In the early, online edition of the Proceedings of the National Academy of Sciences, researchers from Emory University, Northwestern University, and Case Western Reserve University describe a computational approach that ties together traditional cancer survival models with adaptive machine-learning algorithms. After demonstrating the "survival convolutional neural networks (SCNNs)" methods with available data for diffuse glioma, the team established a "genomic survival convolutional neural network (GSCNN)" that further integrated histologic and genomic data for predicting cancer outcomes. "Our approach surpasses the prognostic accuracy of human experts using the current clinical standard for classifying brain tumors and presents an innovative approach for objective, accurate, and integrated prediction of patient outcomes," the authors say.
A team from Germany and the US introduces an RNA-sequencing pipeline that makes it possible to track the transcriptomes of plant pathogens during infection and exposure to the plant's innate immune system. The team applied the approach to Arabidopsis thaliana plants infected with the leaf-infecting pathogen Pseudomonas syringae, following transcriptional patterns in different strains of the bacterial pathogen over time in plants with a range of immune-related mutations. "Using various combinations of the P. syringae strains and immune-compromised plant genotypes, we showed that expression patterns of the early infection stage had a high predictive power for later bacterial growth at 48 [hours post infection]," the authors report.
Finally, Monsanto and Colorado State University researchers explore auxin herbicide resistance mechanisms in the invasive weed Kochia scoparia L. Schrad, commonly known as kochia. The team focused on a kochia biotype found in western Nebraska that has documented resistance to herbicides such as dicamba, 2,4-D, and fluroxypyr, using transcriptome sequencing to narrow in on a suspicious mutation in a conserved region of the AUX or IAA gene, which codes for an indole-3-acetic acid protein. Through a series of follow-up experiments, including yeast two-hybrid analyses, the authors verified the resistance allele and tracked down potential markers for predicting herbicide resistance.