bioRxiv Science⌕ Search

Biology subjects

Armstrong, D. J.

Publications and source records attributed to Armstrong, D. J..

2 recordsLinked to original sources
bioRxiv2026

VFB-MCP: Natural-Language Access to Drosophila Neuroscience Grounded by an Expert-Curated Ontology-Led Knowledgebase

Biological databases store curated knowledge that researchers traditionally access through web interfaces or APIs. To move beyond casual browsing requires domain-specific knowledge and expertise to frame the queries necessary to explore this data. This generates barriers to both experienced users trying to integrate wide-ranging sources of information and for new users entering scientific fields undergoing paradigm shifts such as connectomics. A potentially powerful method to lower these barriers is to facilitate natural-language access to these databases by exposing them to large language models (LLMs) via the Model Context Protocol (MCP). Here we implement this for Virtual Fly Brain (VFB), an expert-curated and ontology-backed knowledgebase of Drosophila neuroscience, providing the precision needed to make recently integrated connectome and molecular data accessible. Benchmarked on 30 neuroscience tasks against a bare LLM and a web-search-assisted LLM, the VFB-MCP-equipped LLM produces precise, verifiable and appropriately quantified answers on 25/30 tasks vs 14/30 for web and 2/30 for bare. The MCP advantage is largest for tasks where data quantification is required (89% vs 11% web). This work establishes MCP layered over ontology-backed knowledge graphs as an effective method to improve LLM response quality for neuroscience and connectomics data, promoting accessibility and accelerating the reliable analysis of complex integrated datasets.

neuroscience↗
bioRxiv2026

RAPID DECLINE OF NESTING PEREGRINE FALCONS IN THE SAN FRANCISCO BAY REGION OF CALIFORNIA SYNCHRONOUS WITH AN H5N1 OUTBREAK

After rebounding from near extirpation during the organochlorine era, breeding Peregrine Falcons (Falco peregrinus) in California are again facing adversity, this time consistent with an outbreak of a highly pathogenic avian influenza. Following the first detection of the H5N1 variant clade 2.3.4.4b virus in California wild birds in July 2022, we assembled data from long-term monitoring (2000-2025) of peregrine breeding territory occupancy in the broad vicinity of San Francisco Bay to examine possible impacts on falcon populations. Prior to the outbreak, 47 focal breeding territories had shown nearly complete occupancy by pairs (98.5% of 390 site-years), with very few vacancies, single birds in attendance, or subadult pair members. Within 8 mo of the outbreak, occupancy had dropped to 65.1%, and 2 yrs later (2025), only 36.2% of sites remained occupied. An uptick in site-occupancy by single birds also occurred after the outbreak, but it is unclear whether these were survivors or floaters attempting to fill vacant territories where both pair members had perished. The high vacancy rates signaled an impact upon floaters (nonbreeding adults) that normally buffer breeding site-occupancy in healthy peregrine populations. From October 2022 through November 2025, 17 peregrine fatalities were diagnosed with H5N1 within the counties comprising our study area. Evidence that H5N1 caused these territory vacancies includes, (1) the striking temporal coincidence of occupancy loss with the outbreak, and (2) the lethality of the virus to peregrines and its confirmed presence in peregrine prey in our study area. Our study reaffirms the value of long-term territory occupancy monitoring in this sentinel species.

ecology↗