Search for: All records

Abstract Acrasids are amoebae with the capacity to form multicellular fruiting bodies in a process known as aggregative multicellularity (AGM). This makes acrasids the only known example of multicellularity among the earliest branches of eukaryotes (the former Excavata). Here, we report theAcrasis konagenome sequence plus transcriptomes from pre-, mid- and post-developmental stages. The genome is rich in novelty and genes with strong signatures of horizontal transfer, and multigene families encode nearly half of the amoeba’s predicted proteome. Development inA. konaappears molecularly simple relative to the AGM model,Dictyostelium discoideum. However, the acrasid also differs from the dictyostelid in that it does not appear to be starving during development. Instead, developingA. konaappears to be very metabolically active, does not induce autophagy and does not up-regulate its proteasomal genes. Together, these observations strongly suggest that starvation is not essential for AGM development. Nonetheless, development in the two amoebae appears to employ remarkably similar pathways for signaling, motility and, potentially, construction of an extracellular matrix surrounding the developing cell mass. Much of this similarity is also shared with animal development, suggesting that much of the basic tool kit for multicellular development arose early in eukaryote evolution. 

Recent advances in model piracy have uncovered a new security hole for malicious attacks endangering the Intellectual Property (IP) of Deep Learning (DL) systems. This manuscript features our research titled “DeepAttest: An End-toEnd Attestation Framework for Deep Neural Networks” [1] that is selected for the 2021 Top Picks in hardware and embedded security. DeepAttest is the first end-to-end framework that achieves reliable and efficient IP protection of DL devices with hardware-bounded usage control. We leverage device-specific model fingerprinting and Trusted Execution Environment (TEE) to ensure that only DL models with the device-specific fingerprint can run inference on protected hardware