Bioinfoysis Technical Report
The DeepAutonomy Team introduces Bioinfoysis, a multi-agent harness that treats a bioinformatics request as a persistent analysis run whose conclusions stay attached to the artifacts that produced them, reaching 82.4% on BixBench.
Ask this paper
Artifacts instead of transcripts. Most agent systems treat planning, tool calls, and code execution as transient conversation. Bioinfoysis persists scripts, tables, and figures as first-class objects, so a final claim can be traced to the computation behind it.
Structured handoffs prevent stale evidence. Each worker returns a handoff that binds its result to the responsible agent, the checklist step, and the plan generation. When the planner replans, results from a superseded generation cannot be silently reused, which is the specific failure this design targets.
Validation before use, not after. A controlled runtime checks generated scripts, tables, and figures before downstream steps consume them, so a broken intermediate stops the trajectory instead of propagating.
82.4% on BixBench, state of the art, with gains holding across four different underlying language models. Consistency across backbones indicates the harness supplies the improvement rather than one model's idiosyncrasies.
Directly transferable design. Executable checklists, generation-tagged handoffs, and pre-use artifact validation are harness-level ideas that apply to any long-horizon coding or analysis agent, not only to bioinformatics.
Abstract
Large language model agents have shown promise in bioinformatics, but most existing systems focus primarily on producing final answers, treating planning, tool use, and code execution as transient interactions. This design is poorly suited to long-horizon bioinformatics tasks, where conclusions must remain connected to the data, computations, and intermediate evidence that support them. We introduce \textbf{Bioinfoysis}, a multi-agent harness that represents each request as a persistent, artifact-grounded analysis run. Bioinfoysis combines global planning with step-wise, evidence-driven replanning: the planner maintains an executable checklist and revises pending steps using structured handoffs returned after each worker execution. These handoffs bind intermediate results to their responsible agent, checklist step, and plan generation, preventing stale evidence from being silently reused after replanning. A controlled runtime validates generated scripts, tables, and figures before they are used in downstream analysis or reporting, while role-specific context, persistent memory, and governed bioinformatics skills support reliable execution over long analysis trajectories. We evaluate Bioinfoysis on BixBench and two question-answering tracks of LAB-Bench 2. On BixBench, Bioinfoysis achieves state-of-the-art accuracy of 82.4\%. Across four underlying language models, Bioinfoysis increases average accuracy from 27.81\% to 64.13\% on SeqQA2 and from 3.13\% to 31.25\% on DbQA2. These results demonstrate that reliable bioinformatics automation depends not only on model capability, but also on the harness that governs planning, execution, memory, and evidence flow. We hope that the emergence of Bioinfoysis will play a driving and leading role in the development of the bioinformatics community. Our demo website can be seen in https://report.bioinfoysis.com/.