Year: 2026
Type: preprint
Abstract: AI agents now run sequencing pipelines, but their outputs are checked by rubrics or language-model judges, not against an independent measurement. 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.
Authors: Yang Liu, Qiuyang Zhang, Sheng Li
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7223060573