Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 52 min ago
PreprintCC BY-NC-NDbioRxiv

LongVerse: verified AI agents for Nanopore and PacBio DNA methylation analysis

  • Y. Liu
  • Q. Zhang
  • S. Li1
  1. 1University of Southern California
Read full textPDFDOICite

Abstract

AI agents now run sequencing pipelines, but their outputs are checked by proxy, with rubrics or language-model judges. LongVerse turns one English sentence into haplotype-resolved DNA methylation from Nanopore or PacBio reads, through agents that read the data, supply arguments to a pinned pipeline and never write code. On HG002, a healthy reference genome, agent and human expert produced byte-identical calls (r = 1.000) that agreed with bisulfite sequencing (r = 0.943, 0.884), at $0.12 to $0.54 of conversation per run. A consensus agent that admits any per-site caller reduced error 10.7 percent against the best published caller.

Related

  1. Epi-PoreC: a nanopore-based method for simultaneous profiling of chromatin conformation, DNA methylation, and chromatin accessibility
    A. McKinlay, W. Zong, C. Pikaard
    bioRxivPreprintCC BYFull textGenetics & genomicsbioRxiv
  2. Research Progress on the Application of Nanopore Long-Read Sequencing Combined with DNA Methylation Marking in Complex Forensic Evidence.
    Qiannan Xu, Xuanman Cui, Shuyan Mei +3
    GenesPeer-reviewedCC BYFull textGenetics & genomicsEurope PMC
  3. A Framework for Quantifying DNA Methylation Heterogeneity and Detecting Co-methylated loci from Native Nanopore Sequencing
    Y. J. Kim, N. R. Zabet
    bioRxivPreprintCC BYFull textBioinformatics & computational biologybioRxiv
  4. Linking somatic variants to altered methylation profiles in nanopore-derived breast and pancreatic cancer genomes using unduliner
    H. C. Atanda, A. D. Ewing
    bioRxivPreprintCC BYFull textBioinformatics & computational biologybioRxiv
  5. Long-read PacBio HiFi sequencing resolves the genomic integration site and structure of the Nestin-CreERT2KEisc/J transgene
    A. M. Charest, A. Charest
    bioRxivPreprintCC BYFull textGenetics & genomicsbioRxiv

Cite

BibTeXRIS
BibTeX
@misc{liu2026longverse,
  title = {{LongVerse: verified AI agents for Nanopore and PacBio DNA methylation analysis}},
  author = {Y. Liu and Q. Zhang and S. Li},
  howpublished = {bioRxiv},
  year = {2026},
  month = oct,
  doi = {10.64898/2026.10.13.758965},
  url = {https://www.biorxiv.org/content/10.64898/2026.10.13.758965v1},
  note = {CC BY-NC-ND},
}
RIS
TY  - UNPB
TI  - LongVerse: verified AI agents for Nanopore and PacBio DNA methylation analysis
AU  - Y. Liu
AU  - Q. Zhang
AU  - S. Li
PY  - 2026
DA  - 2026/10/13
PB  - bioRxiv
DO  - 10.64898/2026.10.13.758965
UR  - https://www.biorxiv.org/content/10.64898/2026.10.13.758965v1
AB  - AI agents now run sequencing pipelines, but their outputs are checked by proxy, with rubrics or language-model judges. LongVerse turns one English sentence into haplotype-resolved DNA methylation from Nanopore or PacBio reads, through agents that read the data, supply arguments to a pinned pipeline and never write code. On HG002, a healthy reference genome, agent and human expert produced byte-identical calls (r = 1.000) that agreed with bisulfite sequencing (r = 0.943, 0.884), at $0.12 to $0.54 of conversation per run. A consensus agent that admits any per-site caller reduced error 10.7 percent against the best published caller.
ER  -