Generating multi-scale predictive networks of Northern Corn Leaf Blight resistance

This talk was presented as part of the eLife Online Research Talks series on April 9, 2020: https://elifesciences.org/inside-elife/1a9d9c08/elife-and-covid-19-keeping-communications-open-with-online-research-talks

Northern Corn Leaf Blight (NLB) is one of the most significant corn diseases, causing the largest crop loss of any disease in the Northern United States from 2012-2015. Understanding how gene expression correlates with specific disease phenotypes is critical for developing disease-resistant crops that enhance global food security and limit economic losses. However, the underlying regulatory mechanisms mediating resistance to NLB are not well understood. Thus, the goal of this study is to integrate transcriptome and proteome measurements with disease phenotyping and computational modeling to build predictive regulatory networks of NLB resistance. To address this, 134 Intermated B73xMo17 Doubled Haploid Lines (IBMDHLs) were planted in a field plot using a randomized complete block design. Plants were treated with NLB inoculum at the sixth and seventh vegetative stages. Seven days after inoculation, leaf material was collected from 3 infected plants for each of the lines. Transcriptomic profiling on the infected plant samples was performed using 3’ QuantSeq, and mass spectrometry was performed using 11-plex Tandem Mass Tag (TMT) peptide labeling. Using these methods, we were able to detect over 44,000 transcripts, 11,000 proteins, and 42,000 phosphosites. We then performed Quantitative Trait Loci (QTL) mapping on each of these molecular datasets as well as disease QTL mapping. We identified over 25,000 transcript, 5,000 protein, and 4,500 phosphosite QTL (eQTL, pQTL, and phosQTL, respectively), of which approximately 20% co-localize with disease QTL. In addition, only approximately 20% of pQTL and 10% of phosQTL overlap with eQTL, illustrating that the incorporation of proteomic data identifies complimentary regulatory events. We then inferred a network of the genes involved in these different QTLs using Spatiotemporal Clustering for Integrative Omics Networks (SCION: https://github.com/nmclark2/SCION). Together these data provide a comprehensive view of the gene regulatory landscape of corn leaves in response to NLB.