A machine-learning enhanced data fusion pipeline identifies new G-protein-based regulators of Drosophila wing growth and morphogenesis

The development of multicellular organisms relies on a symphony of spatiotemporally coordinated signals that selectively regulate gene expression. In particular, G protein-coupled receptors (GPCRs), the largest superfamily of transmembrane receptors, play a pivotal role in transducing extracellular signals into physiological outcomes. Notably, neurotransmitter GPCRs, classically associated with neuronal tissue communication, are increasingly emerging as regulators of pattern formation and morphogenesis. However, how these receptors coordinate such morphogenetic processes remains poorly understood. To address this gap, we developed and employed a coupled, machine-learning-based analytical pipeline, MAPPER 2.0, that fuses quantitative and qualitative analyses of Drosophila melanogaster wing phenotypes to robustly identify both severe and more subtle phenotypes generated by RNAi expression. We phenotypically characterized the impact of RNAi-based inhibition of the 111 GPCRs and the G-protein subunits in Drosophila , a genetic model system for investigating conserved protein and gene regulatory pathways. Severe morphological phenotypes resulted from RNAi-mediated knockdown targeting several G-proteins and neuropeptide and neurotransmitter GPCRs, with seven knockdowns exhibiting greater than 80% penetrance. Beyond these strong qualitative hits, MAPPER 2.0 revealed a broader class of more subtle phenotypes, including quantitative differences in wing size, compartmental organization, and vein patterning. Quantitative reverse transcription polymerase chain reaction and meta-analysis of RNA expression data validated that positive hits are expressed in the wing disc. Overall, MAPPER 2.0 provides a phenotypic platform for drug testing and mechanism discovery in GPCR-implicated human diseases, ranging from cancers to neurological conditions.