ViroBioTree: A Tree-Structured Biological Evidence Retrieval Framework for Viral Protein Function Annotation
Tinglian Lai, Fuguo Liu, Guodong Li, Liyan Hua · 2026
Read ViroBioTree: A Tree-Structured Biological Evidence Retrieval Framework for Viral Protein Function Annotation by Tinglian Lai, Fuguo Liu, Guodong Li, Liyan Hua (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13307837.txt.
Description
Tinglian Lai and colleagues present ViroBioTree in 2026, a system that organises evidence about viral proteins as a tree of linked facts instead of chopped text passages. Standard retrieval tools fragment relationships between gene regions, domains, motifs, and annotations. Their framework converts sequence, structure, and model attention evidence into typed nodes with traceable connections.
Retrieval then runs through deterministic recall, evidence-aware reranking, balanced selection, and rule-based checks before writing cited reports. Demo tests scored perfectly on structure explanation tasks, though the authors stress those scores measure agreement with expected tags, not biological truth. A full SARS-CoV-2 index held nearly two hundred thousand nodes. Against a vector retrieval baseline, performance varied by task. The system works offline and avoids claiming wet-lab validation it has not done.
Catalog record
| CREATOR | Tinglian Lai, Fuguo Liu, Guodong Li, Liyan Hua |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61820 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |