We have gained insights into network resilience and susceptibility through induction of a variety of random and targeted perturbations in hundreds of biological networks, followed by investigation of their responses under attack. This has enabled us to develop tools for predicting robustness in real world networks, and for partitioning complex networks into biologically meaningful communities. We are also developing deep learning models for deriving regulatory inferences from large scale gene expression data. These models have been used to decipher the floral transition network of Arabidopsis, rice master regulators of abiotic stress and more recently, evidence for prion-like proteins as mediators of cross-talk between transient and trans-generational memory in plants. (Work Funded by DBT)