Orchestra: Corroboration-Based Regulatory Candidate Discovery via Composed Bioinformatics MCP Agents

Abstract

Orchestra composes two independently built bioinformatics MCP servers -- RegNetAgents, which infers regulatory network topology, and CASCADE, which provides independent functional evidence (LINCS knockdown, DepMap essentiality, super-enhancer status, DoRothEA transcription-factor confidence) -- into one multi-agent LangGraph workflow. Its central claim is that requiring the two systems to agree on a candidate regulator yields a more trustworthy candidate than either alone. We test this on the TCGA tumor-acquired regulator tier, selecting each query's ten regulators by ARACNe mutual-information edge weight. Across BRCA/COAD focal genes and matched negative controls, requiring at least 2 of 4 evidence sources predicts known-driver status (IntOGen) among focal genes (odds ratio 3.15, Benjamini-Hochberg-adjusted p=0.016; 2.92 after deduplication) but shows no association among negative controls (p=0.70); a single source does not in BRCA/COAD. In a second panel (STAD) the pattern reproduces and is stronger (OR 6.71, adjusted p=0.0047), again with no association in controls. Corroboration adds value beyond MI edge weight in both panels (BRCA/COAD p=0.039; STAD p=0.0001), though modestly in BRCA/COAD. Deduplication, a permutation test and a reproduction of RegNetAgents' published results agree; a second, separately derived driver list (Vogelstein et al. 2013) is directionally consistent but too sparse to reach significance. All experiments invoke Orchestra's real agentic entry point.

Transcriptomic identification of high-altitude illness candidate genes and physiological mechanisms: a multidimensional bioinformatics study with 106 drug-food homologous substance target prediction.
Defang Liu, Jiao Yan, Ting Luo +4
Frontiers in physiology14 Oct 2026Peer-reviewedCC BYFull textGenetics & genomicsEurope PMCPMC
Integrated 16S rRNA Bioinformatics Analysis of Gut Microbiota, Predicted Pathways, and Candidate Biomarkers in Inflammatory Bowel Disease
A. Zehra, E. Natarajan
bioRxiv16 Oct 2026PreprintCC BYFull textBioinformatics & computational biologybioRxiv
Targeting PRMT5 Across Cancers and AI-Driven Discovery of Candidate Compounds
S. Rana, M. S. Hossan, A. A. Rafi +5
bioRxiv19 Oct 2026PreprintAll rights reservedFull textBioinformatics & computational biologybioRxiv
Comprehensive in silico analysis reveals candidate regulatory mechanisms underlying selective cerebellar vulnerability in pontocerebellar hypoplasia
L. Branco, S. Mayer
bioRxiv28 Sep 2026PreprintCC BYFull textSystems & synthetic biologybioRxiv
Genome-wide association study identifies candidate loci and a functional regulatory variant associated with egg-production traits in Wenchang chickens
Lingzhi Fan, Xing Hong, Mingze Li +4
Poultry science5 Oct 2026Peer-reviewedCC BY-NC-NDFull textGenetics & genomicsEurope PMCPMC

Cite

BibTeX RIS

BibTeX

@misc{bird2026orchestra,
  title = {{Orchestra: Corroboration-Based Regulatory Candidate Discovery via Composed Bioinformatics MCP Agents}},
  author = {Jose A. Bird},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2609.05496},
  eprint = {2609.05496},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.05496},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Orchestra: Corroboration-Based Regulatory Candidate Discovery via Composed Bioinformatics MCP Agents
AU  - Jose A. Bird
PY  - 2026
DA  - 2026/10/16
PB  - arXiv
DO  - 10.48550/arxiv.2609.05496
UR  - https://arxiv.org/abs/2609.05496
AB  - Orchestra composes two independently built bioinformatics MCP servers -- RegNetAgents, which infers regulatory network topology, and CASCADE, which provides independent functional evidence (LINCS knockdown, DepMap essentiality, super-enhancer status, DoRothEA transcription-factor confidence) -- into one multi-agent LangGraph workflow. Its central claim is that requiring the two systems to agree on a candidate regulator yields a more trustworthy candidate than either alone. We test this on the TCGA tumor-acquired regulator tier, selecting each query's ten regulators by ARACNe mutual-information edge weight. Across BRCA/COAD focal genes and matched negative controls, requiring at least 2 of 4 evidence sources predicts known-driver status (IntOGen) among focal genes (odds ratio 3.15, Benjamini-Hochberg-adjusted p=0.016; 2.92 after deduplication) but shows no association among negative controls (p=0.70); a single source does not in BRCA/COAD. In a second panel (STAD) the pattern reproduces and is stronger (OR 6.71, adjusted p=0.0047), again with no association in controls. Corroboration adds value beyond MI edge weight in both panels (BRCA/COAD p=0.039; STAD p=0.0001), though modestly in BRCA/COAD. Deduplication, a permutation test and a reproduction of RegNetAgents' published results agree; a second, separately derived driver list (Vogelstein et al. 2013) is directionally consistent but too sparse to reach significance. All experiments invoke Orchestra's real agentic entry point.
ER  -

Abstract

Related

Cite