Machine Learning Uncovers the Transcriptional Regulatory Network for the Production Host Streptomyces albidoflavus J1074

Streptomyces albidoflavus J1074 is a popular and genetically tractable platform strain used for drug discovery and production via the expression of heterologous biosynthetic gene clusters (BGCs). However, its transcriptional regulation network (TRN) and its impact on secondary metabolism are poorly understood. Here we characterized its TRN by applying an independent component analysis to a compendium of 142 high quality RNA-Seq transcriptomes, from both in-house and public sources spanning 58 unique growth conditions. We obtained 63 independently modulated sets of genes (iModulons), that quantitatively describe the TRN and its activity state across diverse conditions. Through analyses of condition-dependent TRN activity states, we (i) describe how the TRN adapts to different growth conditions, (ii) conduct a cross-species iModulon comparison, uncovering shared features and unique characteristics of the TRN across lineages, (iii) identify putative regulators of several biosynthetic gene clusters, including Surugamide, Cyclofaulknamycin and Dudomycin, and (iv) infer potential functions of 40% of the uncharacterized genes in the S. albidoflavus J1074 genome. Our findings provide a comprehensive and quantitative understanding of the TRN of S. albidoflavus J1074, and suggest new strategies for rational strain design to optimize its production capabilities. ### Competing Interest Statement The authors have declared no competing interest.