YR-MPE: a modular desktop platform for agent-operable and reproducible phylogenetic workflows

Phylogenomic analyses combine many tools, file formats and execution modes. Desktop graphical user interfaces make individual steps accessible but hide their state from scripts; command-line pipelines make steps scriptable but leave parameter discovery, provenance and recovery to the user; and emerging LLM agents can call command-line tools but cannot inspect a GUI session or its artifacts. We present YR-MPE, a modular desktop platform whose YPhylo Workbench unifies these modes through a typed workspace, a plugin contract and a semantic action layer. Every operation in the action registry is exposed as a machine-callable action; controls without a CLI equivalent are declared GUI-only. The workspace records every imported plugin result as a run with its inputs, parameters, tool identity and output artifacts, the latter with SHA-256 hashes; each compiled workflow step becomes its own run. A placeholder compiler turns JSONL command sequences into reproducible workflow DAGs, while a skill layer maps each declared control to the corresponding command-line option; plugins with tool-specific contracts expose tool-specific actions. YR-MPE runs natively on Windows and bundles alignment, model selection, maximum-likelihood, Bayesian and species-tree inference, molecular dating, selection and visualization tools. We describe the interface and its provenance/workflow substrate, and report three representative recipes: an organelle-genome workflow, a single-locus model-to-tree analysis, and an agent-driven run. YR-MPE is released under GPL-3.0 at https://github.com/Gipsy-The-Sheller/YR-MPE.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-18
DOI
https://doi.org/10.21203/rs.3.rs-11771753/v1
Primary Topic
Genomics and Phylogenetic Studies
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

YR-MPE: a modular desktop platform for agent-operable and reproducible phylogenetic workflows

Zhihan Gu, Yang Jin, Yunlong Zhang, Zhi-Jie Xu et al.
Research Square
Genomics and Phylogenetic Studies
preprint

YR-MPE: a modular desktop platform for agent-operable and reproducible phylogenetic workflows

Zhihan Gu, Yang Jin, Yunlong Zhang, Zhi-Jie Xu, Rui Zhang, Yiyang Jia, Yi-Tong Lei, Shuo Han, Xin-Bo Cheng, Ming-Rui Shang
preprint en

Abstract

Phylogenomic analyses combine many tools, file formats and execution modes. Desktop graphical user interfaces make individual steps accessible but hide their state from scripts; command-line pipelines make steps scriptable but leave parameter discovery, provenance and recovery to the user; and emerging LLM agents can call command-line tools but cannot inspect a GUI session or its artifacts. We present YR-MPE, a modular desktop platform whose YPhylo Workbench unifies these modes through a typed workspace, a plugin contract and a semantic action layer. Every operation in the action registry is exposed as a machine-callable action; controls without a CLI equivalent are declared GUI-only. The workspace records every imported plugin result as a run with its inputs, parameters, tool identity and output artifacts, the latter with SHA-256 hashes; each compiled workflow step becomes its own run. A placeholder compiler turns JSONL command sequences into reproducible workflow DAGs, while a skill layer maps each declared control to the corresponding command-line option; plugins with tool-specific contracts expose tool-specific actions. YR-MPE runs natively on Windows and bundles alignment, model selection, maximum-likelihood, Bayesian and species-tree inference, molecular dating, selection and visualization tools. We describe the interface and its provenance/workflow substrate, and report three representative recipes: an organelle-genome workflow, a single-locus model-to-tree analysis, and an agent-driven run. YR-MPE is released under GPL-3.0 at https://github.com/Gipsy-The-Sheller/YR-MPE.

Research Square
Donghua University (CN), Shanghai Jiao Tong University (CN)
National Natural Science Foundation of China
Genomics and Phylogenetic Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon