2,000 zones, indicating fundamental genomic segmental structures. Human chromosomes exhibit Giemsa-banding patterns comprising 850 bands at prometaphase and ∼2,000 at prophase. Using single-nucleotide coordinates available for the 850 bands, we identified diagnostic oligonucleotides distinguishing Giemsa-negative and Giemsa-positive regions. Computational pseudo-band reconstruction based on these oligonucleotides generated genome segmentations that more closely paralleled the AI-derived ∼2,000 zones than the 850 bands. These findings indicate that AI captures characteristic sequence features of chromosomal bands and can predict high-resolution banding from genome sequences alone, thereby bridging classical cytogenetics and modern AI-based genomics.

Related

  1. Cell-resolved pharmacodynamic profiling reveals tissue- and cell-population-specific activity of antisense oligonucleotide chemical conjugates in vivo
    S. Cheng, J. Frommer, A. Eddaoudi +4
    bioRxivPreprintAll rights reservedFull textMolecular biology & gene regulationbioRxiv
  2. Disentangling polymer confinement from specific-folding interactions reveals the drivers of Escherichia coli chromosome organization
    Pourya Delafrouz, Tanya Nesterova, Frances E Harris +6
    Nucleic Acids ResearchPeer-reviewedCC BY-NC 4.0Full textMolecular biology & gene regulationCrossrefEurope PMC
  3. Chromosome-level assembly of Lemnamoeba sinaloensis gen. nov. et sp. nov. (Apofilosa) reveals dynamic genome evolution in Rhizaria
    Markus Hiltunen Thorén, Ioana Onuț-Brännström, Vasily Zlatogursky +5
    BMC GenomicsPeer-reviewedCC BY 4.0Molecular biology & gene regulationCrossref
  4. Advances in therapeutic oligonucleotide conjugates
    Ya Ying Zheng, Yonger Xue, Xucheng Hou +1
    Aaps OpenPeer-reviewedCC BY 4.0Full textMolecular biology & gene regulationPMCEurope PMC
  5. Modular multifunctional 5HP2O linkers for the versatile assembly of protein-oligonucleotide conjugates
    Jan H. Meffert, Alexandros Kostakis, Joshua McHale +4
    Communications ChemistryPeer-reviewedCC BY 4.0Full textMolecular biology & gene regulationCrossref

Cite

BibTeXRIS
BibTeX
@article{ikemura2026unsupervised,
  title = {{An unsupervised explainable AI reveals oligonucleotide-usage zones matching 2,000 chromosome bands}},
  author = {Toshimichi Ikemura and Yuki Iwasaki and Kennosuke Wada and Yoshiko Wada and Takashi Abe},
  journal = {Life Science Alliance},
  year = {2026},
  month = nov,
  doi = {10.26508/lsa.202603782},
  url = {https://www.life-science-alliance.org/lookup/doi/10.26508/lsa.202603782},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - An unsupervised explainable AI reveals oligonucleotide-usage zones matching 2,000 chromosome bands
AU  - Toshimichi Ikemura
AU  - Yuki Iwasaki
AU  - Kennosuke Wada
AU  - Yoshiko Wada
AU  - Takashi Abe
PY  - 2026
DA  - 2026/11/05
JO  - Life Science Alliance
DO  - 10.26508/lsa.202603782
UR  - https://www.life-science-alliance.org/lookup/doi/10.26508/lsa.202603782
AB  - Unsupervised and explainable AI can uncover genomic features beyond human expectations. Oligonucleotides such as penta- and hexanucleotides often function as core binding motifs for regulatory proteins, and their usage provides powerful tools in functional genomics. We applied an unsupervised, explainable AI to their odds-ratio profiles of all 1-Mb euchromatic fragments in the human genome. Clustering using penta- or hexanucleotide odds ratios (1,024 or 4,096 variables) unexpectedly revealed nearly 2,000 distinct zones. CG-containing penta- and hexanucleotides (244 and 1,185 variables) produced comparable ∼2,000 zones, indicating fundamental genomic segmental structures. Human chromosomes exhibit Giemsa-banding patterns comprising 850 bands at prometaphase and ∼2,000 at prophase. Using single-nucleotide coordinates available for the 850 bands, we identified diagnostic oligonucleotides distinguishing Giemsa-negative and Giemsa-positive regions. Computational pseudo-band reconstruction based on these oligonucleotides generated genome segmentations that more closely paralleled the AI-derived ∼2,000 zones than the 850 bands. These findings indicate that AI captures characteristic sequence features of chromosomal bands and can predict high-resolution banding from genome sequences alone, thereby bridging classical cytogenetics and modern AI-based genomics.
SN  - 2575-1077
ER  -